pg_logical_slot_get_changes waits continously for a partial WAL record spanning across 2 pages

From 9029482c8dca9762b016743ca693b8e50396f253 Mon Sep 17 00:00:00 2001
From: Vignesh C <vignesh21@gmail.com>
Date: Thu, 26 Jun 2025 16:36:40 +0530
Subject: [PATCH v2] Fix infinite wait when reading partially written WAL
 record after crash

If a crash occurs while writing a WAL record that spans multiple pages,
the recovery process marks the page with the XLP_FIRST_IS_OVERWRITE_CONTRECORD
flag. However, logical decoding currently attempts to read the full WAL record
based on its expected size before checking this flag, which can lead to an
infinite wait if the remaining data is never written (e.g., no activity after crash).

This patch updates the logic to first read the page header and check for the
XLP_FIRST_IS_OVERWRITE_CONTRECORD flag before attempting to reconstruct the
full WAL record. If the flag is set, decoding correctly identifies the record
as incomplete and avoids waiting for WAL data that will never arrive.
---
 src/backend/access/transam/xlogreader.c       |  22 ++-
 src/test/recovery/meson.build                 |   1 +
 .../recovery/t/046_checkpoint_logical_slot.pl | 136 ++++++++++++++++++
 3 files changed, 154 insertions(+), 5 deletions(-)
 create mode 100644 src/test/recovery/t/046_checkpoint_logical_slot.pl

diff --git a/src/backend/access/transam/xlogreader.c b/src/backend/access/transam/xlogreader.c
index 2790ade1f91..cba00f6149b 100644
--- a/src/backend/access/transam/xlogreader.c
+++ b/src/backend/access/transam/xlogreader.c
@@ -723,11 +723,14 @@ restart:
 			/* Calculate pointer to beginning of next page */
 			targetPagePtr += XLOG_BLCKSZ;
 
-			/* Wait for the next page to become available */
-			readOff = ReadPageInternal(state, targetPagePtr,
-									   Min(total_len - gotlen + SizeOfXLogShortPHD,
-										   XLOG_BLCKSZ));
-
+			/*
+			 * Read the page header before processing the WAL record data. This
+			 * is necessary to correctly handle cases where the previous record
+			 * ended as a partial record. Attempting to read the full
+			 * record without checking the header may result in waiting
+			 * indefinitely for data that doesn't exist.
+			 */
+			readOff = ReadPageInternal(state, targetPagePtr, SizeOfXLogShortPHD);
 			if (readOff == XLREAD_WOULDBLOCK)
 				return XLREAD_WOULDBLOCK;
 			else if (readOff < 0)
@@ -775,6 +778,15 @@ restart:
 									  LSN_FORMAT_ARGS(RecPtr));
 				goto err;
 			}
+			
+			/* Wait for the next page to become available */
+			readOff = ReadPageInternal(state, targetPagePtr,
+									   Min(total_len - gotlen + SizeOfXLogShortPHD,
+										   XLOG_BLCKSZ));
+			if (readOff == XLREAD_WOULDBLOCK)
+				return XLREAD_WOULDBLOCK;
+			else if (readOff < 0)
+				goto err;
 
 			/* Append the continuation from this page to the buffer */
 			pageHeaderSize = XLogPageHeaderSize(pageHeader);
diff --git a/src/test/recovery/meson.build b/src/test/recovery/meson.build
index 6e78ff1a030..52993c32dbb 100644
--- a/src/test/recovery/meson.build
+++ b/src/test/recovery/meson.build
@@ -54,6 +54,7 @@ tests += {
       't/043_no_contrecord_switch.pl',
       't/044_invalidate_inactive_slots.pl',
       't/045_archive_restartpoint.pl',
+      't/046_checkpoint_logical_slot.pl',
       't/047_checkpoint_physical_slot.pl',
       't/048_vacuum_horizon_floor.pl'
     ],
diff --git a/src/test/recovery/t/046_checkpoint_logical_slot.pl b/src/test/recovery/t/046_checkpoint_logical_slot.pl
new file mode 100644
index 00000000000..d67c5108d78
--- /dev/null
+++ b/src/test/recovery/t/046_checkpoint_logical_slot.pl
@@ -0,0 +1,136 @@
+# Copyright (c) 2025, PostgreSQL Global Development Group
+#
+# This test verifies the case when the logical slot is advanced during
+# checkpoint. The test checks that the logical slot's restart_lsn still refers
+# to an existed WAL segment after immediate restart.
+#
+use strict;
+use warnings FATAL => 'all';
+
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+
+use Test::More;
+
+if ($ENV{enable_injection_points} ne 'yes')
+{
+	plan skip_all => 'Injection points not supported by this build';
+}
+
+my ($node, $result);
+
+$node = PostgreSQL::Test::Cluster->new('mike');
+$node->init;
+$node->append_conf('postgresql.conf', "wal_level = 'logical'");
+$node->start;
+
+# Check if the extension injection_points is available, as it may be
+# possible that this script is run with installcheck, where the module
+# would not be installed by default.
+if (!$node->check_extension('injection_points'))
+{
+	plan skip_all => 'Extension injection_points not installed';
+}
+
+$node->safe_psql('postgres', q(CREATE EXTENSION injection_points));
+
+# Create the two slots we'll need.
+$node->safe_psql('postgres',
+	q{select pg_create_logical_replication_slot('slot_logical', 'test_decoding')}
+);
+$node->safe_psql('postgres',
+	q{select pg_create_physical_replication_slot('slot_physical', true)});
+
+# Advance both slots to the current position just to have everything "valid".
+$node->safe_psql('postgres',
+	q{select count(*) from pg_logical_slot_get_changes('slot_logical', null, null)}
+);
+$node->safe_psql('postgres',
+	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
+);
+
+# Run checkpoint to flush current state to disk and set a baseline.
+$node->safe_psql('postgres', q{checkpoint});
+
+# Generate some transactions to get RUNNING_XACTS.
+my $xacts = $node->background_psql('postgres');
+$xacts->query_until(
+	qr/run_xacts/,
+	q(\echo run_xacts
+SELECT 1 \watch 0.1
+\q
+));
+
+$node->advance_wal(20);
+
+# Run another checkpoint to set a new restore LSN.
+$node->safe_psql('postgres', q{checkpoint});
+
+$node->advance_wal(20);
+
+# Run another checkpoint, this time in the background, and make it wait
+# on the injection point) so that the checkpoint stops right before
+# removing old WAL segments.
+note('starting checkpoint');
+
+my $checkpoint = $node->background_psql('postgres');
+$checkpoint->query_safe(
+	q(select injection_points_attach('checkpoint-before-old-wal-removal','wait'))
+);
+$checkpoint->query_until(
+	qr/starting_checkpoint/,
+	q(\echo starting_checkpoint
+checkpoint;
+\q
+));
+
+# Wait until the checkpoint stops right before removing WAL segments.
+note('waiting for injection_point');
+$node->wait_for_event('checkpointer', 'checkpoint-before-old-wal-removal');
+note('injection_point is reached');
+
+# Try to advance the logical slot, but make it stop when it moves to the next
+# WAL segment (this has to happen in the background, too).
+my $logical = $node->background_psql('postgres');
+$logical->query_safe(
+	q{select injection_points_attach('logical-replication-slot-advance-segment','wait');}
+);
+$logical->query_until(
+	qr/get_changes/,
+	q(
+\echo get_changes
+select count(*) from pg_logical_slot_get_changes('slot_logical', null, null) \watch 1
+\q
+));
+
+# Wait until the slot's restart_lsn points to the next WAL segment.
+note('waiting for injection_point');
+$node->wait_for_event('client backend',
+	'logical-replication-slot-advance-segment');
+note('injection_point is reached');
+
+# OK, we're in the right situation: time to advance the physical slot, which
+# recalculates the required LSN, and then unblock the checkpoint, which
+# removes the WAL still needed by the logical slot.
+$node->safe_psql('postgres',
+	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
+);
+
+# Continue the checkpoint.
+$node->safe_psql('postgres',
+	q{select injection_points_wakeup('checkpoint-before-old-wal-removal')});
+
+# Abruptly stop the server (1 second should be enough for the checkpoint
+# to finish; it would be better).
+$node->stop('immediate');
+
+$node->start;
+
+eval {
+	$node->safe_psql('postgres',
+		q{select count(*) from pg_logical_slot_get_changes('slot_logical', null, null);}
+	);
+};
+is($@, '', "Logical slot still valid");
+
+done_testing();
-- 
2.43.0

