From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: with ECARTIS (v1.0.0; list xfs); Wed, 19 Jul 2006 07:11:26 -0700 (PDT) Received: from orca.ele.uri.edu (orca.ele.uri.edu [131.128.51.63]) by oss.sgi.com (8.12.10/8.12.10/SuSE Linux 0.7) with ESMTP id k6JEB7DW009520 for ; Wed, 19 Jul 2006 07:11:07 -0700 Subject: Re: stable xfs From: Ming Zhang Reply-To: mingz@ele.uri.edu In-Reply-To: <20060719055621.GA1491@tuatara.stupidest.org> References: <1153150223.4532.24.camel@localhost.localdomain> <17595.47312.720883.451573@base.ty.sabi.co.UK> <1153262166.2669.267.camel@localhost.localdomain> <17597.27469.834961.186850@base.ty.sabi.co.UK> <1153272044.2669.282.camel@localhost.localdomain> <20060719055621.GA1491@tuatara.stupidest.org> Content-Type: text/plain Date: Wed, 19 Jul 2006 10:10:06 -0400 Message-Id: <1153318206.2691.43.camel@localhost.localdomain> Mime-Version: 1.0 Content-Transfer-Encoding: 7bit Sender: xfs-bounce@oss.sgi.com Errors-To: xfs-bounce@oss.sgi.com List-Id: xfs To: Chris Wedgwood Cc: Peter Grandi , Linux XFS On Tue, 2006-07-18 at 22:56 -0700, Chris Wedgwood wrote: > On Tue, Jul 18, 2006 at 09:20:44PM -0400, Ming Zhang wrote: > > > when u say large parallel storage system, you mean independent > > spindles right? but most people will have all disks configured in > > one RAID5/6 and thus it is not parallel any more. > > it depends, you might have 100s of spindles in groups, you don't make > a giant raid5/6 array with that many disks, you make a number of > smaller arrays right > > > i think with write barrier support, system without UPS should be ok. > > with barrier support a UPS shouldn't be necessary > > > considering even u have UPS, kernel oops in other parts still can > > take the FS down. > i mean with UPS and huge write cache, but no write barrier. > but a crash won't cause writes to be 'reordered' > > > reordering is bad because the fs pushes writes down in a manner that > means when it comes back it will be able to make it self consistent, > so if you have a number of writes pending and some of them are lost, > and those that are lost are not the most recent writes because of > reordering, you can end up with a corrupt fs