X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [64.117.224.130]) by svr1.postgresql.org (Postfix) with ESMTP id 4489FD1B8AE for ; Fri, 29 Aug 2003 00:50:28 +0000 (GMT) Received: from svr1.postgresql.org ([64.117.224.193]) by localhost (neptune.hub.org [64.117.224.130]) (amavisd-new, port 10024) with ESMTP id 28252-08 for ; Thu, 28 Aug 2003 21:50:00 -0300 (ADT) Received: from harrier.cohaesio.com (harrier.cohaesio.com [212.97.128.50]) by svr1.postgresql.org (Postfix) with ESMTP id 49349D1B52C for ; Thu, 28 Aug 2003 21:49:57 -0300 (ADT) Received: by harrier.cohaesio.com (Postfix, from userid 88) id 2D3D086; Fri, 29 Aug 2003 02:49:45 +0200 (CEST) X-Original-To: news2mail@news.cohaesio.com From: "Anders K. Pedersen" Subject: Re: Queries sometimes take 1000 times the normal time Date: Fri, 29 Aug 2003 02:49:44 +0200 Organization: Cohaesio A/S Lines: 14 Message-ID: References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit X-Trace: harrier.cohaesio.com 1062118184 13692 212.97.128.160 (29 Aug 2003 00:49:44 GMT) X-Complaints-To: newsmaster@news.cohaesio.com X-Accept-Language: da, en-us, en To: pgsql-performance@postgresql.org X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200308/451 X-Sequence-Number: 3267 scott.marlowe wrote: > Just to add to the clutch here, also check your bdflush settings (if > you're on linux) or equivalent (if you're not.) > > Many times the swapping algo in linux can be quite bursty if you have it > set to move too many pages at a time during cleanup / flush. According to vmstat it doesn't swap near the spikes, so I don't think this is the problem. I posted a vmstat sample in another reply, where you can see an example of what happens. Regards, Anders K. Pedersen