From 6824ac619506fdc762ae49726f95b1d5468c84dc Mon Sep 17 00:00:00 2001
From: Vignesh C <vignesh21@gmail.com>
Date: Sat, 28 Jun 2025 20:14:40 +0530
Subject: [PATCH v3] Fix infinite wait when reading partially written WAL
 record after crash

If a crash occurs while writing a WAL record that spans multiple pages,
the recovery process marks the page with the XLP_FIRST_IS_OVERWRITE_CONTRECORD
flag. However, logical decoding currently attempts to read the full WAL record
based on its expected size before checking this flag, which can lead to an
infinite wait if the remaining data is never written (e.g., no activity after crash).

This patch updates the logic to first read the page header and check for the
XLP_FIRST_IS_OVERWRITE_CONTRECORD flag before attempting to reconstruct the
full WAL record. If the flag is set, decoding correctly identifies the record
as incomplete and avoids waiting for WAL data that will never arrive.
---
 src/backend/access/transam/xlogreader.c | 20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)

diff --git a/src/backend/access/transam/xlogreader.c b/src/backend/access/transam/xlogreader.c
index f84569cdbab..11b1d8facaf 100644
--- a/src/backend/access/transam/xlogreader.c
+++ b/src/backend/access/transam/xlogreader.c
@@ -407,11 +407,14 @@ restart:
 			/* Calculate pointer to beginning of next page */
 			targetPagePtr += XLOG_BLCKSZ;
 
-			/* Wait for the next page to become available */
-			readOff = ReadPageInternal(state, targetPagePtr,
-									   Min(total_len - gotlen + SizeOfXLogShortPHD,
-										   XLOG_BLCKSZ));
-
+			/*
+			 * Read the page header before processing the WAL record data.
+			 * This is necessary to correctly handle cases where the previous
+			 * record ended as a partial record. Attempting to read the full
+			 * record without checking the header may result in waiting
+			 * indefinitely for data that doesn't exist.
+			 */
+			readOff = ReadPageInternal(state, targetPagePtr, SizeOfXLogShortPHD);
 			if (readOff < 0)
 				goto err;
 
@@ -459,6 +462,13 @@ restart:
 				goto err;
 			}
 
+			/* Wait for the next page to become available */
+			readOff = ReadPageInternal(state, targetPagePtr,
+									   Min(total_len - gotlen + SizeOfXLogShortPHD,
+										   XLOG_BLCKSZ));
+			if (readOff < 0)
+				goto err;
+
 			/* Append the continuation from this page to the buffer */
 			pageHeaderSize = XLogPageHeaderSize(pageHeader);
 
-- 
2.43.0

From 6d9f8c360b6ebe652260b7fa7dfcd07e54e01db3 Mon Sep 17 00:00:00 2001
From: Vignesh C <vignesh21@gmail.com>
Date: Thu, 26 Jun 2025 16:36:40 +0530
Subject: [PATCH v3] Fix infinite wait when reading partially written WAL
 record after crash

If a crash occurs while writing a WAL record that spans multiple pages,
the recovery process marks the page with the XLP_FIRST_IS_OVERWRITE_CONTRECORD
flag. However, logical decoding currently attempts to read the full WAL record
based on its expected size before checking this flag, which can lead to an
infinite wait if the remaining data is never written (e.g., no activity after crash).

This patch updates the logic to first read the page header and check for the
XLP_FIRST_IS_OVERWRITE_CONTRECORD flag before attempting to reconstruct the
full WAL record. If the flag is set, decoding correctly identifies the record
as incomplete and avoids waiting for WAL data that will never arrive.
---
 src/backend/access/transam/xlogreader.c       |  22 ++-
 src/test/recovery/meson.build                 |   1 +
 .../recovery/t/046_checkpoint_logical_slot.pl | 136 ++++++++++++++++++
 3 files changed, 154 insertions(+), 5 deletions(-)
 create mode 100644 src/test/recovery/t/046_checkpoint_logical_slot.pl

diff --git a/src/backend/access/transam/xlogreader.c b/src/backend/access/transam/xlogreader.c
index 2790ade1f91..bf4a13ba55f 100644
--- a/src/backend/access/transam/xlogreader.c
+++ b/src/backend/access/transam/xlogreader.c
@@ -723,11 +723,14 @@ restart:
 			/* Calculate pointer to beginning of next page */
 			targetPagePtr += XLOG_BLCKSZ;
 
