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 E085ED1B52C for ; Tue, 12 Aug 2003 04:37:30 +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 18660-02 for ; Tue, 12 Aug 2003 01:35:58 -0300 (ADT) Received: from home.samurai.com (home.samurai.com [205.207.28.66]) by svr1.postgresql.org (Postfix) with ESMTP id 1DE93D1B550 for ; Tue, 12 Aug 2003 01:37:20 -0300 (ADT) Received: by home.samurai.com (Postfix, from userid 1014) id 8866A26203; Tue, 12 Aug 2003 00:37:19 -0400 (EDT) Date: Tue, 12 Aug 2003 00:37:19 -0400 From: Neil Conway To: Bruce Momjian Cc: Reece Hart , "pgsql-performance@postgresql.org" Subject: Re: Perfomance Tuning Message-ID: <20030812043719.GB76772@home.samurai.com> References: <1060642567.15483.220.camel@tallac> <200308112259.h7BMxUo14439@candle.pha.pa.us> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <200308112259.h7BMxUo14439@candle.pha.pa.us> User-Agent: Mutt/1.4.1i X-Virus-Scanned: by amavisd-new at postgresql.org X-Archive-Number: 200308/195 X-Sequence-Number: 3012 On Mon, Aug 11, 2003 at 06:59:30PM -0400, Bruce Momjian wrote: > Uh, the ext2 developers say it isn't 100% reliable --- at least that is > that was told. I don't know any personally, but I mentioned it while I > was visiting Red Hat, and they didn't refute it. IMHO, if we're going to say "don't use X on production PostgreSQL systems", we need to have some better evidene than "no one has said anything to the contrary, and I heard X is bad". If we can't produce such evidence, we shouldn't say anything at all, and users can decide what to use for themselves. (Not that I'm agreeing or disagreeing about ext2 in particular...) > > My > > untested interpretation was that the update bookkeeping as well as data > > update were all getting journalled, the journal space would fill, get > > sync'd, then repeat. In effect, all blocks were being written TWICE just > > for the journalling, never mind the overhead for PostgreSQL > > transactions. Journalling may or may not have been the culprit, but I doubt everything was being written to disk twice: (a) ext3 does metadata-only journalling by default (b) PostgreSQL only fsyncs WAL records to disk, not the data itself -Neil