agora inbox for [email protected]  

help / color / mirror / Atom feed

[PATCH v3] Avoid orphaned objects dependencies
144+ messages / 2 participants
[nested] [flat]

* [PATCH v3] Avoid orphaned objects dependencies
@ 2024-03-29 15:43  Bertrand Drouvot <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Bertrand Drouvot @ 2024-03-29 15:43 UTC (permalink / raw)

It's currently possible to create orphaned objects dependencies, for example:

Scenario 1:

session 1: begin; drop schema schem;
session 2: create a function in the schema schem
session 1: commit;

With the above, the function created in session 2 would be linked to a non
existing schema.

Scenario 2:

session 1: begin; create a function in the schema schem
session 2: drop schema schem;
session 1: commit;

With the above, the function created in session 1 would be linked to a non
existing schema.

To avoid those scenarios, a new lock (that conflicts with a lock taken by DROP)
has been put in place when the dependencies are being recorded. With this in
place, the drop schema in scenario 2 would be locked.

Also, after locking the object, the patch checks that the object still exists:
with this in place session 2 in scenario 1 would be locked and would report an
error once session 1 committs (that would not be the case should session 1 abort
the transaction).

The patch adds a few tests for some dependency cases (that would currently produce
orphaned objects):

- schema and function (as the above scenarios)
- function and type
- table and type
---
 src/backend/catalog/dependency.c              | 54 ++++++++++++++
 src/backend/catalog/objectaddress.c           | 70 +++++++++++++++++++
 src/backend/catalog/pg_depend.c               |  6 ++
 src/include/catalog/dependency.h              |  1 +
 src/include/catalog/objectaddress.h           |  1 +
 src/test/modules/Makefile                     |  1 +
 src/test/modules/meson.build                  |  1 +
 .../test_dependencies_locks/.gitignore        |  3 +
 .../modules/test_dependencies_locks/Makefile  | 14 ++++
 .../expected/test_dependencies_locks.out      | 49 +++++++++++++
 .../test_dependencies_locks/meson.build       | 12 ++++
 .../specs/test_dependencies_locks.spec        | 39 +++++++++++
 12 files changed, 251 insertions(+)
  40.5% src/backend/catalog/
  29.6% src/test/modules/test_dependencies_locks/expected/
  18.7% src/test/modules/test_dependencies_locks/specs/
   8.3% src/test/modules/test_dependencies_locks/

diff --git a/src/backend/catalog/dependency.c b/src/backend/catalog/dependency.c
index d4b5b2ade1..a49357bbe2 100644
--- a/src/backend/catalog/dependency.c
+++ b/src/backend/catalog/dependency.c
@@ -1517,6 +1517,60 @@ AcquireDeletionLock(const ObjectAddress *object, int flags)
 	}
 }
 
+/*
+ * depLockAndCheckObject
+ *
+ * Lock the object that we are about to record a dependency on.
+ * After it's locked, verify that it hasn't been dropped while we
+ * weren't looking.  If the object has been dropped, this function
+ * does not return!
+ */
+void
+depLockAndCheckObject(const ObjectAddress *object)
+{
+	char	   *object_description;
+
+	/*
+	 * Those don't rely on LockDatabaseObject() when being dropped (see
+	 * AcquireDeletionLock()). Also it looks like they can not produce
+	 * orphaned dependent objects when being dropped.
+	 */
+	if (object->classId == RelationRelationId || object->classId == AuthMemRelationId)
+		return;
+
+	object_description = getObjectDescription(object, true);
+
+	/* assume we should lock the whole object not a sub-object */
+	LockDatabaseObject(object->classId, object->objectId, 0, AccessShareLock);
+
+	/* check if object still exists */
+	if (!ObjectByIdExist(object, false))
+	{
+		/*
+		 * It might be possible that we are creating it (for example creating
+		 * a composite type while creating a relation), so bypass the syscache
+		 * lookup and use a dirty snaphot instead to cover this scenario.
+		 */
+		if (!ObjectByIdExist(object, true))
+		{
+			/*
+			 * If the object has been dropped before we get a chance to get
+			 * its description, then emit a generic error message. That looks
+			 * like a good compromise over extra complexity.
+			 */
+			if (object_description)
+				ereport(ERROR, errmsg("%s does not exist", object_description));
+			else
+				ereport(ERROR, errmsg("a dependent object does not exist"));
+		}
+	}
+
+	if (object_description)
+		pfree(object_description);
+
+	return;
+}
+
 /*
  * ReleaseDeletionLock - release an object deletion lock
  *
diff --git a/src/backend/catalog/objectaddress.c b/src/backend/catalog/objectaddress.c
index 7b536ac6fd..c2b873dd81 100644
--- a/src/backend/catalog/objectaddress.c
+++ b/src/backend/catalog/objectaddress.c
@@ -2590,6 +2590,76 @@ get_object_namespace(const ObjectAddress *address)
 	return oid;
 }
 
