On Wed, 3 May 2023 at 15:59, Amit Kapila <amit.kapil...@gmail.com> wrote:
>
> On Tue, May 2, 2023 at 4:52 PM Drouvot, Bertrand
> <bertranddrouvot...@gmail.com> wrote:
> >
> >
> > As per your suggestion, changing the insert ordering (like in V8 attached)
> > makes it now work on the failing environment too.
> >
>
> I think it is better to use wait_for_replay_catchup() to wait for
> standby to catch up. I have changed that and a comment in the
> attached. I'll push this tomorrow unless there are further comments.
Thanks for posting the updated patch, I had run this test in a loop of
100 times to verify that there was no failure because of race
conditions. The 100 times execution passed successfully.
One suggestion:
"wal file" should be changed to "WAL file":
+# Request a checkpoint on the standby to trigger the WAL file(s) removal
+$node_standby->safe_psql('postgres', 'checkpoint;');
+
+# Verify that the wal file has not been retained on the standby
+my $standby_walfile = $node_standby->data_dir . '/pg_wal/' . $walfile_name;
Regards,
Vignesh
