Re: Orphaned records in pg_replication_origin_status after subscription drop

From: Michael Paquier <michael(at)paquier(dot)xyz>
To: Masahiko Sawada <sawada(dot)mshk(at)gmail(dot)com>
Cc: "Zhijie Hou (Fujitsu)" <houzj(dot)fnst(at)fujitsu(dot)com>, Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>, Postgres hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Orphaned records in pg_replication_origin_status after subscription drop
Date: 2025-12-22 23:21:39
Message-ID: aUnSg50Y7tsASWvt@paquier.xyz
Views: Whole Thread | Raw Message | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Mon, Dec 22, 2025 at 01:22:38PM -0800, Masahiko Sawada wrote:
> Thank you for making the patch! The patch looks good to me.

Creating the origin at the end of the same transaction that sets the
state to SUBREL_STATE_DATASYNC seems sensible here. I'll spend a
couple of extra hours playing with all that across all the branches,
see if I can wrap it. This includes some more error injection to
cross-check the state of all these transactions with the states in
the catalogs while we drop the subscription.

The test addition is interesting, nice. I didn't notice that it would
be possible to use this trick in 004_sync..

Also, I have double-checked the names of the folks who have reported
the bug, giving the following list for the commit logs (in case I
don't finish this stuff, feel free to use that):
Daisuke Higuchi <higudai(at)amazon(dot)com>
Tenglong Gu <brucegu(at)amazon(dot)com>
--
Michael

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Masahiko Sawada 2025-12-22 23:24:40 Re: [PATCH] Add memory usage reporting to VACUUM VERBOSE
Previous Message Thomas Munro 2025-12-22 23:02:03 Re: [PATCH] O_CLOEXEC not honored on Windows - handle inheritance chain