Horton

Re: Testing autovacuum wraparound (including failsafe)

Daniel Gustafsson <daniel@yesql.se>

View thread View on postgr.es
From: Daniel Gustafsson <daniel@yesql.se>
To: Masahiko Sawada <sawada.mshk@gmail.com>
Cc: Noah Misch <noah@leadboat.com>, John Naylor <john.naylor@enterprisedb.com>, Heikki Linnakangas <hlinnaka@iki.fi>, Ian Lawrence Barwick <barwick@gmail.com>, Andres Freund <andres@anarazel.de>, Anastasia Lubennikova <lubennikovaav@gmail.com>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>, Peter Geoghegan <pg@bowt.ie>
Date: 2023-11-27T13:40:09Z
Lists: pgsql-hackers 
> On 27 Nov 2023, at 14:06, Masahiko Sawada <sawada.mshk@gmail.com> wrote:

> Is it true that we can modify the timeout after creating
> BackgroundPsql object? If so, it seems we don't need to introduce the
> new timeout argument. But how?

I can't remember if that's leftovers that incorrectly remains from an earlier
version of the BackgroundPsql work, or if it's a very bad explanation of
->set_query_timer_restart().  The timeout will use the timeout_default value
and that cannot be overridden, it can only be reset per query.

With your patch the timeout still cannot be changed, but at least set during
start which seems good enough until there are tests warranting more complexity.
The docs should be corrected to reflect this in your patch.

--
Daniel Gustafsson

Commits

  1. Fix meson installation of xid_wraparound test.

  2. Add tests for XID wraparound.

  3. Add option to specify timeout seconds to BackgroundPsql.pm.

  4. Fix wrong description of BackgroundPsql's timeout.

  5. Refactor background psql TAP functions

  6. Consider triggering VACUUM failsafe during scan.

  7. Fix autovacuum log output heap truncation issue.