[PATCH] xfstests 278: Ensure lost+found is not created with wrong link count (V4)

[PATCH] xfstests 278: Ensure lost+found is not created with wrong link count (V4)

[Carlos Maiolino]

xfs_repair was leaving lost+found directory with a wrong link count when a
cleaned inode was re-used to create lost+found. This test case confirm that,
after xfs_repair is executed, the lost+found inode is left in a consistent
state.

Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
---
 278     |   87 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
 278.out |    2 +
 group   |    1 +
 3 files changed, 90 insertions(+), 0 deletions(-)
 create mode 100755 278
 create mode 100644 278.out

diff --git a/278 b/278
new file mode 100755
index 0000000..89cae35
--- /dev/null
+++ b/278
@@ -0,0 +1,87 @@
+#! /bin/bash
+# FS QA Test No. 278
+#
+# Test xfs_repair to ensure it fixes the lost+found link count
+# at the first run. See also commit 198b747f255346bca64408875763b6ca0ed3d57d
+# from xfsprogs tree.
+#
+#
+#-----------------------------------------------------------------------
+# Copyright (c) 2011 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
+#-----------------------------------------------------------------------
+#
+# creator
+owner=cmaiolino@redhat.com
+
+seq=`basename $0`
+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.*
+}
+
+# get standard environment, filters and checks
+. ./common.rc
+. ./common.filter
+
+# real QA test starts here
+_supported_fs xfs
+_supported_os Linux
+_require_scratch
+_scratch_mkfs >/dev/null 2>&1
+_scratch_mount
+
+mkdir -p $SCRATCH_MNT/dir/subdir
+
+#get inodes to be changed
+DIR_INO=`ls -i $SCRATCH_MNT |awk '{print $1}'`
+SUBDIR_INO=`ls -i $SCRATCH_MNT/dir |awk '{print $1}'`
+
+_scratch_unmount
+
+echo "Silence is goodness..."
+
+# Corrupt DIR
+xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].inumber.i4 0" $SCRATCH_DEV >> $seq.full
+xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].name 0" $SCRATCH_DEV >> $seq.full
+xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].offset 0" $SCRATCH_DEV >> $seq.full
+xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].namelen 0" $SCRATCH_DEV >> $seq.full
+xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.hdr.parent.i4 0" $SCRATCH_DEV >> $seq.full
+xfs_db -x -c "inode $DIR_INO" -c "write core.nlinkv2 0" $SCRATCH_DEV >> $seq.full
+
+# Corrupt SUBDIR
+xfs_db -x -c "inode $SUBDIR_INO" -c "write u.sfdir2.hdr.parent.i4 0" $SCRATCH_DEV >> $seq.full
+xfs_db -x -c "inode $SUBDIR_INO" -c "write core.nlinkv2 0" $SCRATCH_DEV >> $seq.full
+
+echo "===== BEGIN of xfs_repair =====" >> $seq.full
+echo "" >>$seq.full
+
+xfs_repair  $SCRATCH_DEV >> $seq.full 2>&1
+echo "===== END of xfs_repair =====" >> $seq.full
+
+if _check_scratch_fs; then
+	status=0
+else
+	status=1
+fi
+exit
diff --git a/278.out b/278.out
new file mode 100644
index 0000000..2acd770
--- /dev/null
+++ b/278.out
@@ -0,0 +1,2 @@
+QA output created by 278
+Silence is goodness...
diff --git a/group b/group
index 99592d3..6f2d748 100644
--- a/group
+++ b/group
@@ -391,3 +391,4 @@ deprecated
 275 auto rw
 276 auto rw metadata
 277 auto ioctl quick metadata
+278 repair auto
-- 
1.7.6.4

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: [PATCH] xfstests 278: Ensure lost+found is not created with wrong link count (V4)

[Carlos Maiolino]

Mark, can you test it and see if the statement I used to send repair
output to $seq.full will work on bash3? I haven't any system here using
bash 3.