+/*
+ * ObjectByIdExist
+ *
+ * Return whether the given object exists.
+ *
+ * Works for most catalogs, if no special processing is needed.
+ */
+bool
+ObjectByIdExist(const ObjectAddress *address, bool use_dirty_snapshot)
+{
+	HeapTuple	tuple;
+	int			cache = -1;
+	const ObjectPropertyType *property;
+
+	if (!use_dirty_snapshot)
+	{
+		property = get_object_property_data(address->classId);
+
+		cache = property->oid_catcache_id;
+	}
+
+	if (cache >= 0)
+	{
+		/* Fetch tuple from syscache. */
+		tuple = SearchSysCache1(cache, ObjectIdGetDatum(address->objectId));
+
+		if (!HeapTupleIsValid(tuple))
+		{
+			return false;
+		}
+
+		ReleaseSysCache(tuple);
+
+		return true;
+	}
+	else
+	{
+		Relation	rel;
+		ScanKeyData skey[1];
+		SysScanDesc scan;
+		SnapshotData DirtySnapshot;
+		Snapshot	snapshot;
+
+		if (use_dirty_snapshot)
+		{
+			InitDirtySnapshot(DirtySnapshot);
+			snapshot = &DirtySnapshot;
+		}
+		else
+			snapshot = NULL;
+
+		rel = table_open(address->classId, AccessShareLock);
+
+		ScanKeyInit(&skey[0],
+					get_object_attnum_oid(address->classId),
+					BTEqualStrategyNumber, F_OIDEQ,
+					ObjectIdGetDatum(address->objectId));
+
+		scan = systable_beginscan(rel, get_object_oid_index(address->classId), true,
+								  snapshot, 1, skey);
+
+		/* we expect exactly one match */
+		tuple = systable_getnext(scan);
+		systable_endscan(scan);
+		table_close(rel, AccessShareLock);
+
+		return (HeapTupleIsValid(tuple));
+	}
+}
+
 /*
  * Return ObjectType for the given object type as given by
  * getObjectTypeDescription; if no valid ObjectType code exists, but it's a
diff --git a/src/backend/catalog/pg_depend.c b/src/backend/catalog/pg_depend.c
index f85a898de8..6bb218f5dd 100644
--- a/src/backend/catalog/pg_depend.c
+++ b/src/backend/catalog/pg_depend.c
@@ -106,6 +106,12 @@ recordMultipleDependencies(const ObjectAddress *depender,
 		if (isObjectPinned(referenced))
 			continue;
 
+		/*
+		 * Acquire a lock and check object still exists while recording the
+		 * dependency. XXX - Should we do so only for DEPENDENCY_NORMAL?
+		 */
+		depLockAndCheckObject(referenced);
+
 		if (slot_init_count < max_slots)
 		{
 			slot[slot_stored_count] = MakeSingleTupleTableSlot(RelationGetDescr(dependDesc),
diff --git a/src/include/catalog/dependency.h b/src/include/catalog/dependency.h
index ec654010d4..8915548711 100644
--- a/src/include/catalog/dependency.h
+++ b/src/include/catalog/dependency.h
@@ -94,6 +94,7 @@ typedef struct ObjectAddresses ObjectAddresses;
 /* in dependency.c */
 
 extern void AcquireDeletionLock(const ObjectAddress *object, int flags);
+extern void depLockAndCheckObject(const ObjectAddress *object);
 
 extern void ReleaseDeletionLock(const ObjectAddress *object);
 
diff --git a/src/include/catalog/objectaddress.h b/src/include/catalog/objectaddress.h
index 3a70d80e32..04891abcc1 100644
--- a/src/include/catalog/objectaddress.h
+++ b/src/include/catalog/objectaddress.h
@@ -53,6 +53,7 @@ extern void check_object_ownership(Oid roleid,
 								   Node *object, Relation relation);
 
 extern Oid	get_object_namespace(const ObjectAddress *address);
+extern bool ObjectByIdExist(const ObjectAddress *address, bool use_dirty_snapshot);
 
