Received: from localhost (unknown [200.46.204.183]) by postgresql.org (Postfix) with ESMTP id 72DC664FD40 for ; Fri, 29 Aug 2008 08:09:26 -0300 (ADT) Received: from postgresql.org ([200.46.204.86]) by localhost (mx1.hub.org [200.46.204.183]) (amavisd-maia, port 10024) with ESMTP id 69238-07 for ; Fri, 29 Aug 2008 08:09:18 -0300 (ADT) X-Greylist: from auto-whitelisted by SQLgrey-1.7.6 Received: from mx00.pub.collaborativefusion.com (mx00.pub.collaborativefusion.com [206.210.89.199]) by postgresql.org (Postfix) with ESMTP id 8E9A564FD20 for ; Fri, 29 Aug 2008 08:09:22 -0300 (ADT) Received: from vanquish.ws.pitbpa0.priv.collaborativefusion.com (vanquish.ws.pitbpa0.priv.collaborativefusion.com [192.168.2.162]) (SSL: TLSv1/SSLv3,256bits,AES256-SHA) by wingspan with esmtp; Fri, 29 Aug 2008 07:09:21 -0400 id 00056403.0000000048B7D8E1.00001859 Date: Fri, 29 Aug 2008 07:08:52 -0400 From: Bill Moran To: Greg Smith Cc: pgsql-performance@postgresql.org Subject: Re: select on 22 GB table causes "An I/O error occured while sending to the backend." exception Message-Id: <20080829070852.0000022f.wmoran@collaborativefusion.com> In-Reply-To: References: <82bpze3ou8.fsf@mid.bfk.de> <20080828095325.3c95f867.wmoran@collaborativefusion.com> Organization: Collaborative Fusion X-Mailer: Sylpheed 2.5.0 (GTK+ 2.12.11; i386-portbld-freebsd7.0) Mime-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit X-Virus-Scanned: Maia Mailguard 1.0.1 X-Spam-Status: No, hits=0 tagged_above=0 required=5 tests=none X-Spam-Level: X-Archive-Number: 200808/420 X-Sequence-Number: 31301 In response to Greg Smith : > On Thu, 28 Aug 2008, Bill Moran wrote: > > > In linux, it's possible to tell the OOM killer never to consider > > certain processes for the axe, using /proc magic. See this page: > > http://linux-mm.org/OOM_Killer > > > > Perhaps this should be in the PostgreSQL docs somewhere? > > The fact that > http://www.postgresql.org/docs/current/static/kernel-resources.html#AEN22218 > tells you to flat-out turn off overcommit is the right conservative thing > to be in the documentation as I see it. Sure, it's possible to keep it on > but disable the worst side-effect in some kernels (looks like 2.6.11+, so > no RHEL4 for example). Trying to get into all in the manual is kind of > pushing what's appropriate for the PostgreSQL docs I think. I don't know, Greg. First off, the solution of making the postmaster immune to the OOM killer seems better than disabling overcommit to me anyway; and secondly, I don't understand why we should avoid making the PG documentation as comprehensive as possible, which seems to be what you are saying: "we shouldn't make the PG documentation too comprehensive, because then it will get very big" -- Bill Moran Collaborative Fusion Inc. http://people.collaborativefusion.com/~wmoran/ wmoran@collaborativefusion.com Phone: 412-422-3463x4023