* [PATCH v3 1/2] common/rc: add functions to check or write objects under /sys/fs/$FSTYP
@ 2016-05-29 4:28 Zorro Lang
2016-05-29 4:28 ` [PATCH v3 2/2] xfs/006: new case to test xfs fail_at_unmount error handling Zorro Lang
2016-06-11 14:18 ` [PATCH v3 1/2] common/rc: add functions to check or write objects under /sys/fs/$FSTYP Eryu Guan
0 siblings, 2 replies; 4+ messages in thread
From: Zorro Lang @ 2016-05-29 4:28 UTC (permalink / raw)
To: fstests; +Cc: sandeen, Zorro Lang
XFS add more configurations in /sys/fs/xfs recently. For use
them, this patch add some common functions for:
1. "require" a file/dir in /sys/fs/${FSTYP}.
2. write a file in /sys/fs/${FSTYP}.
For common use, these functions can be used by other filesystems.
Signed-off-by: Zorro Lang <zlang@redhat.com>
---
Hi,
V3 revert dname=$(basename $(readlink -f $dev)) line.
I think we don't need to check if /sys/fs/${FSTYP}/${dname}/$target
is exist in _set_fs_sys_fs_param(). Because if no this file,
*echo* will report error, and it will disturb golden file.
Thanks,
Zorro
common/rc | 40 ++++++++++++++++++++++++++++++++++++++++
1 file changed, 40 insertions(+)
diff --git a/common/rc b/common/rc
index 51092a0..759ec30 100644
--- a/common/rc
+++ b/common/rc
@@ -3556,6 +3556,46 @@ run_fsx()
fi
}
+_require_fs_sys_fs()
+{
+ local dev=$1
+ local target=$2
+ local tmp_mnt=`mktemp -d`
+
+ if [ ! -b "$dev" -o -z "$target" ];then
+ _fail "Usage: _require_fs_sys_fs <device> <sysfs_path>"
+ fi
+
+ local dname=$(basename $(readlink -f $dev))
+ _mount -t $FSTYP `_common_dev_mount_options` $dev $tmp_mnt
+ if [ $? -ne 0 ];then
+ rm -f $tmp_mnt
+ _fail "could not mount; mkfs first in your test?"
+ elif [ ! -e /sys/fs/${FSTYP}/${dname}/$target ];then
+ umount $tmp_mnt
+ rm -f $tmp_mnt
+ _notrun "/sys/fs/${FSTYP}/${dname}/$target: No such file or directory"
+ fi
+ umount $tmp_mnt
+ rm -rf $tmp_mnt
+}
+
+_set_fs_sys_fs_param()
+{
+ local dev=$1
+ shift
+ local target=$1
+ shift
+ local content="$*"
+
+ if [ ! -b "$dev" -o -z "$target" -o -z "$content" ];then
+ _fail "Usage: _set_sys_fs_param <mounted_device> <target> <content>"
+ fi
+
+ local dname=$(basename $(readlink -f $dev))
+ echo "$content" > /sys/fs/${FSTYP}/${dname}/$target
+}
+
init_rc
################################################################################
--
2.5.5
^ permalink raw reply related [flat|nested] 4+ messages in thread* [PATCH v3 2/2] xfs/006: new case to test xfs fail_at_unmount error handling 2016-05-29 4:28 [PATCH v3 1/2] common/rc: add functions to check or write objects under /sys/fs/$FSTYP Zorro Lang @ 2016-05-29 4:28 ` Zorro Lang 2016-06-11 14:39 ` Eryu Guan 2016-06-11 14:18 ` [PATCH v3 1/2] common/rc: add functions to check or write objects under /sys/fs/$FSTYP Eryu Guan 1 sibling, 1 reply; 4+ messages in thread From: Zorro Lang @ 2016-05-29 4:28 UTC (permalink / raw) To: fstests; +Cc: sandeen, Zorro Lang XFS bring in a new configuration under /sys/fs/xfs/error, named fail_at_unmount. It's used to stop unmount retrying forever when it hit IO error. This case try to unmount an faulty dm device, and to sure unmount won't retry forever if fail_at_unmount=1. Signed-off-by: Zorro Lang <zlang@redhat.com> --- Hi, Nothing changed in this V3. Due to I modify [PATCH v3 1/2], so I think if I send this 2/2 patch with 1/2 together, maybe easy to