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 <bfoster@xxxxxxxxxx> > --- > > 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. > + _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 Thanks, Eryu _______________________________________________ xfs mailing list xfs@xxxxxxxxxxx http://oss.sgi.com/mailman/listinfo/xfs
