From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Mon, 29 Oct 2007 14:10:38 -0700 (PDT) Received: from larry.melbourne.sgi.com (larry.melbourne.sgi.com [134.14.52.130]) by oss.sgi.com (8.12.11.20060308/8.12.10/SuSE Linux 0.7) with SMTP id l9TLAUrr010243 for ; Mon, 29 Oct 2007 14:10:33 -0700 Date: Tue, 30 Oct 2007 08:10:22 +1100 From: David Chinner Subject: Re: Crash with XFS on top of DRBD (DRBD 8.0.6 svn / Kernel 2.6.22) Message-ID: <20071029211021.GK995458@sgi.com> References: <20071029101329.ietaisee@trusted.lncsa.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20071029101329.ietaisee@trusted.lncsa.com> Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: Laurent Caron Cc: drbd-user@linbit.com, xfs@oss.sgi.com On Mon, Oct 29, 2007 at 10:14:19AM +0100, Laurent Caron wrote: > Hi, > > I'm back with my crash, oomkiller..... problems on my DRBD cluster of 2 > servers. > > I compiled a 2.6.22 kernel with slab/slab debugging turned on. > > Here is the last oom-killer message I got on that server. You ran out of low memory. Slab didn't use it all, so I'm not sure what would be using it. You should post OOM invokations to LKML. Given DRDB is te unusual thing your are using, I'd be looking there first... Cheers, Dave. -- Dave Chinner Principal Engineer SGI Australian Software Group