review and merge? :) Thanks, Zorro tests/xfs/006 | 81 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ tests/xfs/006.out | 2 ++ tests/xfs/group | 1 + 3 files changed, 84 insertions(+) create mode 100755 tests/xfs/006 create mode 100644 tests/xfs/006.out diff --git a/tests/xfs/006 b/tests/xfs/006 new file mode 100755 index 0000000..c134b1c --- /dev/null +++ b/tests/xfs/006 @@ -0,0 +1,81 @@ +#! /bin/bash +# FS QA Test 006 +# +# Test xfs' "fail at unmount" error handling configuration. Stop +# XFS retrying to writeback forever when unmount. +# +#----------------------------------------------------------------------- +# Copyright (c) 2016 YOUR NAME HERE. 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.* + _dmerror_cleanup +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter +. ./common/dmerror + +# remove previous $seqres.full before test +rm -f $seqres.full + +# real QA test starts here +_supported_fs xfs +_supported_os Linux +_require_dm_target error +_require_scratch + +_scratch_mkfs > $seqres.full 2>&1 +_require_fs_sys_fs $SCRATCH_DEV error/fail_at_unmount + +# The device is still a linear device at here +_dmerror_init +_dmerror_mount + +# Enable fail_at_unmount +_set_fs_sys_fs_param $DMERROR_DEV error/fail_at_unmount 1 + +# make some IO at here. If these data still not writeback from +# journal to data, it will hit writeback error after load error +# tabel in next step. +_pwrite_byte 0x61 0 100M $SCRATCH_MNT/file >/dev/null +_dmerror_load_error_table + +# Even above data already all writeback, umount will cause XFS +# try to writeback something to root inode. It still can trigger +# umount fail. +_dmerror_unmount + +echo "Silence is golden" +_dmerror_cleanup + +# success, all done +status=0 +exit diff --git a/tests/xfs/006.out b/tests/xfs/006.out new file mode 100644 index 0000000..675c1b7 --- /dev/null +++ b/tests/xfs/006.out @@ -0,0 +1,2 @@ +QA output created by 006 +Silence is golden diff --git a/tests/xfs/group b/tests/xfs/group index f4c6816..39169ea 100644 --- a/tests/xfs/group +++ b/tests/xfs/group @@ -4,6 +4,7 @@ 004 db auto quick 005 auto quick 007 auto quota quick +006 auto quick unmount 008 rw ioctl auto quick 009 rw ioctl auto prealloc quick 010 auto quick repair -- 2.5.5 ^ permalink raw reply related [flat|nested] 4+ messages in thread
* Re: [PATCH v3 2/2] xfs/006: new case to test xfs fail_at_unmount error handling 2016-05-29 4:28 ` [PATCH v3 2/2] xfs/006: new case to test xfs fail_at_unmount error handling Zorro Lang @ 2016-06-11 14:39 ` Eryu Guan 0 siblings, 0 replies; 4+ messages in thread From: Eryu Guan @ 2016-06-11 14:39 UTC (permalink / raw) To: Zorro Lang; +Cc: fstests, sandeen On Sun, May 29, 2016 at 12:28:41PM +0800, Zorro Lang wrote: > XFS bring in a new configuration under /sys/fs/xfs/error, named > fail_at_unmount. It's used to stop unmount retrying forever when > it hit IO error. Refering to the kernel commit IDs in commit log would be good for future reference. > > This case try to unmount an faulty dm device, and to sure unmount > won't retry forever if fail_at_unmount=1. > > Signed-off-by: Zorro Lang <zlang@redhat.com> > --- > > Hi, > > Nothing changed in this V3. Due to I modify [PATCH v3 1/2], so I think > if I send this 2/2 patch with 1/2 together, maybe easy to review and > merge? :) > > Thanks, > Zorro > > tests/xfs/006 | 81 +++++++++++++++++++++++++++++++++++++++++++++++++++++++ > tests/xfs/006.out | 2 ++ > tests/xfs/group | 1 + > 3 files changed, 84 insertions(+) > create mode 100755 tests/xfs/006 > create mode 100644 tests/xfs/006.out > > diff --git a/tests/xfs/006 b/tests/xfs/006 > new file mode 100755 > index 0000000..c134b1c > --- /dev/null > +++ b/tests/xfs/006 > @@ -0,0 +1,81 @@ > +#! /bin/bash > +# FS QA Test 006 > +# > +# Test xfs' "fail at unmount" error handling configuration. Stop > +# XFS retrying to writeback forever when unmount. > +# > +#----------------------------------------------------------------------- > +# Copyright (c) 2016 YOUR NAME HERE. All Rights Reserved. ^^^^^^^^^^^^^^ Red Hat Inc. :) > +# > +# 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.* > + _dmerror_cleanup > +} > + > +# get standard environment, filters and checks > +. ./common/rc > +. ./common/filter > +. ./common/dmerror > + > +# remove previous $seqres.full before test > +rm -f $seqres.full > + > +# real QA test starts here > +_supported_fs xfs > +_supported_os Linux > +_require_dm_target error > +_require_scratch > + > +_scratch_mkfs > $seqres.full 2>&1 > +_require_fs_sys_fs $SCRATCH_DEV error/fail_at_unmount > + > +# The device is still a linear device at here > +_dmerror_init > +_dmerror_mount > + > +# Enable fail_at_unmount > +_set_fs_sys_fs_param $DMERROR_DEV error/fail_at_unmount 1 I tested this by setting "fail_at_unmount 0" on 4.7-rc1 kernel, I think test should hang at umount time in this case, but test still passed. Seems the case didn't test what it supposed to test. > + > +# make some IO at here. If these data still not writeback from > +# journal to data, it will hit writeback error after load error > +# tabel in next step. > +_pwrite_byte 0x61 0 100M $SCRATCH_MNT/file >/dev/null > +_dmerror_load_error_table > + > +# Even above data already all writeback, umount will cause XFS > +# try to writeback something to root inode. It still can trigger > +# umount fail. > +_dmerror_unmount > + > +echo "Silence is golden" > +_dmerror_cleanup It's called in _cleanup, not needed here. Thanks, Eryu > + > +# success, all done > +status=0 > +exit > diff --git a/tests/xfs/006.out b/tests/xfs/006.out > new file mode 100644 > index 0000000..675c1b7 > --- /dev/null > +++ b/tests/xfs/006.out > @@ -0,0 +1,2 @@ > +QA output created by 006 > +Silence is golden > diff --git a/tests/xfs/group b/tests/xfs/group > index f4c6816..39169ea 100644 > --- a/tests/xfs/group > +++ b/tests/xfs/group > @@ -4,6 +4,7 @@ > 004 db auto quick > 005 auto quick > 007 auto quota quick > +006 auto quick unmount > 008 rw ioctl auto quick > 009 rw ioctl auto prealloc quick > 010 auto quick repair > -- > 2.5.5 > > -- > To unsubscribe from this list: send the line "unsubscribe fstests" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html ^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH v3 1/2] common/rc: add functions to check or write objects under /sys/fs/$FSTYP 2016-05-29 4:28 [PATCH v3 1/2] common/rc: add functions to check or write objects under /sys/fs/$FSTYP Zorro Lang 2016-05-29 4:28 ` [PATCH v3 2/2] xfs/006: new case to test xfs fail_at_unmount error handling Zorro Lang @ 2016-06-11 14:18 ` Eryu Guan 1 sibling, 0 replies; 4+ messages in thread From: Eryu Guan @ 2016-06-11 14:18 