Re: Berserk Autovacuum (let's save next Mandrill)

From: Darafei "Komяpa" Praliaskouski <me(at)komzpa(dot)net>
To: Michael Banck <mbanck(at)gmx(dot)net>
Cc: David Rowley <david(dot)rowley(at)2ndquadrant(dot)com>, Alvaro Herrera <alvherre(at)2ndquadrant(dot)com>, PostgreSQL Developers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Berserk Autovacuum (let's save next Mandrill)
Date: 2019-03-31 10:19:53
Message-ID: CAC8Q8tJVO+2hZMrLyv=AaHKA45S8dcWC7kARUX6xU2qJSeoLqQ@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

>
> By the way, the Routine Vacuuming chapter of the documentation says:
>
> "The sole disadvantage of increasing autovacuum_freeze_max_age (and
> vacuum_freeze_table_age along with it) is that the pg_xact and
> pg_commit_ts subdirectories of the database cluster will take more space
>
> [...]
>
> If [pg_xact and pg_commit_ts taking 0.5 and 20 GB, respectively]
> is trivial compared to your total database size, setting
> autovacuum_freeze_max_age to its maximum allowed value is recommended."
>
> Maybe this should be qualified with "unless you have trouble with your
> autovacuum keeping up" or so; or generally reworded?

This recommendation is in the mindset of "wraparound never happens".
If your database is large, you have more chances to hit it painfully, and
if it's append-only even more so.

Alternative point of "if your database is super large and actively written,
you may want to set autovacuum_freeze_max_age to even smaller values so
that autovacuum load is more evenly spread over time" may be needed.

--
Darafei Praliaskouski
Support me: http://patreon.com/komzpa

In response to

Responses

Browse pgsql-hackers by date

  From Date Subject
Next Message Darafei Komяpa Praliaskouski 2019-03-31 10:30:12 Re: Berserk Autovacuum (let's save next Mandrill)
Previous Message Darafei Komяpa Praliaskouski 2019-03-31 10:12:21 Re: Berserk Autovacuum (let's save next Mandrill)