mirrors/postgres
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity

Add retries for further investigation of 019_replslot_limit.pl failures.

Browse Source 
Tom noticed evidence in the buildfarm suggesting the failures might just be
really slow process exits. To investigate further, instead of giving up after
seeing multiple walsender pids once, retry. For now continue to report test
failure if a retry succeeds.

See also commit afdeff10526 and fe0972ee5e6.

Per suggestion from Tom Lane.

Discussion: https://postgr.es/m/3042597.1648148740@sss.pgh.pa.us
This commit is contained in:
Andres Freund 2022-03-24 17:12:09 -07:00
parent 26ebb0e280
commit f28bf667f6
1 changed files with 28 additions and 10 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

38
src/test/recovery/t/019_replslot_limit.pl
View File

@ -333,23 +333,41 @@ $node_standby3->init_from_backup($node_primary3, $backup_name,
$node_standby3->append_conf('postgresql.conf', "primary_slot_name = 'rep3'"); $node_standby3->append_conf('postgresql.conf', "primary_slot_name = 'rep3'");
$node_standby3->start; $node_standby3->start;
$node_primary3->wait_for_catchup($node_standby3); $node_primary3->wait_for_catchup($node_standby3);
my $senderpid = $node_primary3->safe_psql('postgres',
"SELECT pid FROM pg_stat_activity WHERE backend_type = 'walsender'");
# We've seen occasional cases where multiple walsender pids are active. An my $senderpid;
# immediate shutdown may hide evidence of a locking bug. So if multiple
# walsenders are observed, shut down in fast mode, and collect some more # We've seen occasional cases where multiple walsender pids are active. It
# information. # could be that we're just observing process shutdown being slow. To collect
if (not like($senderpid, qr/^[0-9]+$/, "have walsender pid $senderpid")) # more information, retry a couple times, print a bit of debugging information
# each iteration. For now report a test failure even if later iterations
# succeed.
my $i = 0;
while (1)
{ {
my ($stdout, $stderr); my ($stdout, $stderr);
$senderpid = $node_primary3->safe_psql('postgres',
"SELECT pid FROM pg_stat_activity WHERE backend_type = 'walsender'");
last if like($senderpid, qr/^[0-9]+$/, "have walsender pid $senderpid");
# show information about all active connections
$node_primary3->psql('postgres', $node_primary3->psql('postgres',
"\\a\\t\nSELECT * FROM pg_stat_activity", "\\a\\t\nSELECT * FROM pg_stat_activity",
stdout => \$stdout, stderr => \$stderr); stdout => \$stdout, stderr => \$stderr);
diag $stdout, $stderr; diag $stdout, $stderr;
$node_primary3->stop('fast');
$node_standby3->stop('fast'); # unlikely that the problem would resolve after 15s, so give up at point
die "could not determine walsender pid, can't continue"; if ($i++ == 150)
{
# An immediate shutdown may hide evidence of a locking bug. If
# retrying didn't resolve the issue, shut down in fast mode.
$node_primary3->stop('fast');
$node_standby3->stop('fast');
die "could not determine walsender pid, can't continue";
}
usleep(100_000);
} }
my $receiverpid = $node_standby3->safe_psql('postgres', my $receiverpid = $node_standby3->safe_psql('postgres',