Date: Thu, 30 Jul 2015 15:59:39 -0300
From: Alvaro Herrera <alvherre@2ndquadrant.com>
To: John Scalia <jayknowsunix@gmail.com>
Cc: "pgsql-admin@postgresql.org" <pgsql-admin@postgresql.org>
Subject: Re: Autovacuum behavior
Message-ID: <20150730185939.GT2441@postgresql.org>
References: 
 <CABzCKRDtMfmgKFoMeTFDRGif+siQK9bXM4791ii8ZBJfORSFNA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: 
 <CABzCKRDtMfmgKFoMeTFDRGif+siQK9bXM4791ii8ZBJfORSFNA@mail.gmail.com>
User-Agent: Mutt/1.5.23 (2014-03-12)
Precedence: bulk
Sender: pgsql-admin-owner@postgresql.org

John Scalia wrote:
> Hi all,
> 
> The autovacuum settings for a 9.4.2 database are shown below, I'm not
> absolutely certain if I missed anything:
> 
> autovacuum = on
> log_autovacuum_min_duration = 100
> autovacuum_max_workers = 15
> autovacuum_naptime = 10min
> #autovacuum_vacuum_threshold = 50
> autovacuum_analyze_threshold = 80
> autovacuum_vacuum_scale_factor = 0.1
> autovacuum_analyze_scale_factor = 0.2
> autovacuum_freeze_max_age = 100000000
> autovacuum_vacuum_cost_delay = 20ms
> autovacuum_vacuum_cost_limit = -1

What's the vacuum_cost_limit setting?  Maybe they are sleeping for too
long, and don't have time to get to some of the other tables because all
15 workers are busy.  This gets worse the more workers there are.

I don't think the 10min naptime is doing you any favors, is it?

If you want them to go faster, maybe you need to lower the cost_delay.

-- 
Álvaro Herrera                http://www.2ndQuadrant.com/
PostgreSQL Development, 24x7 Support, Remote DBA, Training & Services


-- 
Sent via pgsql-admin mailing list (pgsql-admin@postgresql.org)
To make changes to your subscription:
http://www.postgresql.org/mailpref/pgsql-admin