Re: Issues with ON CONFLICT UPDATE and REINDEX CONCURRENTLY

From: Mihail Nikalayeu <mihailnikalayeu(at)gmail(dot)com>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: Álvaro Herrera <alvherre(at)kurilemu(dot)de>, Alexander Lakhin <exclusion(at)gmail(dot)com>, Michael Paquier <michael(at)paquier(dot)xyz>, Noah Misch <noah(at)leadboat(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)postgresql(dot)org>
Subject: Re: Issues with ON CONFLICT UPDATE and REINDEX CONCURRENTLY
Date: 2025-12-03 10:22:13
Message-ID: CADzfLwWuZRXP8aPrXEvre-MD3cMeSPwgSCbCwH2wAe02dPeABQ@mail.gmail.com
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

Hi, Tom!

On Wed, Dec 3, 2025 at 5:43 AM Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> wrote:

> BF member prion has been failing since 5dee7a603 went in. Apparently
> that solution doesn't work under -DRELCACHE_FORCE_RELEASE and/or
> -DCATCACHE_FORCE_RELEASE (so I'm expecting the CLOBBER_CACHE_ALWAYS
> animals to fail too, whenever they next report).

Oh, my bad, sorry.
Attached patch with output variant for
RELCACHE_FORCE_RELEASE\CATCACHE_FORCE_RELEASE case.

But for CLOBBER_CACHE_ALWAYS - it is just a mess, too many different
kinds of wakeup loops need to be done.

So, my proposals:
* either drop that test - currently it looks like a test that will
fail 99% of time due to relcache changes instead of actual issue
* either add some kind of mechanics to skip that test in case of
particular build arguments
* either replace it by some kind of small stress test, without any
injection_points

Best regards,
Mikhail.

Attachment Content-Type Size
Test_output_variant_for_CATCACHE_FORCE_RELEASE.patch text/x-patch 8.1 KB

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Ignat Remizov 2025-12-03 10:37:30 [PATCH] Add enable_copy_program GUC to control COPY PROGRAM
Previous Message Heikki Linnakangas 2025-12-03 09:57:32 Re: Serverside SNI support in libpq