Re: log_autovacuum in Postgres 14 -- ordering issue - Mailing list pgsql-hackers

From Michael Paquier
Subject Re: log_autovacuum in Postgres 14 -- ordering issue
Date
Msg-id YSg30TlPHDPwQC4V@paquier.xyz
Whole thread Raw
In response to Re: log_autovacuum in Postgres 14 -- ordering issue  (Alvaro Herrera <alvherre@alvh.no-ip.org>)
List pgsql-hackers
On Wed, Aug 25, 2021 at 08:23:04PM -0400, Alvaro Herrera wrote:
> On 2021-Aug-25, Peter Geoghegan wrote:
>>  This seems like a release housekeeping task to me. I'll come up with
>> a patch targeting 14 and master in a few days.
>
> Agreed, thanks.

Sorry for the late reply here.  Indeed, I can see your point to move
the buffer usage a bit down, grouped with the other information
related to I/O.  Moving down this information gives the attached.  If
you wish to do that yourself, that's fine by me, of course.

Saying this, an ANALYZE-only command does amvacuumcleanup() for all
the indexes and the stats exist.  I am not saying that we should do
that for 14 as that's too late, but we could consider adding the index
information also in this case in 15~?
--
Michael

Attachment

pgsql-hackers by date:

Previous
From: Mark Dilger
Date:
Subject: Re: amcheck/verify_heapam doesn't check for interrupts
Next
From: Julien Rouhaud
Date:
Subject: Re: Mark all GUC variable as PGDLLIMPORT