Re: Test 026_overwrite_contrecord fails on very slow machines (under Valgrind) - Mailing list pgsql-hackers

From Michael Paquier
Subject Re: Test 026_overwrite_contrecord fails on very slow machines (under Valgrind)
Date
Msg-id ZS96oP3mViEk1eqx@paquier.xyz
Whole thread Raw
In response to Re: Test 026_overwrite_contrecord fails on very slow machines (under Valgrind)  (Andres Freund <andres@anarazel.de>)
List pgsql-hackers
On Thu, Oct 12, 2023 at 04:46:02PM -0700, Andres Freund wrote:
> The machine skink is hosted on runs numerous buildfarm animals (24 I think
> right now, about to be 28). While it has plenty resources (16 cores/32
> threads, 128GB RAM), test runtime is still pretty variable depending on what
> other tests are running at the same time...

Okay.  It seems to me that just setting checkpoint_timeout to 1h
should leave plenty of room to make sure that the test is not unstable
compared to the default of 5 mins.  So, why not just do that and see
what happens for a few days?
--
Michael

Attachment

pgsql-hackers by date:

Previous
From: Peter Eisentraut
Date:
Subject: Re: Clean up some pg_dump tests
Next
From: Michael Paquier
Date:
Subject: Re: Remove wal_level settings for subscribers in tap tests