 extern bool is_objectclass_supported(Oid class_id);
 extern const char *get_object_class_descr(Oid class_id);
diff --git a/src/test/modules/Makefile b/src/test/modules/Makefile
index 256799f520..75f357100f 100644
--- a/src/test/modules/Makefile
+++ b/src/test/modules/Makefile
@@ -17,6 +17,7 @@ SUBDIRS = \
 		  test_copy_callbacks \
 		  test_custom_rmgrs \
 		  test_ddl_deparse \
+		  test_dependencies_locks \
 		  test_dsa \
 		  test_dsm_registry \
 		  test_extensions \
diff --git a/src/test/modules/meson.build b/src/test/modules/meson.build
index d8fe059d23..60305dcccd 100644
--- a/src/test/modules/meson.build
+++ b/src/test/modules/meson.build
@@ -16,6 +16,7 @@ subdir('test_bloomfilter')
 subdir('test_copy_callbacks')
 subdir('test_custom_rmgrs')
 subdir('test_ddl_deparse')
+subdir('test_dependencies_locks')
 subdir('test_dsa')
 subdir('test_dsm_registry')
 subdir('test_extensions')
diff --git a/src/test/modules/test_dependencies_locks/.gitignore b/src/test/modules/test_dependencies_locks/.gitignore
new file mode 100644
index 0000000000..bf000faac4
--- /dev/null
+++ b/src/test/modules/test_dependencies_locks/.gitignore
@@ -0,0 +1,3 @@
+# Generated subdirectories
+/log/
+/output_iso
diff --git a/src/test/modules/test_dependencies_locks/Makefile b/src/test/modules/test_dependencies_locks/Makefile
new file mode 100644
index 0000000000..7491048380
--- /dev/null
+++ b/src/test/modules/test_dependencies_locks/Makefile
@@ -0,0 +1,14 @@
+# src/test/modules/test_dependencies_locks/Makefile
+
+ISOLATION = test_dependencies_locks
+
+ifdef USE_PGXS
+PG_CONFIG = pg_config
+PGXS := $(shell $(PG_CONFIG) --pgxs)
+include $(PGXS)
+else
+subdir = src/test/modules/test_dependencies_locks
+top_builddir = ../../../..
+include $(top_builddir)/src/Makefile.global
+include $(top_srcdir)/contrib/contrib-global.mk
+endif
diff --git a/src/test/modules/test_dependencies_locks/expected/test_dependencies_locks.out b/src/test/modules/test_dependencies_locks/expected/test_dependencies_locks.out
new file mode 100644
index 0000000000..d0980f77d5
--- /dev/null
+++ b/src/test/modules/test_dependencies_locks/expected/test_dependencies_locks.out
@@ -0,0 +1,49 @@
+Parsed test spec with 2 sessions
+
+starting permutation: s1_begin s1_create_function_in_schema s2_drop_schema s1_commit
+step s1_begin: BEGIN;
+step s1_create_function_in_schema: CREATE FUNCTION testschema.foo() RETURNS int AS 'select 1' LANGUAGE sql;
+step s2_drop_schema: DROP SCHEMA testschema; <waiting ...>
+step s1_commit: COMMIT;
+step s2_drop_schema: <... completed>
+ERROR:  cannot drop schema testschema because other objects depend on it
+
+starting permutation: s2_begin s2_drop_schema s1_create_function_in_schema s2_commit
+step s2_begin: BEGIN;
+step s2_drop_schema: DROP SCHEMA testschema;
+step s1_create_function_in_schema: CREATE FUNCTION testschema.foo() RETURNS int AS 'select 1' LANGUAGE sql; <waiting ...>
+step s2_commit: COMMIT;
+step s1_create_function_in_schema: <... completed>
+ERROR:  schema testschema does not exist
+
+starting permutation: s1_begin s1_create_function_with_type s2_drop_foo_type s1_commit
+step s1_begin: BEGIN;
+step s1_create_function_with_type: CREATE FUNCTION footype(num foo) RETURNS int AS 'select 1' LANGUAGE sql;
+step s2_drop_foo_type: DROP TYPE public.foo; <waiting ...>
+step s1_commit: COMMIT;
+step s2_drop_foo_type: <... completed>
+ERROR:  cannot drop type foo because other objects depend on it
+
