Re: effective_wal_level is not decreasing after using REPACK (CONCURRENTLY)

From: Álvaro Herrera <alvherre(at)kurilemu(dot)de>
To: Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>
Cc: shveta malik <shveta(dot)malik(at)gmail(dot)com>, Imran Zaheer <imran(dot)zhir(at)gmail(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org>, shveta malik <shvetamalik(at)gmail(dot)com>
Subject: Re: effective_wal_level is not decreasing after using REPACK (CONCURRENTLY)
Date: 2026-05-27 21:20:43
Message-ID: ahdfQfjw_l2OMH-e@alvherre.pgsql
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On 2026-May-27, Masahiko Sawada wrote:

> > Some buildfarm members reported failures[1][2]. The newly added test
> > works only on builds without injection points because the logical slot
> > "test_slot" is removed during tests using injection points, otherwise
> > it still exists on the primary server at the end of
> > 051_effective_wal_level.pl.
> >
>
> I've attached the patch to fix the issue by moving the new test
> earlier in the test script, so it no longer depends on slot state.

Ah, so that's why the submitted patch had the test placed at an awkward
place in the file! I moved it to the end because I failed to realize
this problem. I didn't see this failure locally, and CI didn't see it
either. I wonder if we're lacking coverage of the no-injection-points
case in CI.

Thanks for the patch, please feel free to push it.

--
Álvaro Herrera PostgreSQL Developer — https://www.EnterpriseDB.com/
"People get annoyed when you try to debug them." (Larry Wall)

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Amit Kapila 2026-05-27 21:26:47 Re: Proposal: Conflict log history table for Logical Replication
Previous Message Jelte Fennema-Nio 2026-05-27 20:33:30 Re: Heads Up: cirrus-ci is shutting down June 1st