From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jan Kara Subject: Re: [PATCH, RFC] ext3: don't clear orphan list on ro mount with errors Date: Tue, 4 Sep 2012 23:27:06 +0200 Message-ID: <20120904212706.GA10636@quack.suse.cz> References: <503BC685.7090707@redhat.com> <503BCAE0.7010900@redhat.com> <20120828080225.GB5146@quack.suse.cz> <50464DCD.2020209@redhat.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Jan Kara , ext4 development To: Eric Sandeen Return-path: Received: from cantor2.suse.de ([195.135.220.15]:55029 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751032Ab2IDV1I (ORCPT ); Tue, 4 Sep 2012 17:27:08 -0400 Content-Disposition: inline In-Reply-To: <50464DCD.2020209@redhat.com> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Tue 04-09-12 13:51:57, Eric Sandeen wrote: > On 8/28/12 3:02 AM, Jan Kara wrote: > > On Mon 27-08-12 14:30:40, Eric Sandeen wrote: > >> When we have a filesystem with an orphan inode list *and* in error > >> state, things behave differently if: > >> > >> 1) e2fsck -p is done prior to mount: e2fsck fixes things and exits > >> happily (barring other significant problems) > >> > >> vs. > >> > >> 2) mount is done first, then e2fsck -p: due to the orphan inode > >> list removal, more errors are found and e2fsck exits with > >> UNEXPECTED INCONSISTENCY. > >> > >> The 2nd case above, on the root filesystem, has the tendency to halt > >> the boot process, which is unfortunate. > >> > >> The situation can be improved by not clearing the orphan > >> inode list when the fs is mounted readonly. > > Yeah, makes sense. I've added the patch to my tree. Thanks. > > > > Honza > > After a little more investigation, I'm now wondering if this is really > worth doing. > > e2fsck zaps the orphan list just like the kernel does: > > * If the filesystem contains errors, don't run the orphan > * list, since the orphan list can't be trusted; and we're > * going to be running a full e2fsck run anyway... > > and my 1) and 2) differences above were due to testing an older version > of e2fsck which didn't properly propagate the error flag. (Sorry...) > > Since upstream e2fsck will _also_ ignore the orphan inode list, there's > probably no great reason for preserving it on a readonly mount after all, > unless it's just to minimize changes when mounting RO (which may be a > sufficient reason, I suppose). So feel free to take it or leave it, > I guess. Since I've already pushed this to Linus and minimizing changes on RO filesystem makes sense anyway, I'll leave the patch in... Thanks for the update. Honza -- Jan Kara SUSE Labs, CR