+starting permutation: s2_begin s2_drop_foo_type s1_create_function_with_type s2_commit
+step s2_begin: BEGIN;
+step s2_drop_foo_type: DROP TYPE public.foo;
+step s1_create_function_with_type: CREATE FUNCTION footype(num foo) RETURNS int AS 'select 1' LANGUAGE sql; <waiting ...>
+step s2_commit: COMMIT;
+step s1_create_function_with_type: <... completed>
+ERROR:  type foo does not exist
+
+starting permutation: s1_begin s1_create_table_with_type s2_drop_footab_type s1_commit
+step s1_begin: BEGIN;
+step s1_create_table_with_type: CREATE TABLE tabtype(a footab);
+step s2_drop_footab_type: DROP TYPE public.footab; <waiting ...>
+step s1_commit: COMMIT;
+step s2_drop_footab_type: <... completed>
+ERROR:  cannot drop type footab because other objects depend on it
+
+starting permutation: s2_begin s2_drop_footab_type s1_create_table_with_type s2_commit
+step s2_begin: BEGIN;
+step s2_drop_footab_type: DROP TYPE public.footab;
+step s1_create_table_with_type: CREATE TABLE tabtype(a footab); <waiting ...>
+step s2_commit: COMMIT;
+step s1_create_table_with_type: <... completed>
+ERROR:  type footab does not exist
diff --git a/src/test/modules/test_dependencies_locks/meson.build b/src/test/modules/test_dependencies_locks/meson.build
new file mode 100644
index 0000000000..92a978ab93
--- /dev/null
+++ b/src/test/modules/test_dependencies_locks/meson.build
@@ -0,0 +1,12 @@
+# Copyright (c) 2024, PostgreSQL Global Development Group
+
+tests += {
+  'name': 'test_dependencies_locks',
+  'sd': meson.current_source_dir(),
+  'bd': meson.current_build_dir(),
+  'isolation': {
+    'specs': [
+      'test_dependencies_locks',
+    ],
+  },
+}
diff --git a/src/test/modules/test_dependencies_locks/specs/test_dependencies_locks.spec b/src/test/modules/test_dependencies_locks/specs/test_dependencies_locks.spec
new file mode 100644
index 0000000000..fd15bd2a78
--- /dev/null
+++ b/src/test/modules/test_dependencies_locks/specs/test_dependencies_locks.spec
@@ -0,0 +1,39 @@
+setup
+{
+  CREATE SCHEMA testschema;
+  CREATE TYPE public.foo as enum ('one', 'two');
+  CREATE TYPE public.footab as enum ('three', 'four');
+}
+
+teardown
+{
+  DROP FUNCTION IF EXISTS testschema.foo();
+  DROP FUNCTION IF EXISTS footype(num foo);
+  DROP TABLE IF EXISTS tabtype;
+  DROP SCHEMA IF EXISTS testschema;
+  DROP TYPE IF EXISTS public.foo;
+  DROP TYPE IF EXISTS public.footab;
+}
+
+session "s1"
+
+step "s1_begin" { BEGIN; }
+step "s1_create_function_in_schema" { CREATE FUNCTION testschema.foo() RETURNS int AS 'select 1' LANGUAGE sql; }
+step "s1_create_function_with_type" { CREATE FUNCTION footype(num foo) RETURNS int AS 'select 1' LANGUAGE sql; }
+step "s1_create_table_with_type" { CREATE TABLE tabtype(a footab); }
+step "s1_commit" { COMMIT; }
+
+session "s2"
+
+step "s2_begin" { BEGIN; }
+step "s2_drop_schema" { DROP SCHEMA testschema; }
+step "s2_drop_foo_type" { DROP TYPE public.foo; }
+step "s2_drop_footab_type" { DROP TYPE public.footab; }
+step "s2_commit" { COMMIT; }
+
+permutation "s1_begin" "s1_create_function_in_schema" "s2_drop_schema" "s1_commit"
+permutation "s2_begin" "s2_drop_schema" "s1_create_function_in_schema" "s2_commit"
+permutation "s1_begin" "s1_create_function_with_type" "s2_drop_foo_type" "s1_commit"
+permutation "s2_begin" "s2_drop_foo_type" "s1_create_function_with_type" "s2_commit"
+permutation "s1_begin" "s1_create_table_with_type" "s2_drop_footab_type" "s1_commit"
+permutation "s2_begin" "s2_drop_footab_type" "s1_create_table_with_type" "s2_commit"
-- 
2.34.1


