From: Jamie Lokier Subject: Re: [PATCH 0/4] (RESEND) ext3[34] barrier changes Date: Wed, 21 May 2008 20:36:33 +0100 Message-ID: <20080521193633.GA26780@shareable.org> References: <482DDA56.6000301@redhat.com> <20080518211140.b29bee30.akpm@linux-foundation.org> <200805191316.27551.chris.mason@oracle.com> <200805191439.36577.chris.mason@oracle.com> <20080521112224.GD5028@ucw.cz> <20080521110324.668048e0.akpm@linux-foundation.org> <20080521182949.GL8581@mit.edu> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii To: Theodore Tso , Andrew Morton , Pavel Machek , Chris Mason , Eric Sandeen , Andi Return-path: Received: from mail2.shareable.org ([80.68.89.115]:50178 "EHLO mail2.shareable.org" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1759063AbYEUTiS (ORCPT ); Wed, 21 May 2008 15:38:18 -0400 Content-Disposition: inline In-Reply-To: <20080521182949.GL8581@mit.edu> Sender: linux-ext4-owner@vger.kernel.org List-ID: Theodore Tso wrote: > On Wed, May 21, 2008 at 11:03:24AM -0700, Andrew Morton wrote: > > > > afaik there is no need to enable this feature if the machine (actually > > the disks) are on a UPS, yes? > > Yes, as long as you're confident that there won't be a kernel > bug/regression causing a lockup while the server is under severe > memory pressure while doing lots of fsync's, file creations, renames, > etc. And as long as your 100% confident that UPS's will never fail, > janitors will never accidentally hit the Emergency Power Office Can a kernel lockup cause this kind of corruption? Will a system reboot wipe the disk's write cache? I had imagined only power loss would prevent the disk from writing it's cache eventually; is that wrong? -- Jamie