Horton

Re: Fixes inconsistent behavior in vacuum when it processes multiple relations

Nathan Bossart <nathandbossart@gmail.com>

View thread
From: Nathan Bossart <nathandbossart@gmail.com>
To: Michael Paquier <michael@paquier.xyz>
Cc: shihao zhong <zhong950419@gmail.com>, PostgreSQL-development <pgsql-hackers@postgresql.org>
Date: 2025-06-23T15:38:40Z
Lists: pgsql-hackers

Commits

Same data as JSON: GET /api/v1/messages/:b64id/commits the thread's linked commits as JSON, with link sources. API reference →

  1. Rationalize handling of VacuumParams

  2. Avoid scribbling of VACUUM options

On Mon, Jun 23, 2025 at 08:48:35AM +0900, Michael Paquier wrote:
> Anyway, here is an attempt at leaning all that.  I am really tempted
> to add a couple of const markers to force VacuumParams to be in
> read-only mode, even if it means doing so for vacuum() but not touch
> at vacuum_rel() where we double-check the reloptions for the truncate
> and index cleanup options.  That would be of course v19-only material.
> Thoughts or opinions?

Is the idea to do something like this for v19 and this [0] for the
back-branches?

I think the recurse-to-TOAST-table case is still broken with your patch.
We should probably move the memcpy() for toast_vacuum_params to the very
top of vacuum_rel().  Otherwise, your patch looks generally reasonable to
me.

[0] https://postgr.es/m/aFRzYhOTZcRgKPLu%40nathan

-- 
nathan