X-Original-To: pgsql-performance-postgresql.org@localhost.postgresql.org Received: from localhost (unknown [200.46.204.144]) by svr1.postgresql.org (Postfix) with ESMTP id 0E5E352A3A for ; Tue, 19 Jul 2005 13:50:28 -0300 (ADT) Received: from svr1.postgresql.org ([200.46.204.71]) by localhost (av.hub.org [200.46.204.144]) (amavisd-new, port 10024) with ESMTP id 74720-04 for ; Tue, 19 Jul 2005 16:50:26 +0000 (GMT) Received: from sherman.stortek.com (sherman.stortek.com [129.80.22.146]) by svr1.postgresql.org (Postfix) with ESMTP id D189752A1B for ; Tue, 19 Jul 2005 13:50:18 -0300 (ADT) Received: from sherman.stortek.com (localhost [127.0.0.1]) by sherman.stortek.com (8.12.10/8.12.0) with ESMTP id j6JGoHD4003067 for ; Tue, 19 Jul 2005 10:50:17 -0600 (MDT) Received: from unix.stortek.com (burma.stortek.com [129.80.16.110]) by sherman.stortek.com (8.12.10/8.12.0) with ESMTP id j6JGoGDC003064; Tue, 19 Jul 2005 10:50:16 -0600 (MDT) Received: from C118181.stortek.com (c118181.stortek.com [129.80.81.172]) by unix.stortek.com (8.12.10/8.12.0) with ESMTP id j6JGoFvn025709; Tue, 19 Jul 2005 10:50:16 -0600 (MDT) Date: Tue, 19 Jul 2005 10:50:14 -0600 From: Robert Creager To: Tom Lane Cc: pgsql-performance@postgresql.org Subject: Re: Huge performance problem between 7.4.1 and 8.0.3 - CS issue? Message-ID: <20050719105014.00007635@C118181.stortek.com> In-Reply-To: <15472.1121709173@sss.pgh.pa.us> References: <20050718114952.00004ac5@C118181.stortek.com> <15472.1121709173@sss.pgh.pa.us> Organization: StorageTek X-Mailer: Sylpheed-Claws 1.0.4.1 Win32 (GTK+ 1.3.0; Win32) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=0 tagged_above=0 required=5 tests= X-Spam-Level: X-Archive-Number: 200507/290 X-Sequence-Number: 13531 On Mon, 18 Jul 2005 13:52:53 -0400 Tom Lane wrote: > > Start a fresh psql session and "SHOW vacuum_cost_delay" to verify what > the active setting is. > Alright. Restarted the 803 database. Cron based vacuum analyze is running every 5 minutes. vacuum_cost_delay is 0. The problem showed up after about 1/2 hour of running. I've got vacuum jobs stacked from the last 35 minutes, with 2 vacuums running at the same time. CS is around 73k. What do I do now? I can bring the db back to normal and not run any cron based vacuum to see if it still happens, but I suspect nothing will happen without the vacuum. I'll leave it in it's current semi-catatonic state as long as possible in case there is something to look at? Cheers, Rob