[prev in list] [next in list] [prev in thread] [next in thread] 

List:       postgresql-general
Subject:    Re: Automatic autovacuum to prevent wraparound - PG13.5
From:       Peter Geoghegan <pg () bowt ! ie>
Date:       2022-06-24 17:11:11
Message-ID: CAH2-Wzn4WXBLM8M5p7Sf3noub8fVoRL-ZCRrPp1=-X9Tt1cs+A () mail ! gmail ! com
[Download RAW message or body]

On Wed, Jun 15, 2022 at 4:13 AM Mauro Farracha <farracha@gmail.com> wrote:
> The scenario:
> - Out of nowhere (during the weekend), without database activity load or batches \
> running, with previous nightly run of vacuum freeze, in the middle of the day, with \
> xids and mxids below 20M we are seeing autovacuum being triggered to prevent \
> wraparound. 
> My question is why this is occurring, which condition might be responsible for this \
> behaviour?

There is a behavior that seems like it might be relevant: VACUUM
interprets autovacuum_multixact_freeze_max_age in a way that accounts
for both MultiXactId consumption and the consumption of "member space"
by MultiXacts. Technically there are 2 SLRUs for MultiXacts, either of
which can wraparound.

This behavior was established by commit 53bb309d2d. It is documented.
Admittedly this whole area of the documentation is in dire need of an
overhaul.  :-(

-- 
Peter Geoghegan


[prev in list] [next in list] [prev in thread] [next in thread]