--+Mfbf1cGg0Kwp7Pc--





^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread

* [PATCH v2 3/4] tests: Fix race condition in postmaster/002_connection_limits
@ 2025-03-07 14:44  Andres Freund <[email protected]>
  0 siblings, 0 replies; 144+ messages in thread

From: Andres Freund @ 2025-03-07 14:44 UTC (permalink / raw)

The test occasionally failed due to unexpected connection limit errors being
encountered after having waited for FATAL errors on another connection. These
spurious failures were caused by the the backend reporting FATAL errors to the
client before detaching from the PGPROC entry. Adding a sleep(1) before
proc_exit() makes it easy to reproduce that problem.

To fix the issue, add a helper function that waits for postmaster to notice
the process having exited. For now this is implemented by waiting for the
DEBUG2 message that postmaster logs in that case. That's not the prettiest
fix, but simple. If we notice this problem elsewhere, it might be worthwhile
to make this more general, e.g. by adding an injection point.

Reported-by: Tomas Vondra <[email protected]>
Diagnosed-by: Heikki Linnakangas <[email protected]>
Reviewed-by: Heikki Linnakangas <[email protected]>
Discussion: https://postgr.es/m/ggflhkciwdyotpoie323chu2c2idpjk5qimrn462encwx2io7s@thmcxl7i6dpw
---
 .../postmaster/t/002_connection_limits.pl     | 35 +++++++++++++++++--
 1 file changed, 32 insertions(+), 3 deletions(-)

diff --git a/src/test/postmaster/t/002_connection_limits.pl b/src/test/postmaster/t/002_connection_limits.pl
index 8cfa6e0ced5..2c185eef6eb 100644
--- a/src/test/postmaster/t/002_connection_limits.pl
+++ b/src/test/postmaster/t/002_connection_limits.pl
@@ -20,6 +20,7 @@ $node->append_conf('postgresql.conf', "max_connections = 6");
 $node->append_conf('postgresql.conf', "reserved_connections = 2");
 $node->append_conf('postgresql.conf', "superuser_reserved_connections = 1");
 $node->append_conf('postgresql.conf', "log_connections = on");
+$node->append_conf('postgresql.conf', "log_min_messages=debug2");
 $node->start;
 
 $node->safe_psql(
@@ -45,13 +46,39 @@ sub background_psql_as_user
 		extra_params => [ '-U', $user ]);
 }
 
+# Like connect_fails(), except that we also wait for the failed backend to
+# have exited.
+#
+# This tests needs to wait for client processes to exit because the error
+# message for a failed connection is reported before the backend has detached
+# from shared memory. If we didn't wait, subsequent tests might hit connection
+# limits spuriously.
+#
+# This can't easily be generalized, as detecting process exit requires
+# log_min_messages to be at least DEBUG2 and is not concurrency safe, as we
+# can't easily be sure the right process exited. In this test that's not a
+# problem though, we only have one new connection at a time.
+sub connect_fails_wait
+{
+	local $Test::Builder::Level = $Test::Builder::Level + 1;
+	my ($node, $connstr, $test_name, %params) = @_;
+
+	my $log_location = -s $node->logfile;
+
+	$node->connect_fails($connstr, $test_name, %params);
+	$node->wait_for_log(qr/DEBUG:  client backend.*exited with exit code 1/,
+		$log_location);
+	ok(1, "$test_name: client backend process exited");
+}
+
 my @sessions = ();
 my @raw_connections = ();
 
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
 push(@sessions, background_psql_as_user('regress_regular'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -60,7 +87,8 @@ $node->connect_fails(
 
 push(@sessions, background_psql_as_user('regress_reserved'));
 push(@sessions, background_psql_as_user('regress_reserved'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_regular",
 	"reserved_connections limit",
 	expected_stderr =>
@@ -68,7 +96,8 @@ $node->connect_fails(
 );
 
 push(@sessions, background_psql_as_user('regress_superuser'));
-$node->connect_fails(
+connect_fails_wait(
+	$node,
 	"dbname=postgres user=regress_superuser",
 	"superuser_reserved_connections limit",
 	expected_stderr => qr/FATAL:  sorry, too many clients already/);
-- 
2.48.1.76.g4e746b1a31.dirty


--q4d36jgrlhdhudrc
Content-Type: text/x-diff; charset=us-ascii
Content-Disposition: attachment;
	filename="v2-0004-tests-Don-t-fail-due-to-high-default-timeout-in-p.patch"



^ permalink  raw  reply  [nested|flat] 144+ messages in thread