Received: from localhost (unknown [200.46.204.187]) by postgresql.org (Postfix) with ESMTP id 89D9C9FA60E for ; Mon, 15 Oct 2007 23:04:43 -0300 (ADT) Received: from postgresql.org ([200.46.204.71]) by localhost (mx1.hub.org [200.46.204.187]) (amavisd-maia, port 10024) with ESMTP id 85807-09 for ; Mon, 15 Oct 2007 23:04:31 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.5 Received: from sss.pgh.pa.us (sss.pgh.pa.us [66.207.139.130]) by postgresql.org (Postfix) with ESMTP id A27549FA610 for ; Mon, 15 Oct 2007 23:04:36 -0300 (ADT) Received: from sss2.sss.pgh.pa.us (tgl@localhost [127.0.0.1]) by sss.pgh.pa.us (8.14.1/8.14.1) with ESMTP id l9G24V7E000742; Mon, 15 Oct 2007 22:04:31 -0400 (EDT) To: Darren Reed cc: Scott Marlowe , "Joshua D. Drake" , pgsql-admin@postgresql.org Subject: Re: Is my database now too big? In-reply-to: <4714151F.4050307@fastmail.net> References: <1191730382.18880.1214536197@webmail.messagingengine.com> <47087723.7040403@commandprompt.com> <1191744007.14918.1214548241@webmail.messagingengine.com> <4708A9DC.5060306@fastmail.net> <47097EA7.902@fastmail.net> <4712D97E.7010004@fastmail.net> <4712E15B.8000405@fastmail.net> <17093.1192464190@sss.pgh.pa.us> <4713CA8C.5020303@fastmail.net> <23809.1192485371@sss.pgh.pa.us> <4714151F.4050307@fastmail.net> Comments: In-reply-to Darren Reed message dated "Mon, 15 Oct 2007 18:34:23 -0700" Date: Mon, 15 Oct 2007 22:04:31 -0400 Message-ID: <741.1192500271@sss.pgh.pa.us> From: Tom Lane X-Virus-Scanned: Maia Mailguard 1.0.1 X-Archive-Number: 200710/133 X-Sequence-Number: 27047 Darren Reed writes: > Tom Lane wrote: >> Indeed. And there shouldn't even be anything in MessageContext until >> the first client command has been received. Maybe you have something >> in ~/.psqlrc that you haven't told us about? > That's easy - I don't even have one of these files! Then the behavior you showed is impossible ;-) There is *something* that is completely broken about your machine, and the rest of us really don't have enough context to tell what. You haven't told us anything about the hardware or operating system, or how you built or obtained the Postgres executables. I don't think you should dismiss the possibility of a hardware problem, especially since the failures aren't 100% reproducible (AFAICT from your previous remarks). We've seen more than one case where Postgres stressed a system more than anything else that was being run, and thereby exposed hardware problems that didn't manifest otherwise. For instance, a bit of bad RAM up near the end of physical memory might not get used at all until Postgres starts eating up memory. Another line of thought is that you built Postgres with a buggy compiler and thereby got buggy executables. Have you tried running the PG regression tests? regards, tom lane