UTC (permalink / raw) To: Zorro Lang; +Cc: fstests, sandeen On Sun, May 29, 2016 at 12:28:40PM +0800, Zorro Lang wrote: > XFS add more configurations in /sys/fs/xfs recently. For use > them, this patch add some common functions for: > 1. "require" a file/dir in /sys/fs/${FSTYP}. > 2. write a file in /sys/fs/${FSTYP}. > > For common use, these functions can be used by other filesystems. > > Signed-off-by: Zorro Lang <zlang@redhat.com> > --- > > Hi, > > V3 revert dname=$(basename $(readlink -f $dev)) line. > > I think we don't need to check if /sys/fs/${FSTYP}/${dname}/$target > is exist in _set_fs_sys_fs_param(). Because if no this file, > *echo* will report error, and it will disturb golden file. > > Thanks, > Zorro > > common/rc | 40 ++++++++++++++++++++++++++++++++++++++++ > 1 file changed, 40 insertions(+) > > diff --git a/common/rc b/common/rc > index 51092a0..759ec30 100644 > --- a/common/rc > +++ b/common/rc > @@ -3556,6 +3556,46 @@ run_fsx() > fi > } > > +_require_fs_sys_fs() _require_fs_sysfs seems better to me :) And a comment about the purpose of this helper would be fine, e.g. the comment Eric wrote in his review: "Test for the existence of a sysfs entry at /sys/fs/$FSTYP/$DEV/$ENTRY" > +{ > + local dev=$1 > + local target=$2 > + local tmp_mnt=`mktemp -d` > + > + if [ ! -b "$dev" -o -z "$target" ];then > + _fail "Usage: _require_fs_sys_fs <device> <sysfs_path>" > + fi > + > + local dname=$(basename $(readlink -f $dev)) We do have a helper "_short_dev" to do this, e.g. local dname=`_short_dev $dev` > + _mount -t $FSTYP `_common_dev_mount_options` $dev $tmp_mnt > + if [ $? -ne 0 ];then > + rm -f $tmp_mnt > + _fail "could not mount; mkfs first in your test?" Having function name as prefix in error message makes debugging easier, I think. > + elif [ ! -e /sys/fs/${FSTYP}/${dname}/$target ];then > + umount $tmp_mnt Use $UMOUNT_PROG here, and in all other umount calls. > + rm -f $tmp_mnt > + _notrun "/sys/fs/${FSTYP}/${dname}/$target: No such file or directory" > + fi > + umount $tmp_mnt > + rm -rf $tmp_mnt > +} > + > +_set_fs_sys_fs_param() _set_fs_sysfs_param :) And a comment about its purpose. Thanks, Eryu > +{ > + local dev=$1 > + shift > + local target=$1 > + shift > + local content="$*" > + > + if [ ! -b "$dev" -o -z "$target" -o -z "$content" ];then > + _fail "Usage: _set_sys_fs_param <mounted_device> <target> <content>" > + fi > + > + local dname=$(basename $(readlink -f $dev)) > + echo "$content" > /sys/fs/${FSTYP}/${dname}/$target > +} > + > init_rc > > ################################################################################ > -- > 2.5.5 > > -- > To unsubscribe from this list: send the line "unsubscribe fstests" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html ^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2016-06-11 14:39 UTC | newest] Thread overview: 4+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2016-05-29 4:28 [PATCH v3 1/2] common/rc: add functions to check or write objects under /sys/fs/$FSTYP Zorro Lang 2016-05-29 4:28 ` [PATCH v3 2/2] xfs/006: new case to test xfs fail_at_unmount error handling Zorro Lang 2016-06-11 14:39 ` Eryu Guan 2016-06-11 14:18 ` [PATCH v3 1/2] common/rc: add functions to check or write objects under /sys/fs/$FSTYP Eryu Guan
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox