On 2024-05-16 Th 16:18, Tom Lane wrote:
> Thomas Munro <thomas.munro@gmail.com> writes:
>> For citext_utf8, I pushed cff4e5a3. Hamerkop runs infrequently, so
>> here's hoping for 100% green on master by Tuesday or so.
> Meanwhile, back at the ranch, it doesn't seem that changed anything:
>
> https://buildfarm.postgresql.org/cgi-bin/show_log.pl?nm=hamerkop&dt=2024-05-16%2011%3A00%3A32
>
> ... and now that I look more closely, the reason why it didn't
> change anything is that hamerkop is still building 0294df2
> on HEAD. All its other branches are equally stuck at the
> end of March. So this is a flat-out-broken animal, and I
> plan to just ignore it until its owner un-sticks it.
> (In particular, I think we shouldn't be in a hurry to push
> the patch discussed downthread.)
>
> Andrew: maybe the buildfarm server could be made to flag
> animals building exceedingly old commits? This is the second
> problem of this sort that I've noticed this month, and you
> really have to look closely to realize it's happening.
>
>
Yeah, that should be doable. Since we have the git ref these days we
should be able to mark it as old, or maybe just reject builds for very
old commits (the latter would be easier).
cheers
andrew
--
Andrew Dunstan
EDB: https://www.enterprisedb.com