-			/* Wait for the next page to become available */
-			readOff = ReadPageInternal(state, targetPagePtr,
-									   Min(total_len - gotlen + SizeOfXLogShortPHD,
-										   XLOG_BLCKSZ));
-
+			/*
+			 * Read the page header before processing the WAL record data.
+			 * This is necessary to correctly handle cases where the previous
+			 * record ended as a partial record. Attempting to read the full
+			 * record without checking the header may result in waiting
+			 * indefinitely for data that doesn't exist.
+			 */
+			readOff = ReadPageInternal(state, targetPagePtr, SizeOfXLogShortPHD);
 			if (readOff == XLREAD_WOULDBLOCK)
 				return XLREAD_WOULDBLOCK;
 			else if (readOff < 0)
@@ -776,6 +779,15 @@ restart:
 				goto err;
 			}
 
+			/* Wait for the next page to become available */
+			readOff = ReadPageInternal(state, targetPagePtr,
+									   Min(total_len - gotlen + SizeOfXLogShortPHD,
+										   XLOG_BLCKSZ));
+			if (readOff == XLREAD_WOULDBLOCK)
+				return XLREAD_WOULDBLOCK;
+			else if (readOff < 0)
+				goto err;
+
 			/* Append the continuation from this page to the buffer */
 			pageHeaderSize = XLogPageHeaderSize(pageHeader);
 
diff --git a/src/test/recovery/meson.build b/src/test/recovery/meson.build
index 6e78ff1a030..52993c32dbb 100644
--- a/src/test/recovery/meson.build
+++ b/src/test/recovery/meson.build
@@ -54,6 +54,7 @@ tests += {
       't/043_no_contrecord_switch.pl',
       't/044_invalidate_inactive_slots.pl',
       't/045_archive_restartpoint.pl',
+      't/046_checkpoint_logical_slot.pl',
       't/047_checkpoint_physical_slot.pl',
       't/048_vacuum_horizon_floor.pl'
     ],
diff --git a/src/test/recovery/t/046_checkpoint_logical_slot.pl b/src/test/recovery/t/046_checkpoint_logical_slot.pl
new file mode 100644
index 00000000000..d67c5108d78
--- /dev/null
+++ b/src/test/recovery/t/046_checkpoint_logical_slot.pl
@@ -0,0 +1,136 @@
+# Copyright (c) 2025, PostgreSQL Global Development Group
+#
+# This test verifies the case when the logical slot is advanced during
+# checkpoint. The test checks that the logical slot's restart_lsn still refers
+# to an existed WAL segment after immediate restart.
+#
+use strict;
+use warnings FATAL => 'all';
+
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+
+use Test::More;
+
+if ($ENV{enable_injection_points} ne 'yes')
+{
+	plan skip_all => 'Injection points not supported by this build';
+}
+
+my ($node, $result);
+
+$node = PostgreSQL::Test::Cluster->new('mike');
+$node->init;
+$node->append_conf('postgresql.conf', "wal_level = 'logical'");
+$node->start;
+
+# Check if the extension injection_points is available, as it may be
+# possible that this script is run with installcheck, where the module
+# would not be installed by default.
+if (!$node->check_extension('injection_points'))
+{
+	plan skip_all => 'Extension injection_points not installed';
+}
+
+$node->safe_psql('postgres', q(CREATE EXTENSION injection_points));
+
+# Create the two slots we'll need.
+$node->safe_psql('postgres',
+	q{select pg_create_logical_replication_slot('slot_logical', 'test_decoding')}
+);
+$node->safe_psql('postgres',
+	q{select pg_create_physical_replication_slot('slot_physical', true)});
+
+# Advance both slots to the current position just to have everything "valid".
+$node->safe_psql('postgres',
+	q{select count(*) from pg_logical_slot_get_changes('slot_logical', null, null)}
+);
+$node->safe_psql('postgres',
+	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
+);
+
+# Run checkpoint to flush current state to disk and set a baseline.
+$node->safe_psql('postgres', q{checkpoint});
+
+# Generate some transactions to get RUNNING_XACTS.
+my $xacts = $node->background_psql('postgres');
+$xacts->query_until(
+	qr/run_xacts/,
+	q(\echo run_xacts
+SELECT 1 \watch 0.1
+\q
+));
+
+$node->advance_wal(20);
+
+# Run another checkpoint to set a new restore LSN.
+$node->safe_psql('postgres', q{checkpoint});
+
+$node->advance_wal(20);
+
+# Run another checkpoint, this time in the background, and make it wait
+# on the injection point) so that the checkpoint stops right before
+# removing old WAL segments.
+note('starting checkpoint');
+
+my $checkpoint = $node->background_psql('postgres');
+$checkpoint->query_safe(
+	q(select injection_points_attach('checkpoint-before-old-wal-removal','wait'))
+);
+$checkpoint->query_until(
+	qr/starting_checkpoint/,
+	q(\echo starting_checkpoint
+checkpoint;
+\q
+));
+
+# Wait until the checkpoint stops right before removing WAL segments.
+note('waiting for injection_point');
+$node->wait_for_event('checkpointer', 'checkpoint-before-old-wal-removal');
+note('injection_point is reached');
+
+# Try to advance the logical slot, but make it stop when it moves to the next
+# WAL segment (this has to happen in the background, too).
+my $logical = $node->background_psql('postgres');
+$logical->query_safe(
+	q{select injection_points_attach('logical-replication-slot-advance-segment','wait');}
+);
+$logical->query_until(
+	qr/get_changes/,
+	q(
+\echo get_changes
+select count(*) from pg_logical_slot_get_changes('slot_logical', null, null) \watch 1
+\q
+));
+
+# Wait until the slot's restart_lsn points to the next WAL segment.
+note('waiting for injection_point');
+$node->wait_for_event('client backend',
+	'logical-replication-slot-advance-segment');
+note('injection_point is reached');
+
+# OK, we're in the right situation: time to advance the physical slot, which
+# recalculates the required LSN, and then unblock the checkpoint, which
+# removes the WAL still needed by the logical slot.
+$node->safe_psql('postgres',
+	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
+);
+
+# Continue the checkpoint.
+$node->safe_psql('postgres',
+	q{select injection_points_wakeup('checkpoint-before-old-wal-removal')});
+
+# Abruptly stop the server (1 second should be enough for the checkpoint
+# to finish; it would be better).
+$node->stop('immediate');
+
+$node->start;
+
+eval {
+	$node->safe_psql('postgres',
+		q{select count(*) from pg_logical_slot_get_changes('slot_logical', null, null);}
+	);
+};
+is($@, '', "Logical slot still valid");
+
+done_testing();
-- 
2.43.0

