Disabling vacuum truncate for autovacuum - Mailing list pgsql-general

From Will Storey
Subject Disabling vacuum truncate for autovacuum
Date
Msg-id Z2DE4lDX4tHqNGZt@dev.null
Whole thread Raw
Responses Re: Disabling vacuum truncate for autovacuum
List pgsql-general
Hi!

I would like to disable vacuum's truncate behaviour for autovacuum.
Previously I had an outage due to its access exclusive lock when it was
replicated to a hot standby.

When that outage happened it was from a VACUUM call in a cronjob rather
than autovacuum. I now run such VACUUMs with TRUNCATE false which avoids
the issue for these. However I've realized that autovacuum could cause this
as well. This is of interest to me because I'm looking at tuning autovacuum
and getting rid of the cronjob, but I've realized relying on autovacuum
could be dangerous because of the truncates.

I believe the only way to disable this for autovacuum is by changing the
vacuum_truncate storage parameters on tables. (Ignoring the now removed
old_snapshot_threshold option). I'm thinking of altering all my tables to
turn it off. Is this a horrible idea? I expect I would need to monitor
tables for problematic growth, but that might be better than a surprise
outage. I suppose the growth could cause an outage too, but I'm thinking it
would be more controllable.

Would I need to disable the settings on catalog tables too? (To rule out
any possibility of it happening). Are there any other things I might be
missing?

I am also wondering if having an autovacuum setting to control it would be
a good idea for a feature. That would be simpler for me than altering all
my tables and help me avoid missing any (e.g. catalogs, new tables).

I might be worrying needlessly about this as maybe it is unlikely to
happen. I suppose it is workload dependent.

Thank you!

Will



pgsql-general by date:

Previous
From: Bharani SV-forum
Date:
Subject: Re: Additional Help in vetting my steps for Postgres DB upgrade from Ver 13.X to ver 15.X
Next
From: Ivan Kurnosov
Date:
Subject: How to deal with dangling files after aborted `pg_restore`?