From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Wed, 10 Oct 2007 14:57:46 -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 l9ALvfFL031487 for ; Wed, 10 Oct 2007 14:57:43 -0700 Date: Thu, 11 Oct 2007 07:57:33 +1000 From: David Chinner Subject: Re: running xfs_repair on large partitions Message-ID: <20071010215733.GL995458@sgi.com> References: <20071010143337.GA2815@apartia.fr> <20071010191452.1fdad848@galadriel.home> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline Content-Transfer-Encoding: 8bit In-Reply-To: <20071010191452.1fdad848@galadriel.home> Sender: xfs-bounce@oss.sgi.com Errors-to: xfs-bounce@oss.sgi.com List-Id: xfs To: Emmanuel Florac Cc: Louis-David Mitterrand , xfs-dev , xfs-oss On Wed, Oct 10, 2007 at 07:14:52PM +0200, Emmanuel Florac wrote: > Le Wed, 10 Oct 2007 16:33:38 +0200 vous écriviez: > > > 1) it would be nice to have a way to know xfs_repair's version (-V ?), > > when using it from a rescue disk I'm never sure if I should get a newer > > one. > > Did you try "xfs_repair -V" as you suggest? On my system, it replies with > the version... > > > 2) on a 32bit system, using xfsprogs version 2.9.0 , it seems xfs_repair > > will fail if its process exceeds 4G. Is that right? Is there a way to > > circumvent that limitation? > > No, this is precisely what 32 bits mean. A 32 bits process can't address > more than 2^32 bits, which is 4GB. I think it's 2GB for a process by default on linux, because the other 2GB is used by the kernel. The split is configurable IIRC. > However I'm surprised you have this > problem; I've xfs_repaired up to 16TB filesystem (maximum manageable with a > 32 bits kernel ) without such problem. Memory usage depends on the number of inodes in the filesystem as well. Cheers, Dave. -- Dave Chinner Principal Engineer SGI Australian Software Group