Date: Thu, 26 Oct 2006 16:59:36 -0500
From: "Jim C. Nasby" <jim@nasby.net>
To: Gavin Hamill <gdh@laterooms.com>
Cc: pgsql-performance@postgresql.org
Subject: Re: VACUUMs take twice as long across all nodes
Message-ID: <20061026215936.GO26892@nasby.net>
References: <20061026095827.9e85e1c0.gdh@laterooms.com>
	<20196.1161874041@sss.pgh.pa.us>
	<20061026160609.5e94e581.gdh@laterooms.com>
	<20061026191728.GD26892@nasby.net>
	<20061026213556.174bd7b5.gdh@laterooms.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <20061026213556.174bd7b5.gdh@laterooms.com>
User-Agent: Mutt/1.5.13 (2006-08-11)

On Thu, Oct 26, 2006 at 09:35:56PM +0100, Gavin Hamill wrote:
> On Thu, 26 Oct 2006 14:17:29 -0500
> "Jim C. Nasby" <jim@nasby.net> wrote:
> 
> > Are you sure that there's nothing else happening on the machine that
> > could affect the vacuum times? Like, say a backup? Or perhaps updates
> > coming in from Slony that didn't used to be there?
> 
> I'm absolutely certain. The backups run from only one slave, given that
> it is a full copy of node 1. Our overnight traffic has not increased
> any, and the nightly backups show that the overall size of the DB has
> not increased more than usual growth.
> 
> Plus, I have fairly verbose logging, and it's not showing anything out
> of the ordinary. 
> 
> Like I said, it's one of those awful hypothesis/hand-waving problems :)

Well, the fact that it's happening on all your nodes leads me to think
Slony is somehow involved. Perhaps it suddenly decided to change how
often it's issuing syncs? I know it issues vacuums as well, so maybe
that's got something to do with it... (though I'm guessing you've
already looked in pg_stat_activity/logs to see if anything
correlates...) Still, it might be worth asking about this on the slony
list...
-- 
Jim Nasby                                            jim@nasby.net
EnterpriseDB      http://enterprisedb.com      512.569.9461 (cell)