From: | MichaelDBA <MichaelDBA(at)sqlexec(dot)com> |
---|---|
To: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Cc: | Luís Roberto Weck <luisroberto(at)siscobra(dot)com(dot)br>, Michael Lewis <mlewis(at)entrata(dot)com>, pgsql-performance(at)lists(dot)postgresql(dot)org |
Subject: | Re: Slow query on a one-tuple table |
Date: | 2019-09-20 13:09:35 |
Message-ID: | 2d20142b-16b4-32e9-cf63-81d813d4718c@sqlexec.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-performance |
Hi all,
I sometimes set autovacuum_vacuum_scale factor = 0 but only when I also
set autovacuum_vacuum_threshold to some non-zero number to force vacuums
after a certain number of rows are updated. It takes the math out of it
by setting the threshold explicitly.
But in this case he has also set autovacuum_vacuum_threshold to only
25! So I think you have to fix your settings by increasing one or both
accordingly.
Regards,
Michael Vitale
Tom Lane wrote on 9/19/2019 6:57 PM:
> =?UTF-8?Q?Lu=c3=ads_Roberto_Weck?= <luisroberto(at)siscobra(dot)com(dot)br> writes:
>> As fas as autovacuum options, this is what I'm using:
>> autovacuum_vacuum_scale_factor=0,
> Ugh ... maybe I'm misremembering, but I *think* that has the effect
> of disabling autovac completely. You don't want zero.
>
> Check in pg_stat_all_tables.last_autovacuum to see if anything
> is happening. If the dates seem reasonably current, then I'm wrong.
>
> regards, tom lane
>
>
From | Date | Subject | |
---|---|---|---|
Next Message | Mariel Cherkassky | 2019-09-23 10:59:40 | pg12 partitions question |
Previous Message | Tom Lane | 2019-09-19 22:57:11 | Re: Slow query on a one-tuple table |