From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from relay.sgi.com (relay2.corp.sgi.com [137.38.102.29]) by oss.sgi.com (Postfix) with ESMTP id CBC1A7CA2 for ; Mon, 15 Aug 2016 06:46:56 -0500 (CDT) Received: from cuda.sgi.com (cuda3.sgi.com [192.48.176.15]) by relay2.corp.sgi.com (Postfix) with ESMTP id 9D92B3040FD for ; Mon, 15 Aug 2016 04:46:53 -0700 (PDT) Received: from mx1.redhat.com (mx1.redhat.com [209.132.183.28]) by cuda.sgi.com with ESMTP id hVkOqzjcNYhfyzzQ (version=TLSv1.2 cipher=ECDHE-RSA-AES256-GCM-SHA384 bits=256 verify=NO) for ; Mon, 15 Aug 2016 04:46:52 -0700 (PDT) Date: Mon, 15 Aug 2016 07:46:49 -0400 From: Brian Foster Subject: Re: [PATCH] tests/xfs: test log recovery metadata LSN ordering Message-ID: <20160815114649.GA14796@bfoster.bfoster> References: <1471004010-52985-1-git-send-email-bfoster@redhat.com> <20160815052933.GE27776@eguan.usersys.redhat.com> MIME-Version: 1.0 Content-Disposition: inline In-Reply-To: <20160815052933.GE27776@eguan.usersys.redhat.com> List-Id: XFS Filesystem from SGI List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Errors-To: xfs-bounces@oss.sgi.com Sender: xfs-bounces@oss.sgi.com To: Eryu Guan Cc: fstests@vger.kernel.org, xfs@oss.sgi.com On Mon, Aug 15, 2016 at 01:29:33PM +0800, Eryu Guan wrote: > On Fri, Aug 12, 2016 at 08:13:30AM -0400, Brian Foster wrote: > > XFS had a bug that lead to a possible out-of-order log recovery > > situation (e.g., replay a stale modification from the log over more > > recent metadata in destination buffer). This resulted in false > > corruption reports during log recovery and thus mount failure. > > > > This condition is caused by system crash or filesystem shutdown shortly > > after a successful log recovery. Add a test to run a combined workload, > > fs shutdown and log recovery loop known to reproduce the problem on > > affected kernels. > > > > Signed-off-by: Brian Foster > > --- > > > > This test reproduces the problem described and addressed in the > > following patchset: > > > > http://oss.sgi.com/pipermail/xfs/2016-August/050840.html > > > > It runs anywhere from 50-100s in the couple of environments I've tested > > on so far and reproduces the problem for me with 100% reliability. Note > > that the bug only affects crc=1 kernels. > > Looks good overall, and tested with the above patchset applied and test > passed without problems. Some minor issues inline > > > > > Brian > > > > tests/xfs/999 | 87 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ > > tests/xfs/999.out | 2 ++ > > tests/xfs/group | 1 + > > 3 files changed, 90 insertions(+) > > create mode 100755 tests/xfs/999 > > create mode 100644 tests/xfs/999.out > > > > diff --git a/tests/xfs/999 b/tests/xfs/999 > > new file mode 100755 > > index 0000000..f9dd7f7 > > --- /dev/null > > +++ b/tests/xfs/999 > > @@ -0,0 +1,87 @@ > > +#! /bin/bash > > +# FS QA Test No. 999 > > +# > > +# Test XFS log recovery ordering on v5 superblock filesystems. XFS had a problem > > +# where it would incorrectly replay older modifications from the log over more > > +# recent versions of metadata due to failure to update metadata LSNs during log > > +# recovery. This could result in false positive reports of corruption during log > > +# recovery and permanent mount failure. > > +# > > +# To test this situation, run frequent shutdowns immediately after log recovery. > > +# Ensure that log recovery does not recover stale modifications and cause > > +# spurious corruption reports and/or mount failures. > > +# > > +#----------------------------------------------------------------------- > > +# Copyright (c) 2016 Red Hat, Inc. All Rights Reserved. > > +# > > +# This program is free software; you can redistribute it and/or > > +# modify it under the terms of the GNU General Public License as > > +# published by the Free Software Foundation. > > +# > > +# This program is distributed in the hope that it would be useful, > > +# but WITHOUT ANY WARRANTY; without even the implied warranty of > > +# MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the > > +# GNU General Public License for more details. > > +# > > +# You should have received a copy of the GNU General Public License > > +# along with this program; if not, write the Free Software Foundation, > > +# Inc., 51 Franklin St, Fifth Floor, Boston, MA 02110-1301 USA > > +#----------------------------------------------------------------------- > > +# > > + > > +seq=`basename $0` > > +seqres=$RESULT_DIR/$seq > > +echo "QA output created by $seq" > > + > > +here=`pwd` > > +tmp=/tmp/$$ > > +status=1 # failure is the default! > > +trap "_cleanup; exit \$status" 0 1 2 3 15 > > + > > +_cleanup() > > +{ > > + cd / > > + rm -f $tmp.* > > + killall -9 fsstress > /dev/null 2>&1 > > We need a '_require_command "$KILLALL_PROGA" killall' and use > $KILLALL_PROG in the test. > Ok. > > + _scratch_unmount > /dev/null 2>&1 > > +} > > + > > +# get standard environment, filters and checks > > +. ./common/rc > > + > > +# Modify as appropriate. > > +_supported_fs xfs > > I'm wondering if this test can be made generic by adding a > "_require_scratch_shutdown"? Like generic/042 to generic/051 > Hmm, probably. I'll give it a try, thanks! Brian > Thanks, > Eryu _______________________________________________ xfs mailing list xfs@oss.sgi.com http://oss.sgi.com/mailman/listinfo/xfs