Received: with ECARTIS (v1.0.0; list linux-xfs); Wed, 28 May 2003 18:35:47 -0700 (PDT) Received: from shell.wgops.com (postfix@dsl092-192-108.sea1.dsl.speakeasy.net [66.92.192.108]) by oss.sgi.com (8.12.9/8.12.9) with SMTP id h4T1ZV2x005428 for ; Wed, 28 May 2003 18:35:32 -0700 Received: by shell.wgops.com (Postfix, from userid 65534) id 46718DA; Wed, 28 May 2003 19:35:31 -0600 (MDT) Received: from [10.1.2.77] (jobe.wgops.com [10.1.2.77]) by shell.wgops.com (Postfix) with ESMTP id B54A4D1; Wed, 28 May 2003 19:35:29 -0600 (MDT) Date: Wed, 28 May 2003 19:35:30 -0600 From: Michael Loftis To: Pietro Abate , linux-xfs@oss.sgi.com Subject: Re: xfs fails to recover a raid0 root filesystem Message-ID: <163200389.1054150530@jobe> In-Reply-To: <20030529002010.GC1275@anu.edu.au> References: <20030529002010.GC1275@anu.edu.au> X-Mailer: Mulberry/3.0.3 (Win32) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Content-Disposition: inline X-archive-position: 4172 X-ecartis-version: Ecartis v1.0.0 Sender: linux-xfs-bounce@oss.sgi.com Errors-to: linux-xfs-bounce@oss.sgi.com X-original-sender: mloftis@wgops.com Precedence: bulk X-list: linux-xfs Content-Length: 1785 Lines: 50 I just had this happen on a large-ish (150+Gb) array. Had nothing to do with RAID. You have to boot rescue disk and use xfs_repair. You may have to zero the logfile. Also note that if oyu're using quotas it might go off and re-figure everyones quota's at mount time which takes a long time. --On Thursday, May 29, 2003 10:20 AM +1000 Pietro Abate wrote: > Hi all, > I've installed xfs on a netra (sparc) workstation. After a power > failoure xfs just hangs trying to recover the root fs. The fs is > actually sitting on the top of a raid0 configuration. I don't really > understand if the problem is related with xfs or with raid (but I think > it's xfs ...). > > I tried to run the raid in degradeted mode passing raid=noautodetect > md=0,/dev/sda4 ... to silo, to try to isolate the problem, but it didn't > solve the problem. > > my questions now: > - is there a way of forcing xfs to ignore the recovery stage and give me a > shell that I can use to run xfs_repair ? (any silo parameter ?) > > - is there any rescue system (cdrom) for sparc with raid and xfs support > that I can use to recover the fs ? > > - if I manage to boot it from the net, is there any rescue system I > can use ? (if not, my plan is to install a small system on the swap > partition, build a kernel with xfs and raid, boot it, recover the fs, > and boot back to the old installation: is it a good plan ?) > > - is xfs supposed to work properly (I'm using 2.4.19 + xfs snapshot) > with raid0 ? is it a problem having the root fs unsing xfs with a such > configuration ? > > please help... :-) > > p > > > -- > Civilization advances by extending the number > of important operations which we can perform > without thinking. (Alfred North Whitehead) > >