From c213cae3ded6467a7b56715105263ed49b2ae706 Mon Sep 17 00:00:00 2001
From: Vignesh C <vignesh21@gmail.com>
Date: Sat, 28 Jun 2025 18:53:29 +0530
Subject: [PATCH v3] Fix infinite wait when reading partially written WAL
 record after crash

If a crash occurs while writing a WAL record that spans multiple pages,
the recovery process marks the page with the XLP_FIRST_IS_OVERWRITE_CONTRECORD
flag. However, logical decoding currently attempts to read the full WAL record
based on its expected size before checking this flag, which can lead to an
infinite wait if the remaining data is never written (e.g., no activity after crash).

This patch updates the logic to first read the page header and check for the
XLP_FIRST_IS_OVERWRITE_CONTRECORD flag before attempting to reconstruct the
full WAL record. If the flag is set, decoding correctly identifies the record
as incomplete and avoids waiting for WAL data that will never arrive.
---
 src/backend/access/transam/xlogreader.c | 22 +++++++++++++++++-----
 1 file changed, 17 insertions(+), 5 deletions(-)

diff --git a/src/backend/access/transam/xlogreader.c b/src/backend/access/transam/xlogreader.c
index a07f55dfbc9..d3dcccf4949 100644
--- a/src/backend/access/transam/xlogreader.c
+++ b/src/backend/access/transam/xlogreader.c
@@ -722,11 +722,14 @@ restart:
 			/* Calculate pointer to beginning of next page */
 			targetPagePtr += XLOG_BLCKSZ;
 
-			/* Wait for the next page to become available */
-			readOff = ReadPageInternal(state, targetPagePtr,
-									   Min(total_len - gotlen + SizeOfXLogShortPHD,
-										   XLOG_BLCKSZ));
-
+			/*
+			 * Read the page header before processing the WAL record data.
+			 * This is necessary to correctly handle cases where the previous
+			 * record ended as a partial record. Attempting to read the full
+			 * record without checking the header may result in waiting
+			 * indefinitely for data that doesn't exist.
+			 */
+			readOff = ReadPageInternal(state, targetPagePtr, SizeOfXLogShortPHD);
 			if (readOff == XLREAD_WOULDBLOCK)
 				return XLREAD_WOULDBLOCK;
 			else if (readOff < 0)
@@ -775,6 +778,15 @@ restart:
 				goto err;
 			}
 
+			/* Wait for the next page to become available */
+			readOff = ReadPageInternal(state, targetPagePtr,
+									   Min(total_len - gotlen + SizeOfXLogShortPHD,
+										   XLOG_BLCKSZ));
+			if (readOff == XLREAD_WOULDBLOCK)
+				return XLREAD_WOULDBLOCK;
+			else if (readOff < 0)
+				goto err;
+
 			/* Append the continuation from this page to the buffer */
 			pageHeaderSize = XLogPageHeaderSize(pageHeader);
 
-- 
2.43.0

From 4c920861f07e0f4e6e21e24e5f6a9060054c3232 Mon Sep 17 00:00:00 2001
From: Michael Paquier <michael@paquier.xyz>
Date: Tue, 15 Jul 2025 12:01:01 +0900
Subject: [PATCH v4 1/3] Fix infinite wait when reading partially-written WAL
 record

If a crash occurs while writing a WAL record that spans multiple pages,
the recovery process marks the page with the
XLP_FIRST_IS_OVERWRITE_CONTRECORD flag.  However, logical decoding
currently attempts to read the full WAL record based on its expected
size before checking this flag, which can lead to an infinite wait if
the remaining data is never written (e.g., no activity after crash).

This patch updates the logic to first read the page header and check for
the XLP_FIRST_IS_OVERWRITE_CONTRECORD flag before attempting to
reconstruct the full WAL record.  If the flag is set, decoding correctly
identifies the record as incomplete and avoids waiting for WAL data that
will never arrive.
---
 src/backend/access/transam/xlogreader.c | 20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)

diff --git a/src/backend/access/transam/xlogreader.c b/src/backend/access/transam/xlogreader.c
index ac1f801b1eb0..2e5361ff4fdd 100644
--- a/src/backend/access/transam/xlogreader.c
+++ b/src/backend/access/transam/xlogreader.c
@@ -723,11 +723,12 @@ restart:
 			/* Calculate pointer to beginning of next page */
 			targetPagePtr += XLOG_BLCKSZ;
 
-			/* Wait for the next page to become available */
-			readOff = ReadPageInternal(state, targetPagePtr,
-									   Min(total_len - gotlen + SizeOfXLogShortPHD,
-										   XLOG_BLCKSZ));
-
+			/*
+			 * Read the page header before processing the record data, to be
+			 * able to handle the case where the previous record ended as
+			 * being a partial one.
+			 */
+			readOff = ReadPageInternal(state, targetPagePtr, SizeOfXLogShortPHD);
 			if (readOff == XLREAD_WOULDBLOCK)
 				return XLREAD_WOULDBLOCK;
 			else if (readOff < 0)
@@ -776,6 +777,15 @@ restart:
 				goto err;
 			}
 
+			/* Wait for the next page to become available */
+			readOff = ReadPageInternal(state, targetPagePtr,
+									   Min(total_len - gotlen + SizeOfXLogShortPHD,
+										   XLOG_BLCKSZ));
+			if (readOff == XLREAD_WOULDBLOCK)
+				return XLREAD_WOULDBLOCK;
+			else if (readOff < 0)
+				goto err;
+
 			/* Append the continuation from this page to the buffer */
 			pageHeaderSize = XLogPageHeaderSize(pageHeader);
 
-- 
2.50.0

From 68fab6348a60320ff4dda943d69dc3388a5754e2 Mon Sep 17 00:00:00 2001
From: Michael Paquier <michael@paquier.xyz>
Date: Tue, 15 Jul 2025 12:02:08 +0900
Subject: [PATCH v4 2/3] Improve stability of recovery test
 047_checkpoint_physical_slot

