tests/generic: test xfs log recovery metadata LSN ordering
PATCH v2 1/1 tests/generic: test xfs log recovery metadata LSN ordering
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 <[email protected]>
---
v2:
- Use $KILLALL_PROG for killall command.
- Convert to generic test.
v1: http://marc.info/?l=linux-xfs&m=147100402211629&w=2
tests/generic/999 | 88 +++++++++++++++++++++++++++++++++++++++++++++++++++ tests/generic/999.out | 2 ++ tests/generic/group | 1 + 3 files changed, 91 insertions(+) create mode 100755 tests/generic/999 create mode 100644 tests/generic/999.out
tests/generic/999 new file +88 -0
@@ -0,0 +1,88 @@ +#! /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_PROG -9 fsstress > /dev/null 2>&1 + _scratch_unmount > /dev/null 2>&1 +} + +# get standard environment, filters and checks +. ./common/rc + +# Modify as appropriate. +_supported_fs xfs +_supported_os Linux + +_require_scratch +_require_command "$KILLALL_PROG" + +rm -f $seqres.full + +echo "Silence is golden." + +_scratch_mkfs_xfs >> $seqres.full 2>&1 +_scratch_mount || _fail "mount failed" + +for i in $(seq 1 50); do + ($FSSTRESS_PROG -d $SCRATCH_MNT -n 999999 -p 4 >> $seqres.full &) \ + > /dev/null 2>&1 + + # purposely include 0 second sleeps to test shutdown immediately after + # recovery + sleep $((RANDOM % 3)) + $XFS_IO_PROG -xc shutdown $SCRATCH_MNT + + ps -e | grep fsstress > /dev/null 2>&1 + while [ $? == 0 ]; do + $KILLALL_PROG -9 fsstress > /dev/null 2>&1 + wait > /dev/null 2>&1 + ps -e | grep fsstress > /dev/null 2>&1 + done + + # quit if mount fails so we don't shutdown the host fs + _scratch_cycle_mount || _fail "cycle mount failed" +done + +# success, all done +status=0 +exit
tests/generic/999.out new file +2 -0
@@ -0,0 +1,2 @@ +QA output created by 999 +Silence is golden.
tests/generic/group +1 -0
@@ -375,3 +375,4 @@ 370 auto quick richacl 371 auto quick enospc prealloc 372 auto quick clone +999 auto log metadata
Signature · 1 line
2.5.5
Quoted 74 lines · Brian Foster
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 <[email protected]>---
v2:
- Use $KILLALL_PROG for killall command.
- Convert to generic test.
v1: http://marc.info/?l=linux-xfs&m=147100402211629&w=2tests/generic/999 | 88 +++++++++++++++++++++++++++++++++++++++++++++++++++ tests/generic/999.out | 2 ++ tests/generic/group | 1 + 3 files changed, 91 insertions(+) create mode 100755 tests/generic/999 create mode 100644 tests/generic/999.outtests/generic/999 new file +43
-0@@ -0,0 +1,88 @@ +#! /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 /
tests/generic/999 line 53
Quoted 10 lines
tests/generic/999 new file +10
-0+ rm -f $tmp.* + $KILLALL_PROG -9 fsstress > /dev/null 2>&1 + _scratch_unmount > /dev/null 2>&1 +} + +# get standard environment, filters and checks +. ./common/rc + +# Modify as appropriate. +_supported_fs xfs
"xfs" should be "generic"?
tests/generic/999 line 56
Quoted 3 lines
tests/generic/999 new file +3
-0+_supported_os Linux + +_require_scratch
Missed a "_require_scratch_shutdown"?
tests/generic/999 line 63
Quoted 7 lines
tests/generic/999 new file +7
-0+_require_command "$KILLALL_PROG" + +rm -f $seqres.full + +echo "Silence is golden." + +_scratch_mkfs_xfs >> $seqres.full 2>&1
If this is a generic test, _scratch_mkfs should be used here.
tests/generic/999 line 76
Quoted 13 lines
tests/generic/999 new file +13
-0+_scratch_mount || _fail "mount failed" + +for i in $(seq 1 50); do + ($FSSTRESS_PROG -d $SCRATCH_MNT -n 999999 -p 4 >> $seqres.full &) \ + > /dev/null 2>&1 + + # purposely include 0 second sleeps to test shutdown immediately after + # recovery + sleep $((RANDOM % 3)) + $XFS_IO_PROG -xc shutdown $SCRATCH_MNT + + ps -e | grep fsstress > /dev/null 2>&1 + while [ $? == 0 ]; do
Better to use "-eq" for integers.
Thanks,
Eryu
Quoted 36 lines
tests/generic/999 new file +12
-0+ $KILLALL_PROG -9 fsstress > /dev/null 2>&1 + wait > /dev/null 2>&1 + ps -e | grep fsstress > /dev/null 2>&1 + done + + # quit if mount fails so we don't shutdown the host fs + _scratch_cycle_mount || _fail "cycle mount failed" +done + +# success, all done +status=0 +exittests/generic/999.out new file +2
-0@@ -0,0 +1,2 @@ +QA output created by 999 +Silence is golden.tests/generic/group +1
-0@@ -375,3 +375,4 @@ 370 auto quick richacl 371 auto quick enospc prealloc 372 auto quick clone +999 auto log metadataSignature · 6 lines
2.5.5 -- To unsubscribe from this list: send the line "unsubscribe fstests" in the body of a message to [email protected] More majordomo info at http://vger.kernel.org/majordomo-info.html