Failure of subscription tests with topminnow

From: Michael Paquier <michael(at)paquier(dot)xyz>
To: Postgres hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Failure of subscription tests with topminnow
Date: 2021-08-16 03:52:55
Message-ID: YRnhFxa9bo73wfpV@paquier.xyz
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi all,

topminnow has just failed in a weird fashion:
https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=topminnow&dt=2021-08-15%2013%3A24%3A58
# SELECT pid != FROM pg_stat_replication WHERE application_name = 'tap_sub';
# expecting this output:
# t
# last actual query output:
#
# with stderr:
# ERROR: syntax error at or near "FROM"
# LINE 1: SELECT pid != FROM pg_stat_replication WHERE application_na...

Looking at the logs, it seems like the problem boils down to an active
slot when using ALTER SUBSCRIPTION tap_sub CONNECTION:
2021-08-15 18:44:38.027 CEST [16473:2] ERROR: could not start WAL
streaming: ERROR: replication slot "tap_sub" is active for PID 16336

There is only one place in 001_rep_changes.pl where this is used.

Thoughts?
--
Michael

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Bossart, Nathan 2021-08-16 04:02:16 Re: make MaxBackends available in _PG_init
Previous Message Michael Paquier 2021-08-16 03:16:15 Re: Changes to recovery_min_apply_delay are ignored while waiting for delay