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 7243552BD4 for ; Wed, 13 Jul 2005 17:18:19 -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 54607-01 for ; Wed, 13 Jul 2005 20:18:18 +0000 (GMT) Received: from mr2.surnet.cl (smtp2.surnet.cl [216.155.73.163]) by svr1.postgresql.org (Postfix) with ESMTP id 545F952B5B for ; Wed, 13 Jul 2005 17:18:16 -0300 (ADT) Received: from smtp2.surnet.cl (216.155.73.169) by mr2.surnet.cl (7.0.031.3) id 4259ADFA015441EB; Wed, 13 Jul 2005 16:17:57 -0400 Received: from smtp2.surnet.cl (mr2.surnet.cl []) by mr2.surnet.cl ([216.155.73.169]); Wed, 13 Jul 2005 20:17:57 +0000 Received: from cluster.surnet.cl (216.155.73.164) by smtp2.surnet.cl (7.0.031.3) id 4259ADE300CFADA6; Wed, 13 Jul 2005 16:17:57 -0400 Received: from alvh.no-ip.org (200.85.218.66) by cluster.surnet.cl (7.0.043) (authenticated as alvherre@surnet.cl) id 42B3EF6000399CCB; Wed, 13 Jul 2005 16:18:13 -0400 Received: by alvh.no-ip.org (Postfix, from userid 1000) id 925F2C370EE; Wed, 13 Jul 2005 16:18:17 -0400 (CLT) Date: Wed, 13 Jul 2005 16:18:17 -0400 From: Alvaro Herrera To: Dan Harris Cc: pgsql-performance@postgresql.org Subject: Re: Quad Opteron stuck in the mud Message-ID: <20050713201817.GA5450@alvh.no-ip.org> References: <42D56772.6020408@arbash-meinel.com> <862C6AD9-646B-4907-AB18-09F1AD08B5F5@drivefaster.net> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <862C6AD9-646B-4907-AB18-09F1AD08B5F5@drivefaster.net> User-Agent: Mutt/1.5.9i X-Virus-Scanned: by amavisd-new at hub.org X-Spam-Status: No, hits=1.378 tagged_above=0 required=5 tests=AWL, DNS_FROM_RFC_ABUSE, FORGED_RCVD_HELO, RCVD_IN_NJABL_PROXY X-Spam-Level: * X-Archive-Number: 200507/151 X-Sequence-Number: 13392 On Wed, Jul 13, 2005 at 01:16:25PM -0600, Dan Harris wrote: > On Jul 13, 2005, at 1:11 PM, John A Meinel wrote: > > >I might be wrong, but there may be something much more substantially > >wrong than slow i/o. > > Yes, I'm afraid of that too. I just don't know what tools I should > use to figure that out. I have some 20 other databases on this > system, same schema but varying sizes, and the small ones perform > very well. It feels like there is an O(n) increase in wait time that > has recently become very noticeable on the largest of them. I'd guess it's stuck on some lock. Try that EXPLAIN, and when it blocks, watch the pg_locks view for locks not granted to the process executing the EXPLAIN. Then check what else is holding the locks. -- Alvaro Herrera () "La rebeldía es la virtud original del hombre" (Arthur Schopenhauer)