0001-reproduce-BF-failure-in-040_standby_failover_slots_s.patch

From fd1c96dcb3d2aeba279c605a8db960c7d0e7a0ad Mon Sep 17 00:00:00 2001
From: Zhijie Hou <houzj.fnst@cn.fujitsu.com>
Date: Tue, 6 May 2025 12:02:17 +0800
Subject: [PATCH] reproduce BF failure in 040_standby_failover_slots_sync

---
 src/backend/replication/logical/logical.c     |  3 +++
 src/backend/replication/walsender.c           |  4 +++
 .../t/040_standby_failover_slots_sync.pl      | 27 ++++++++++++++++++-
 3 files changed, 33 insertions(+), 1 deletion(-)

diff --git a/src/backend/replication/logical/logical.c b/src/backend/replication/logical/logical.c
index a8d2e024d34..f7cc4cabd49 100644
--- a/src/backend/replication/logical/logical.c
+++ b/src/backend/replication/logical/logical.c
@@ -1819,6 +1819,9 @@ LogicalConfirmReceivedLocation(XLogRecPtr lsn)
 {
 	Assert(lsn != InvalidXLogRecPtr);
 
+	if (lsn < MyReplicationSlot->data.confirmed_flush)
+		elog(LOG, "confirmed_flush move backwards");
+
 	/* Do an unlocked check for candidate_lsn first. */
 	if (MyReplicationSlot->candidate_xmin_lsn != InvalidXLogRecPtr ||
 		MyReplicationSlot->candidate_restart_valid != InvalidXLogRecPtr)
diff --git a/src/backend/replication/walsender.c b/src/backend/replication/walsender.c
index 9fa8beb6103..c74dea40815 100644
--- a/src/backend/replication/walsender.c
+++ b/src/backend/replication/walsender.c
@@ -95,6 +95,7 @@
 #include "utils/ps_status.h"
 #include "utils/timeout.h"
 #include "utils/timestamp.h"
+#include "utils/injection_point.h"
 
 /* Minimum interval used by walsender for stats flushes, in ms */
 #define WALSENDER_STATS_FLUSH_INTERVAL         1000
@@ -2810,6 +2811,9 @@ WalSndLoop(WalSndSendDataCallback send_data)
 			SyncRepInitConfig();
 		}
 
+		if (send_data == XLogSendLogical)
+			INJECTION_POINT("process-replies");
+
 		/* Check for input from the client */
 		ProcessRepliesIfAny();
 
diff --git a/src/test/recovery/t/040_standby_failover_slots_sync.pl b/src/test/recovery/t/040_standby_failover_slots_sync.pl
index 9c8b49e942d..441fd77a157 100644
--- a/src/test/recovery/t/040_standby_failover_slots_sync.pl
+++ b/src/test/recovery/t/040_standby_failover_slots_sync.pl
@@ -841,9 +841,11 @@ $primary->reload;
 # option; subsequent tests will verify if it can be correctly replicated to the
 # subscriber after committing it on the promoted standby.
 ##################################################
-
+$standby1->append_conf('postgresql.conf', qq(sync_replication_slots = off));
 $standby1->start;
 
+$standby1->safe_psql('postgres', "SELECT pg_sync_replication_slots();");
+
 # Prepare a transaction
 $primary->safe_psql(
 	'postgres', qq[
@@ -865,6 +867,12 @@ $primary->poll_query_until(
 	"SELECT COUNT(*) FROM pg_catalog.pg_replication_slots WHERE slot_name = 'lsub1_slot' AND active='f'",
 	1);
 
+$primary->safe_psql(
+	'postgres', qq[
+	CREATE EXTENSION injection_points;
+	SELECT injection_points_attach('process-replies', 'wait');
+]);
+
 # Set two_phase to true and enable the subscription
 $subscriber1->safe_psql(
 	'postgres', qq[
@@ -872,6 +880,23 @@ $subscriber1->safe_psql(
 	ALTER SUBSCRIPTION regress_mysub1 ENABLE;
 ]);
 
+sleep 1;
+
+$primary->safe_psql('postgres',
+	"SELECT injection_points_wakeup('process-replies');");
+
+$standby1->append_conf('postgresql.conf', qq(sync_replication_slots = on));
+$standby1->reload;
+
+
+sleep 2;
+
+$primary->safe_psql(
+	'postgres', qq[
+	SELECT injection_points_detach('process-replies');
+	SELECT injection_points_wakeup('process-replies');
+]);
+
 $primary->wait_for_catchup('regress_mysub1');
 
 my $two_phase_at = $primary->safe_psql('postgres',
-- 
2.31.1