From: Scott Marlowe <scott.marlowe@gmail.com>
To: Sachchida Ojha <sojha@secure-elements.com>
Cc: Michael Glaesemann <grzm@seespotcode.net>
Cc: pgsql-performance@postgresql.org
Subject: Re: Autovacuum is running forever
Date: Tue, 21 Aug 2007 17:01:23 -0500
Message-ID: <dcc563d10708211501p559b73d3q123a35bdbce1b68d@mail.gmail.com> (raw)
In-Reply-To: <E099DB9EF394534CBA0F62ED03CF3BBF05FDBF1F@EXVS01.IceMAIL.com>
References: <C927A262-C7B0-4003-ABAE-570256F76B16@seespotcode.net>
<E099DB9EF394534CBA0F62ED03CF3BBF05FDBF1F@EXVS01.IceMAIL.com>
On 8/21/07, Sachchida Ojha <sojha@secure-elements.com> wrote:
> vacuum_cost_delay = 200
> vacuum_cost_page_hit = 1
> vacuum_cost_page_miss = 10
> vacuum_cost_page_dirty = 20
> vacuum_cost_limit = 200
>
> autovacuum = on
> autovacuum_naptime = 3600
>
> maintenance_work_mem = 256MB # min 1024, size
That's a REALLY long naptime. Better to let autovacuum decide if you
need vacuum more often, and just increase the vacuum_cost_delay and
decrease vacuum_cost_limit so that vacuum doesn't slam your I/O.
Maintenance work mem on the other hand is plenty big. and your fsm
settings seem large enough to handle your freed space.
But making vacuum wait so long between runs may be slowly bloating
your data store, and then vacuum becomes more and more expensive
because it takes longer and longer to run.
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Reply to all the recipients using the --to and --cc options:
reply via email
To: pgsql-performance@postgresql.org
Cc: scott.marlowe@gmail.com, sojha@secure-elements.com, grzm@seespotcode.net
Subject: Re: Autovacuum is running forever
In-Reply-To: <dcc563d10708211501p559b73d3q123a35bdbce1b68d@mail.gmail.com>
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
This inbox is served by DDX for PostgreSQL; see mirroring instructions
for how to clone and mirror all data and code used for this inbox