Hi,
On 2024-03-26 00:00:38 -0400, Tom Lane wrote:
> Andres Freund <andres@anarazel.de> writes:
> > I think there must be some actual regression involved. The frequency of
> > failures on HEAD vs failures on 16 - both of which run the tests concurrently
> > via meson - is just vastly different.
>
> Are you sure it's not just that the total time to run the core
> regression tests has grown to a bit more than what the test timeout
> allows for?
You're right, that could be it - in a way at least, the issue is replay not
catching up within 180s, so it'd have to be the data volume growing, I think.
But it doesn't look like the regression volume meaningfully grew around that
time?
I guess I'll try to write a buildfarm database query to extract how long that
phase of the test took from all runs on my menagerie, not just the failing
one, and see if there's a visible trend.
Greetings,
Andres Freund