From: Marc Meledandri Subject: Re: Oops at shutdown: kernel BUG at fs/ext4/super.c:804! Date: Mon, 9 Sep 2013 13:05:00 -0400 Message-ID: References: <20130909162319.GG1612@quack.suse.cz> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8 To: Jan Kara , linux-ext4@vger.kernel.org Return-path: Received: from mail-pd0-f180.google.com ([209.85.192.180]:47675 "EHLO mail-pd0-f180.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755091Ab3IIRFb (ORCPT ); Mon, 9 Sep 2013 13:05:31 -0400 Received: by mail-pd0-f180.google.com with SMTP id y10so6395330pdj.25 for ; Mon, 09 Sep 2013 10:05:30 -0700 (PDT) In-Reply-To: <20130909162319.GG1612@quack.suse.cz> Sender: linux-ext4-owner@vger.kernel.org List-ID: On Mon, Sep 9, 2013 at 12:23 PM, Jan Kara wrote: > On Sun 08-09-13 18:28:40, Marc Meledandri wrote: >> Been seeing this problem at shutdown on 3.10.x kernels. >> >> I saw a previous similar thread, but it seemed related to NFS whereas this is >> affecting my raid device (dm-0). > Well, this is likely caused by inodes in filesystem on dm-0 being still > held by someone. Are you exporting dm-0 via NFS? Because there have been > recently bugs in NFS in this area (this patch should have fixed them: > http://www.spinics.net/lists/kernel/msg1592640.html). > Thank you. I'd seen that patch, but don't have a remotely similar configuration as specified in the STR. I also thought NFS was out of the picture at that stage of the shutdown, but after re-reading the patch notes, it looks like the earlier delegation may do the trick for my standard NFS implementation as well. I'll test with this patch. Looks hopeful.