Re: [PATCH] fstests: test platform_check_ismounted do its work

[Eryu Guan <eguan@redhat.com>]

On Wed, Sep 14, 2016 at 04:02:09PM +0800, Zorro Lang wrote:
> There's a bug, the ustat() sysetm call is not implemented on someone
> arch, but xfsprogs uses this to detect whether a filesystem is
> mounted in platform_check_ismounted() function.
> 
> If ustat() is not implemented, platform_check_ismounted() return
> "isn't mounted" if a filesystem is mounted, that's a big problem, it
> will cause corruption.
> 
> Signed-off-by: Zorro Lang <zlang@redhat.com>
> ---
>  tests/xfs/284     | 113 ++++++++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/xfs/284.out |   8 ++++
>  tests/xfs/group   |   1 +
>  3 files changed, 122 insertions(+)
>  create mode 100755 tests/xfs/284
>  create mode 100644 tests/xfs/284.out
> 
> diff --git a/tests/xfs/284 b/tests/xfs/284
> new file mode 100755
> index 0000000..53fbcb1
> --- /dev/null
> +++ b/tests/xfs/284
> @@ -0,0 +1,113 @@
> +#! /bin/bash
> +# FS QA Test 284
> +#
> +# Make sure platform_check_ismounted() do its work. A bug cause
> +# platform_check_ismounted return "isn't mounted" status, even if
> +# the filesystem is mounted. That maybe cause corruption, so must
> +# make sure this function always do its work.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2016 Red Hat.  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.*
> +	rm -f $METADUMP_FILE 2>/dev/null
> +	rm -f $COPY_FILE 2>/dev/null
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# remove previous $seqres.full before test
> +rm -f $seqres.full
> +
> +# real QA test starts here
> +_supported_fs xfs
> +_supported_os Linux
> +_require_test
> +_require_scratch
> +
> +function filter_mounted()
> +{
> +	grep "mounted" | _filter_scratch | head -1
> +}
> +
> +METADUMP_FILE="${TEST_DIR}/${seq}_metadump"
> +COPY_FILE="${TEST_DIR}/${seq}_copyfile"
> +
> +# Test dump a mounted device
> +_scratch_mount
> +_scratch_metadump $METADUMP_FILE 2>&1 | filter_mounted
> +[ ${PIPESTATUS[0]} -eq 0 ] && echo "Test dump a mounted device [Failed]"

I don't think these status checks are needed, because if these
metadump/copy/repair/mkfs commands proceed on mounted fs, the golden
image will be broken.

> +_scratch_unmount
> +
> +# Test restore a mounted device

"restore to a mounted device"

> +_scratch_metadump $METADUMP_FILE
> +_scratch_mount
> +xfs_mdrestore $METADUMP_FILE $SCRATCH_DEV 2>&1 | filter_mounted
> +[ ${PIPESTATUS[0]} -eq 0 ] && echo "Test restore a mounted device [Failed]"
> +_scratch_unmount
> +
> +# Test xfs_copy to a mounted device
> +# If source is mounted, xfs_copy will print a warning, but still
> +# keep on copying. If target is mounted, xfs_copy will failed.
> +$XFS_COPY_PROG $SCRATCH_DEV $COPY_FILE >/dev/null
> +_scratch_mount
> +$XFS_COPY_PROG $COPY_FILE $SCRATCH_DEV 2>&1 | filter_mounted
> +[ ${PIPESTATUS[0]} -eq 0 ] && echo "Test xfs_copy to a mounted device [Failed]"
> +_scratch_unmount
> +
> +# Test xfs_db a mounted device without readonly (-r) option
> +_scratch_mount
> +$XFS_DB_PROG -c sb $SCRATCH_DEV 2>&1 | filter_mounted
> +[ ${PIPESTATUS[0]} -eq 0 ] &&
> +	echo "Test xfs_db a mounted device(without -r option) [Failed]"
> +_scratch_unmount
> +
> +# Test mkfs.xfs a mounted device
> +# Generally mkfs will report device is mounted, but if it can't find
> +# device is mounted, it'll report device busy.
> +_scratch_mount
> +_scratch_mkfs 2>&1 | filter_mounted
> +[ ${PIPESTATUS[0]} -eq 0 ] && echo "Test mkfs.xfs a mounted device [Failed]"
> +_scratch_unmount
> +
> +# Test xfs_repair (with/without modify flag) a mounted device
> +_scratch_mount
> +_scratch_xfs_repair -n 2>&1 | filter_mounted
> +[ ${PIPESTATUS[0]} -eq 0 ] &&
> +	echo "Test xfs_repair -n a mounted devie [Failed]"
> +_scratch_xfs_repair 2>&1 | filter_mounted
> +[ ${PIPESTATUS[0]} -eq 0 ] && echo "Test xfs_repair a mounted devie [Failed]"
> +_scratch_unmount
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/xfs/284.out b/tests/xfs/284.out
> new file mode 100644
> index 0000000..302d75e
> --- /dev/null
> +++ b/tests/xfs/284.out
> @@ -0,0 +1,8 @@
> +QA output created by 284
> +xfs_metadump: SCRATCH_DEV contains a mounted and writable filesystem
> +xfs_mdrestore: a filesystem is mounted on target device "SCRATCH_DEV", cannot restore to a mounted filesystem.
> +xfs_copy:  a filesystem is mounted on target device "SCRATCH_DEV".
> +xfs_db: SCRATCH_DEV contains a mounted filesystem
> +mkfs.xfs: SCRATCH_DEV contains a mounted filesystem
> +xfs_repair: SCRATCH_DEV contains a mounted and writable filesystem
> +xfs_repair: SCRATCH_DEV contains a mounted filesystem
> diff --git a/tests/xfs/group b/tests/xfs/group
> index baa0ff6..0ac4bd5 100644
> --- a/tests/xfs/group
> +++ b/tests/xfs/group
> @@ -281,6 +281,7 @@
>  281 dump ioctl auto quick
>  282 dump ioctl auto quick
>  283 dump ioctl auto quick
> +284 auto quick mount

I'm not sure if this belongs to 'mount' group, it does test against
mounted device, not actually test mount.

But perhaps we can add 'dump' 'copy' 'mkfs' 'db' 'repair' groups?

Thanks,
Eryu

>  287 auto dump quota quick
>  290 auto rw prealloc quick ioctl zero
>  291 auto repair
> -- 
> 2.7.4
> 
> --
> 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