Author: Alexander Korotkov <akorotkov@postgresql.org>
Backpatch-through: 17
---
 src/test/recovery/t/047_checkpoint_physical_slot.pl | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/src/test/recovery/t/047_checkpoint_physical_slot.pl b/src/test/recovery/t/047_checkpoint_physical_slot.pl
index a1332b5d44cb..9e98383e30ea 100644
--- a/src/test/recovery/t/047_checkpoint_physical_slot.pl
+++ b/src/test/recovery/t/047_checkpoint_physical_slot.pl
@@ -94,9 +94,11 @@ $node->safe_psql('postgres',
 	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
 );
 
-# Continue the checkpoint.
+# Continue the checkpoint and wait for its completion.
+my $log_offset = -s $node->logfile;
 $node->safe_psql('postgres',
 	q{select injection_points_wakeup('checkpoint-before-old-wal-removal')});
+$node->wait_for_log(qr/checkpoint complete/, $log_offset);
 
 my $restart_lsn_old = $node->safe_psql('postgres',
 	q{select restart_lsn from pg_replication_slots where slot_name = 'slot_physical'}
@@ -104,8 +106,7 @@ my $restart_lsn_old = $node->safe_psql('postgres',
 chomp($restart_lsn_old);
 note("restart lsn before stop: $restart_lsn_old");
 
-# Abruptly stop the server (1 second should be enough for the checkpoint
-# to finish; it would be better).
+# Abruptly stop the server.
 $node->stop('immediate');
 
 $node->start;
-- 
2.50.0

From 4584664c18bb63a604c6d20eb22831ec6e59a49a Mon Sep 17 00:00:00 2001
From: Michael Paquier <michael@paquier.xyz>
Date: Tue, 15 Jul 2025 12:07:50 +0900
Subject: [PATCH v4 3/3] Reintroduce test 046_checkpoint_logical_slot

This patch is only for HEAD and v18 where the test has been removed,
improvements to the checkpoint completion wait and the extra
pg_logical_emit_message() call to force the creation of a WAL record
that spawns through at least 2 pages needs to be backpatched down to v17
to improve the stability and coverage of the original test.

Author: Alexander Korotkov <akorotkov@postgresql.org>
Co-authored-by: Hayato Kuroda <kuroda.hayato@fujitsu.com>
---
 src/test/recovery/meson.build                 |   1 +
 .../recovery/t/046_checkpoint_logical_slot.pl | 142 ++++++++++++++++++
 2 files changed, 143 insertions(+)
 create mode 100644 src/test/recovery/t/046_checkpoint_logical_slot.pl

diff --git a/src/test/recovery/meson.build b/src/test/recovery/meson.build
index 6e78ff1a030b..52993c32dbba 100644
--- a/src/test/recovery/meson.build
+++ b/src/test/recovery/meson.build
@@ -54,6 +54,7 @@ tests += {
       't/043_no_contrecord_switch.pl',
       't/044_invalidate_inactive_slots.pl',
       't/045_archive_restartpoint.pl',
+      't/046_checkpoint_logical_slot.pl',
       't/047_checkpoint_physical_slot.pl',
       't/048_vacuum_horizon_floor.pl'
     ],
diff --git a/src/test/recovery/t/046_checkpoint_logical_slot.pl b/src/test/recovery/t/046_checkpoint_logical_slot.pl
new file mode 100644
index 000000000000..bdfca3ddf826
--- /dev/null
+++ b/src/test/recovery/t/046_checkpoint_logical_slot.pl
@@ -0,0 +1,142 @@
+# Copyright (c) 2025, PostgreSQL Global Development Group
+#
+# This test verifies the case when the logical slot is advanced during
+# checkpoint. The test checks that the logical slot's restart_lsn still refers
+# to an existed WAL segment after immediate restart.
+
+use strict;
+use warnings FATAL => 'all';
+
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+
+use Test::More;
+
+if ($ENV{enable_injection_points} ne 'yes')
+{
+	plan skip_all => 'Injection points not supported by this build';
+}
+
+my ($node, $result);
+
+$node = PostgreSQL::Test::Cluster->new('mike');
+$node->init;
+$node->append_conf('postgresql.conf', "wal_level = 'logical'");
+$node->start;
+
+# Check if the extension injection_points is available, as it may be
+# possible that this script is run with installcheck, where the module
+# would not be installed by default.
+if (!$node->check_extension('injection_points'))
+{
+	plan skip_all => 'Extension injection_points not installed';
+}
+
+$node->safe_psql('postgres', q(CREATE EXTENSION injection_points));
+
+# Create the two slots we'll need.
+$node->safe_psql('postgres',
+	q{select pg_create_logical_replication_slot('slot_logical', 'test_decoding')}
+);
+$node->safe_psql('postgres',
+	q{select pg_create_physical_replication_slot('slot_physical', true)});
+
+# Advance both slots to the current position just to have everything "valid".
+$node->safe_psql('postgres',
+	q{select count(*) from pg_logical_slot_get_changes('slot_logical', null, null)}
+);
+$node->safe_psql('postgres',
+	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
+);
+
+# Run checkpoint to flush current state to disk and set a baseline.
+$node->safe_psql('postgres', q{checkpoint});
+
+# Generate some transactions to get RUNNING_XACTS.
+my $xacts = $node->background_psql('postgres');
+$xacts->query_until(
+	qr/run_xacts/,
+	q(\echo run_xacts
+SELECT 1 \watch 0.1
+\q
+));
+
+$node->advance_wal(20);
+
+# Run another checkpoint to set a new restore LSN.
+$node->safe_psql('postgres', q{checkpoint});
+
+$node->advance_wal(20);
+
+# Run another checkpoint, this time in the background, and make it wait
+# on the injection point) so that the checkpoint stops right before
+# removing old WAL segments.
+note('starting checkpoint');
+
+my $checkpoint = $node->background_psql('postgres');
+$checkpoint->query_safe(
+	q(select injection_points_attach('checkpoint-before-old-wal-removal','wait'))
+);
+$checkpoint->query_until(
+	qr/starting_checkpoint/,
+	q(\echo starting_checkpoint
+checkpoint;
+\q
+));
+
+# Wait until the checkpoint stops right before removing WAL segments.
+note('waiting for injection_point');
+$node->wait_for_event('checkpointer', 'checkpoint-before-old-wal-removal');
+note('injection_point is reached');
+
+# Try to advance the logical slot, but make it stop when it moves to the next
+# WAL segment (this has to happen in the background, too).
+my $logical = $node->background_psql('postgres');
+$logical->query_safe(
+	q{select injection_points_attach('logical-replication-slot-advance-segment','wait');}
+);
+$logical->query_until(
+	qr/get_changes/,
+	q(
+\echo get_changes
+select count(*) from pg_logical_slot_get_changes('slot_logical', null, null) \watch 1
+\q
+));
+
+# Wait until the slot's restart_lsn points to the next WAL segment.
+note('waiting for injection_point');
+$node->wait_for_event('client backend',
+	'logical-replication-slot-advance-segment');
+note('injection_point is reached');
+
+# OK, we're in the right situation: time to advance the physical slot, which
+# recalculates the required LSN, and then unblock the checkpoint, which
+# removes the WAL still needed by the logical slot.
+$node->safe_psql('postgres',
+	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
+);
+
+# Generate a long WAL record, spawning at least two pages for the follow-up
+# post-recovery check.
+$node->safe_psql('postgres',
+	q{select pg_logical_emit_message(false, '', repeat('123456789', 1000))});
+
+# Continue the checkpoint and wait for its completion.
+my $log_offset = -s $node->logfile;
+$node->safe_psql('postgres',
+	q{select injection_points_wakeup('checkpoint-before-old-wal-removal')});
+$node->wait_for_log(qr/checkpoint complete/, $log_offset);
+
+# Abruptly stop the server.
+$node->stop('immediate');
+
+$node->start;
+
+eval {
+	$node->safe_psql('postgres',
+		q{select count(*) from pg_logical_slot_get_changes('slot_logical', null, null);}
+	);
+};
+is($@, '', "Logical slot still valid");
+
+done_testing();
-- 
2.50.0

