v1-0001-Temporary-patch-to-help-debug-pg_walsummary-test-.patch

From aa9a129de3a7860c9d1db239ee4c23940eb40759 Mon Sep 17 00:00:00 2001
From: Robert Haas <rhaas@postgresql.org>
Date: Thu, 25 Jan 2024 10:02:41 -0500
Subject: [PATCH v1] Temporary patch to help debug pg_walsummary test failures.

The tests in 002_blocks.pl are failing in the buildfarm from time to
time, but we don't know how to reproduce the failure elsewhere. The
most obvious explanation seems to be the unexpected disappearance of a
WAL summary file, so bump up the logging level in
RemoveWalSummaryIfOlderThan to try to help us spot such problems. Also
adjust 002_blocks.pl to dump out a directory listing of the relevant
directory at various points.

This patch should be reverted once we sort out what's happening here.
---
 src/backend/backup/walsummary.c       |  3 ++-
 src/bin/pg_walsummary/t/002_blocks.pl | 14 ++++++++++++++
 2 files changed, 16 insertions(+), 1 deletion(-)

diff --git a/src/backend/backup/walsummary.c b/src/backend/backup/walsummary.c
index b549673a9d..bef75a9b47 100644
--- a/src/backend/backup/walsummary.c
+++ b/src/backend/backup/walsummary.c
@@ -251,7 +251,8 @@ RemoveWalSummaryIfOlderThan(WalSummaryFile *ws, time_t cutoff_time)
 		ereport(ERROR,
 				(errcode_for_file_access(),
 				 errmsg("could not stat file \"%s\": %m", path)));
-	ereport(DEBUG2,
+	/* XXX: was DEBUG2, temporarily increased to LOG */
+	ereport(LOG,
 			(errmsg_internal("removing file \"%s\"", path)));
 }
 
diff --git a/src/bin/pg_walsummary/t/002_blocks.pl b/src/bin/pg_walsummary/t/002_blocks.pl
index d609d2c547..40908da8cb 100644
--- a/src/bin/pg_walsummary/t/002_blocks.pl
+++ b/src/bin/pg_walsummary/t/002_blocks.pl
@@ -48,6 +48,7 @@ SELECT summarized_tli, summarized_lsn FROM pg_get_wal_summarizer_state()
 EOM
 ($summarized_tli, $summarized_lsn) = split(/\|/, $progress);
 note("after insert, summarized TLI $summarized_tli through $summarized_lsn");
+note_wal_summary_dir("after insert", $node1);
 
 # Update a row in the first block of the table and trigger a checkpoint.
 $node1->safe_psql('postgres', <<EOM);
@@ -70,6 +71,7 @@ SELECT tli, start_lsn, end_lsn from pg_available_wal_summaries()
 EOM
 my ($tli, $start_lsn, $end_lsn) = split(/\|/, $details);
 note("examining summary for TLI $tli from $start_lsn to $end_lsn");
+note_wal_summary_dir("after new summary", $node1);
 
 # Reconstruct the full pathname for the WAL summary file.
 my $filename = sprintf "%s/pg_wal/summaries/%08s%08s%08s%08s%08s.summary",
@@ -77,6 +79,7 @@ my $filename = sprintf "%s/pg_wal/summaries/%08s%08s%08s%08s%08s.summary",
 					   split(m@/@, $start_lsn),
 					   split(m@/@, $end_lsn);
 ok(-f $filename, "WAL summary file exists");
+note_wal_summary_dir("after existence check", $node1);
 
 # Run pg_walsummary on it. We expect block 0 to be modified, but depending
 # on where the new tuple ends up, block 1 might also be modified, so we
@@ -84,5 +87,16 @@ ok(-f $filename, "WAL summary file exists");
 my ($stdout, $stderr) = run_command([ 'pg_walsummary', '-i', $filename ]);
 like($stdout, qr/FORK main: block 0$/m, "stdout shows block 0 modified");
 is($stderr, '', 'stderr is empty');
+note_wal_summary_dir("after pg_walsummary run", $node1);
 
 done_testing();
+
+# XXX. Temporary debugging code.
+sub note_wal_summary_dir
+{
+	my ($flair, $node) = @_;
+
+	my $wsdir = sprintf "%s/pg_wal/summaries", $node->data_dir;
+	my @wsfiles = grep { $_ ne '.' && $_ ne '..' } slurp_dir($wsdir);
+	note("$flair pg_wal/summaries has: @wsfiles");
+}
-- 
2.39.3 (Apple Git-145)