Re: A test for replay of regression tests - Mailing list pgsql-hackers

From Thomas Munro
Subject Re: A test for replay of regression tests
Date
Msg-id CA+hUKGJm_uex8rDfTcP6ig_nNTySD9DOKq9iyghTtuE0dg+t7g@mail.gmail.com
Whole thread Raw
In response to Re: A test for replay of regression tests  (Peter Geoghegan <pg@bowt.ie>)
Responses Re: A test for replay of regression tests  (Peter Geoghegan <pg@bowt.ie>)
List pgsql-hackers
On Fri, Mar 25, 2022 at 4:03 PM Peter Geoghegan <pg@bowt.ie> wrote:
> If you want to know whether or not the buildfarm will have problems
> due to VACUUM failing to get a cleanup lock randomly, then I suggest
> that you use an approach like the one from my patch here:
>
> https://postgr.es/m/CAH2-WzkiB-qcsBmWrpzP0nxvrQExoUts1d7TYShg_DrkOHeg4Q@mail.gmail.com
>
> I recently tried it again myself. With the gizmo in place the tests
> fail in exactly the same way you've had problems with on the
> buildfarm. On the first try, even.

Interesting.  IIUC your chaos gizmo shows that particular vacuum test
still failing on master, but that wouldn't happen in real life because
since 383f2221 it's a temp table.  Your gizmo should probably detect
temp rels, as your comment says.  I was sort of thinking that perhaps
if DISABLE_PAGE_SKIPPING is eventually made to do what its name sounds
like it does, we could remove TEMP from that test and it'd still pass
with the gizmo...



pgsql-hackers by date:

Previous
From: Peter Geoghegan
Date:
Subject: Re: Assert in pageinspect with NULL pages
Next
From: Amit Kapila
Date:
Subject: Re: logical decoding and replication of sequences