From c09e67a8e074a3235499bcf891487fdc1293072d Mon Sep 17 00:00:00 2001
From: Michael Paquier <michael@paquier.xyz>
Date: Tue, 15 Jul 2025 12:25:37 +0900
Subject: [PATCH v4] Improve recovery test 046_checkpoint_logical_slot

This commit improves 046_checkpoint_logical_slot on two aspects:
- Add one pg_logical_emit_message() call, to force the creation of a
record that spawns across two pages.
- Make the logic wait for the checkpoint completion.

Author: Alexander Korotkov <akorotkov@postgresql.org>
Co-authored-by: Hayato Kuroda <kuroda.hayato@fujitsu.com>
---
 src/test/recovery/t/046_checkpoint_logical_slot.pl | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/src/test/recovery/t/046_checkpoint_logical_slot.pl b/src/test/recovery/t/046_checkpoint_logical_slot.pl
index 0468d4609e43..d2cf1cb4464e 100644
--- a/src/test/recovery/t/046_checkpoint_logical_slot.pl
+++ b/src/test/recovery/t/046_checkpoint_logical_slot.pl
@@ -119,12 +119,18 @@ $node->safe_psql('postgres',
 	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
 );
 
+# Generate a long WAL record, spawning at least two pages for the follow-up
+# post-recovery check.
+$node->safe_psql('postgres',
+	q{select pg_logical_emit_message(false, '', repeat('123456789', 1000))});
+
 # Continue the checkpoint.
+my $log_offset = -s $node->logfile;
 $node->safe_psql('postgres',
 	q{select injection_points_wakeup('checkpoint-before-old-wal-removal')});
+$node->wait_for_log(qr/checkpoint complete/, $log_offset);
 
-# Abruptly stop the server (1 second should be enough for the checkpoint
-# to finish; it would be better).
+# Abruptly stop the server.
 $node->stop('immediate');
 
 $node->start;
-- 
2.50.0

From 753aa3c063c9fd9f690583cf0a7bde19624e69d1 Mon Sep 17 00:00:00 2001
From: Alexander Korotkov <akorotkov@postgresql.org>
Date: Fri, 18 Jul 2025 00:04:52 +0300
Subject: [PATCH v5-PG17] Improve recovery test 046_checkpoint_logical_slot

This commit improves 046_checkpoint_logical_slot in two aspects:
- Add one pg_logical_emit_message() call to force the creation of a record
  that spawns across two pages.
- Make the logic wait for the checkpoint completion.

Discussion: https://postgr.es/m/CALDaNm34m36PDHzsU_GdcNXU0gLTfFY5rzh9GSQv%3Dw6B%2BQVNRQ%40mail.gmail.com
Author: Alexander Korotkov <akorotkov@postgresql.org>
Co-authored-by: Hayato Kuroda <kuroda.hayato@fujitsu.com>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
---
 src/test/recovery/t/046_checkpoint_logical_slot.pl | 10 ++++++++--
 1 file changed, 8 insertions(+), 2 deletions(-)

diff --git a/src/test/recovery/t/046_checkpoint_logical_slot.pl b/src/test/recovery/t/046_checkpoint_logical_slot.pl
index 0468d4609e4..d2cf1cb4464 100644
--- a/src/test/recovery/t/046_checkpoint_logical_slot.pl
+++ b/src/test/recovery/t/046_checkpoint_logical_slot.pl
@@ -119,12 +119,18 @@ $node->safe_psql('postgres',
 	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
 );
 
+# Generate a long WAL record, spawning at least two pages for the follow-up
+# post-recovery check.
+$node->safe_psql('postgres',
+	q{select pg_logical_emit_message(false, '', repeat('123456789', 1000))});
+
 # Continue the checkpoint.
+my $log_offset = -s $node->logfile;
 $node->safe_psql('postgres',
 	q{select injection_points_wakeup('checkpoint-before-old-wal-removal')});
+$node->wait_for_log(qr/checkpoint complete/, $log_offset);
 
-# Abruptly stop the server (1 second should be enough for the checkpoint
-# to finish; it would be better).
+# Abruptly stop the server.
 $node->stop('immediate');
 
 $node->start;
-- 
2.39.5 (Apple Git-154)

From 8de05d7864af9263ca5d14182d8b57407cb2ff95 Mon Sep 17 00:00:00 2001
From: Alexander Korotkov <akorotkov@postgresql.org>
Date: Thu, 17 Jul 2025 23:42:57 +0300
Subject: [PATCH v5 2/3] Improve the stability of the recovery test
 047_checkpoint_physical_slot

Currently, the comments in 047_checkpoint_physical_slot. It shows an
incomplete intention to wait for checkpoint completion before performing
an immediate database stop.  However, an immediate node stop can occur both
before and after checkpoint completion.  Both cases should work correctly.
But we would like the test to be more stable and deterministic.  This is why
this commit makes this test explicitly wait for the checkpoint completion
log message.

Discussion: https://postgr.es/m/CAPpHfdurV-j_e0pb%3DUFENAy3tyzxfF%2ByHveNDNQk2gM82WBU5A%40mail.gmail.com
Discussion: https://postgr.es/m/aHXLep3OaX_vRTNQ%40paquier.xyz
Author: Alexander Korotkov <akorotkov@postgresql.org>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Backpatch-through: 17
---
 src/test/recovery/t/047_checkpoint_physical_slot.pl | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

diff --git a/src/test/recovery/t/047_checkpoint_physical_slot.pl b/src/test/recovery/t/047_checkpoint_physical_slot.pl
index a1332b5d44c..9e98383e30e 100644
--- a/src/test/recovery/t/047_checkpoint_physical_slot.pl
+++ b/src/test/recovery/t/047_checkpoint_physical_slot.pl
@@ -94,9 +94,11 @@ $node->safe_psql('postgres',
 	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
 );
 
-# Continue the checkpoint.
+# Continue the checkpoint and wait for its completion.
+my $log_offset = -s $node->logfile;
 $node->safe_psql('postgres',
 	q{select injection_points_wakeup('checkpoint-before-old-wal-removal')});
+$node->wait_for_log(qr/checkpoint complete/, $log_offset);
 
 my $restart_lsn_old = $node->safe_psql('postgres',
 	q{select restart_lsn from pg_replication_slots where slot_name = 'slot_physical'}
@@ -104,8 +106,7 @@ my $restart_lsn_old = $node->safe_psql('postgres',
 chomp($restart_lsn_old);
 note("restart lsn before stop: $restart_lsn_old");
 
-# Abruptly stop the server (1 second should be enough for the checkpoint
-# to finish; it would be better).
+# Abruptly stop the server.
 $node->stop('immediate');
 
 $node->start;
-- 
2.39.5 (Apple Git-154)

From 629b7c23501465d8eac186f8f6a1f76f39e34780 Mon Sep 17 00:00:00 2001
From: Alexander Korotkov <akorotkov@postgresql.org>
Date: Thu, 17 Jul 2025 23:40:28 +0300
Subject: [PATCH v5 1/3] Fix infinite wait when reading a partially written WAL
 record

If a crash occurs while writing a WAL record that spans multiple pages, the
recovery process marks the page with the XLP_FIRST_IS_OVERWRITE_CONTRECORD
flag.  However, logical decoding currently attempts to read the full WAL
record based on its expected size before checking this flag, which can lead
to an infinite wait if the remaining data is never written (e.g., no activity
after crash).

This patch updates the logic first to read the page header and check for
the XLP_FIRST_IS_OVERWRITE_CONTRECORD flag before attempting to reconstruct
the full WAL record.  If the flag is set, decoding correctly identifies
the record as incomplete and avoids waiting for WAL data that will never
arrive.

Discussion: https://postgr.es/m/CAAKRu_ZCOzQpEumLFgG_%2Biw3FTa%2BhJ4SRpxzaQBYxxM_ZAzWcA%40mail.gmail.com
Discussion: https://postgr.es/m/CALDaNm34m36PDHzsU_GdcNXU0gLTfFY5rzh9GSQv%3Dw6B%2BQVNRQ%40mail.gmail.com
Author: Vignesh C <vignesh21@gmail.com>
Reviewed-by: Hayato Kuroda <kuroda.hayato@fujitsu.com>
Reviewed-by: Dilip Kumar <dilipbalaut@gmail.com>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
Reviewed-by: Alexander Korotkov <aekorotkov@gmail.com>
---
 src/backend/access/transam/xlogreader.c | 20 +++++++++++++++-----
 1 file changed, 15 insertions(+), 5 deletions(-)

diff --git a/src/backend/access/transam/xlogreader.c b/src/backend/access/transam/xlogreader.c
index ac1f801b1eb..dcc8d4f9c1b 100644
--- a/src/backend/access/transam/xlogreader.c
+++ b/src/backend/access/transam/xlogreader.c
@@ -723,11 +723,12 @@ restart:
 			/* Calculate pointer to beginning of next page */
 			targetPagePtr += XLOG_BLCKSZ;
 
-			/* Wait for the next page to become available */
-			readOff = ReadPageInternal(state, targetPagePtr,
-									   Min(total_len - gotlen + SizeOfXLogShortPHD,
-										   XLOG_BLCKSZ));
-
+			/*
+			 * Read the page header before processing the record data, so we
+			 * can handle the case where the previous record ended as being a
+			 * partial one.
+			 */
+			readOff = ReadPageInternal(state, targetPagePtr, SizeOfXLogShortPHD);
 			if (readOff == XLREAD_WOULDBLOCK)
 				return XLREAD_WOULDBLOCK;
 			else if (readOff < 0)
@@ -776,6 +777,15 @@ restart:
 				goto err;
 			}
 
+			/* Wait for the next page to become available */
+			readOff = ReadPageInternal(state, targetPagePtr,
+									   Min(total_len - gotlen + SizeOfXLogShortPHD,
+										   XLOG_BLCKSZ));
+			if (readOff == XLREAD_WOULDBLOCK)
+				return XLREAD_WOULDBLOCK;
+			else if (readOff < 0)
+				goto err;
+
 			/* Append the continuation from this page to the buffer */
 			pageHeaderSize = XLogPageHeaderSize(pageHeader);
 
-- 
2.39.5 (Apple Git-154)

From d5a4c098a4f7e7bf68b13805e554a15fe997ac3a Mon Sep 17 00:00:00 2001
From: Alexander Korotkov <akorotkov@postgresql.org>
Date: Fri, 18 Jul 2025 00:05:22 +0300
Subject: [PATCH v5 3/3] Reintroduce test 046_checkpoint_logical_slot

This patch is only for HEAD and v18, where the test has been removed,
improvements to the checkpoint completion wait, and the extra
pg_logical_emit_message() call to force the creation of a WAL record that
spawns through at least two pages need to be backpatched down to v17 to
improve the stability and coverage of the original test.

Discussion: https://postgr.es/m/CAAKRu_ZCOzQpEumLFgG_%2Biw3FTa%2BhJ4SRpxzaQBYxxM_ZAzWcA%40mail.gmail.com
Discussion: https://postgr.es/m/CALDaNm34m36PDHzsU_GdcNXU0gLTfFY5rzh9GSQv%3Dw6B%2BQVNRQ%40mail.gmail.com
Author: Alexander Korotkov <akorotkov@postgresql.org>
Co-authored-by: Hayato Kuroda <kuroda.hayato@fujitsu.com>
Reviewed-by: Michael Paquier <michael@paquier.xyz>
---
 src/test/recovery/meson.build                 |   1 +
 .../recovery/t/046_checkpoint_logical_slot.pl | 142 ++++++++++++++++++
 2 files changed, 143 insertions(+)
 create mode 100644 src/test/recovery/t/046_checkpoint_logical_slot.pl

diff --git a/src/test/recovery/meson.build b/src/test/recovery/meson.build
index 6e78ff1a030..52993c32dbb 100644
--- a/src/test/recovery/meson.build
+++ b/src/test/recovery/meson.build
@@ -54,6 +54,7 @@ tests += {
       't/043_no_contrecord_switch.pl',
       't/044_invalidate_inactive_slots.pl',
       't/045_archive_restartpoint.pl',
+      't/046_checkpoint_logical_slot.pl',
       't/047_checkpoint_physical_slot.pl',
       't/048_vacuum_horizon_floor.pl'
     ],
diff --git a/src/test/recovery/t/046_checkpoint_logical_slot.pl b/src/test/recovery/t/046_checkpoint_logical_slot.pl
new file mode 100644
index 00000000000..bdfca3ddf82
--- /dev/null
+++ b/src/test/recovery/t/046_checkpoint_logical_slot.pl
@@ -0,0 +1,142 @@
+# Copyright (c) 2025, PostgreSQL Global Development Group
+#
+# This test verifies the case when the logical slot is advanced during
+# checkpoint. The test checks that the logical slot's restart_lsn still refers
+# to an existed WAL segment after immediate restart.
+
+use strict;
+use warnings FATAL => 'all';
+
+use PostgreSQL::Test::Cluster;
+use PostgreSQL::Test::Utils;
+
+use Test::More;
+
+if ($ENV{enable_injection_points} ne 'yes')
+{
+	plan skip_all => 'Injection points not supported by this build';
+}
+
+my ($node, $result);
+
+$node = PostgreSQL::Test::Cluster->new('mike');
+$node->init;
+$node->append_conf('postgresql.conf', "wal_level = 'logical'");
+$node->start;
+
+# Check if the extension injection_points is available, as it may be
+# possible that this script is run with installcheck, where the module
+# would not be installed by default.
+if (!$node->check_extension('injection_points'))
+{
+	plan skip_all => 'Extension injection_points not installed';
+}
+
+$node->safe_psql('postgres', q(CREATE EXTENSION injection_points));
+
+# Create the two slots we'll need.
+$node->safe_psql('postgres',
+	q{select pg_create_logical_replication_slot('slot_logical', 'test_decoding')}
+);
+$node->safe_psql('postgres',
+	q{select pg_create_physical_replication_slot('slot_physical', true)});
+
+# Advance both slots to the current position just to have everything "valid".
+$node->safe_psql('postgres',
+	q{select count(*) from pg_logical_slot_get_changes('slot_logical', null, null)}
+);
+$node->safe_psql('postgres',
+	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
+);
+
+# Run checkpoint to flush current state to disk and set a baseline.
+$node->safe_psql('postgres', q{checkpoint});
+
+# Generate some transactions to get RUNNING_XACTS.
+my $xacts = $node->background_psql('postgres');
+$xacts->query_until(
+	qr/run_xacts/,
+	q(\echo run_xacts
+SELECT 1 \watch 0.1
+\q
+));
+
+$node->advance_wal(20);
+
+# Run another checkpoint to set a new restore LSN.
+$node->safe_psql('postgres', q{checkpoint});
+
+$node->advance_wal(20);
+
+# Run another checkpoint, this time in the background, and make it wait
+# on the injection point) so that the checkpoint stops right before
+# removing old WAL segments.
+note('starting checkpoint');
+
+my $checkpoint = $node->background_psql('postgres');
+$checkpoint->query_safe(
+	q(select injection_points_attach('checkpoint-before-old-wal-removal','wait'))
+);
+$checkpoint->query_until(
+	qr/starting_checkpoint/,
+	q(\echo starting_checkpoint
+checkpoint;
+\q
+));
+
+# Wait until the checkpoint stops right before removing WAL segments.
+note('waiting for injection_point');
+$node->wait_for_event('checkpointer', 'checkpoint-before-old-wal-removal');
+note('injection_point is reached');
+
+# Try to advance the logical slot, but make it stop when it moves to the next
+# WAL segment (this has to happen in the background, too).
+my $logical = $node->background_psql('postgres');
+$logical->query_safe(
+	q{select injection_points_attach('logical-replication-slot-advance-segment','wait');}
+);
+$logical->query_until(
+	qr/get_changes/,
+	q(
+\echo get_changes
+select count(*) from pg_logical_slot_get_changes('slot_logical', null, null) \watch 1
+\q
+));
+
+# Wait until the slot's restart_lsn points to the next WAL segment.
+note('waiting for injection_point');
+$node->wait_for_event('client backend',
+	'logical-replication-slot-advance-segment');
+note('injection_point is reached');
+
+# OK, we're in the right situation: time to advance the physical slot, which
+# recalculates the required LSN, and then unblock the checkpoint, which
+# removes the WAL still needed by the logical slot.
+$node->safe_psql('postgres',
+	q{select pg_replication_slot_advance('slot_physical', pg_current_wal_lsn())}
+);
+
+# Generate a long WAL record, spawning at least two pages for the follow-up
+# post-recovery check.
+$node->safe_psql('postgres',
+	q{select pg_logical_emit_message(false, '', repeat('123456789', 1000))});
+
+# Continue the checkpoint and wait for its completion.
+my $log_offset = -s $node->logfile;
+$node->safe_psql('postgres',
+	q{select injection_points_wakeup('checkpoint-before-old-wal-removal')});
+$node->wait_for_log(qr/checkpoint complete/, $log_offset);
+
+# Abruptly stop the server.
+$node->stop('immediate');
+
+$node->start;
+
+eval {
+	$node->safe_psql('postgres',
+		q{select count(*) from pg_logical_slot_get_changes('slot_logical', null, null);}
+	);
+};
+is($@, '', "Logical slot still valid");
+
+done_testing();
-- 
2.39.5 (Apple Git-154)