Thanks
On Thu, Jan 12, 2012 at 12:06:27PM -0200, Carlos Maiolino wrote:
> xfs_repair was leaving lost+found directory with a wrong link count when a
> cleaned inode was re-used to create lost+found. This test case confirm that,
> after xfs_repair is executed, the lost+found inode is left in a consistent
> state.
> 
> Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>
> ---
>  278     |   87 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>  278.out |    2 +
>  group   |    1 +
>  3 files changed, 90 insertions(+), 0 deletions(-)
>  create mode 100755 278
>  create mode 100644 278.out
> 
> diff --git a/278 b/278
> new file mode 100755
> index 0000000..89cae35
> --- /dev/null
> +++ b/278
> @@ -0,0 +1,87 @@
> +#! /bin/bash
> +# FS QA Test No. 278
> +#
> +# Test xfs_repair to ensure it fixes the lost+found link count
> +# at the first run. See also commit 198b747f255346bca64408875763b6ca0ed3d57d
> +# from xfsprogs tree.
> +#
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2011 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
> +#-----------------------------------------------------------------------
> +#
> +# creator
> +owner=cmaiolino@redhat.com
> +
> +seq=`basename $0`
> +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.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common.rc
> +. ./common.filter
> +
> +# real QA test starts here
> +_supported_fs xfs
> +_supported_os Linux
> +_require_scratch
> +_scratch_mkfs >/dev/null 2>&1
> +_scratch_mount
> +
> +mkdir -p $SCRATCH_MNT/dir/subdir
> +
> +#get inodes to be changed
> +DIR_INO=`ls -i $SCRATCH_MNT |awk '{print $1}'`
> +SUBDIR_INO=`ls -i $SCRATCH_MNT/dir |awk '{print $1}'`
> +
> +_scratch_unmount
> +
> +echo "Silence is goodness..."
> +
> +# Corrupt DIR
> +xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].inumber.i4 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].name 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].offset 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].namelen 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.hdr.parent.i4 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $DIR_INO" -c "write core.nlinkv2 0" $SCRATCH_DEV >> $seq.full
> +
> +# Corrupt SUBDIR
> +xfs_db -x -c "inode $SUBDIR_INO" -c "write u.sfdir2.hdr.parent.i4 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $SUBDIR_INO" -c "write core.nlinkv2 0" $SCRATCH_DEV >> $seq.full
> +
> +echo "===== BEGIN of xfs_repair =====" >> $seq.full
> +echo "" >>$seq.full
> +
> +xfs_repair  $SCRATCH_DEV >> $seq.full 2>&1
> +echo "===== END of xfs_repair =====" >> $seq.full
> +
> +if _check_scratch_fs; then
> +	status=0
> +else
> +	status=1
> +fi
> +exit
> diff --git a/278.out b/278.out
> new file mode 100644
> index 0000000..2acd770
> --- /dev/null
> +++ b/278.out
> @@ -0,0 +1,2 @@
> +QA output created by 278
> +Silence is goodness...
> diff --git a/group b/group
> index 99592d3..6f2d748 100644
> --- a/group
> +++ b/group
> @@ -391,3 +391,4 @@ deprecated
>  275 auto rw
>  276 auto rw metadata
>  277 auto ioctl quick metadata
> +278 repair auto
> -- 
> 1.7.6.4
> 
> _______________________________________________
> xfs mailing list
> xfs@oss.sgi.com
> http://oss.sgi.com/mailman/listinfo/xfs

-- 
--Carlos

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: [PATCH] xfstests 278: Ensure lost+found is not created with wrong link count (V4)

[Mark Tinguely]

On 01/12/12 08:45, Carlos Maiolino wrote:
> Mark, can you test it and see if the statement I used to send repair
> output to $seq.full will work on bash3? I haven't any system here using
> bash 3.
>
> Thanks

Works under bash3 and bash4. Thank-you,

Reviewed-by: Mark Tinguely <tinguely@sgi.com>

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs

Re: [PATCH] xfstests 278: Ensure lost+found is not created with wrong link count (V4)

[Eric Sandeen]

On 1/12/12 8:06 AM, Carlos Maiolino wrote:
> xfs_repair was leaving lost+found directory with a wrong link count when a
> cleaned inode was re-used to create lost+found. This test case confirm that,
> after xfs_repair is executed, the lost+found inode is left in a consistent
> state.
> 
> Signed-off-by: Carlos Maiolino <cmaiolino@redhat.com>

A couple minor comments below just to consider next time, but I think it's ok.
I'll merge it to xfstests-dev with my signoff and Mark's review.

-Eric

> ---
>  278     |   87 +++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
>  278.out |    2 +
>  group   |    1 +
>  3 files changed, 90 insertions(+), 0 deletions(-)
>  create mode 100755 278
>  create mode 100644 278.out
> 
> diff --git a/278 b/278
> new file mode 100755
> index 0000000..89cae35
> --- /dev/null
> +++ b/278
> @@ -0,0 +1,87 @@
> +#! /bin/bash
> +# FS QA Test No. 278
> +#
> +# Test xfs_repair to ensure it fixes the lost+found link count
> +# at the first run. See also commit 198b747f255346bca64408875763b6ca0ed3d57d
> +# from xfsprogs tree.
> +#
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2011 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
> +#-----------------------------------------------------------------------
> +#
> +# creator
> +owner=cmaiolino@redhat.com
> +
> +seq=`basename $0`
> +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.*
> +}
> +
> +# get standard environment, filters and checks
> +. ./common.rc
> +. ./common.filter
> +
> +# real QA test starts here
> +_supported_fs xfs
> +_supported_os Linux
> +_require_scratch
> +_scratch_mkfs >/dev/null 2>&1

I suppose I'd prefer this go to $seq.full since we have it,
but no big deal, lots of tests don't.

> +_scratch_mount
> +
> +mkdir -p $SCRATCH_MNT/dir/subdir
> +
> +#get inodes to be changed
> +DIR_INO=`ls -i $SCRATCH_MNT |awk '{print $1}'`
> +SUBDIR_INO=`ls -i $SCRATCH_MNT/dir |awk '{print $1}'`

I wonder if these should be more explicit; how about:

+DIR_INO=`ls -id $SCRATCH_MNT/dir |awk '{print $1}'`
+SUBDIR_INO=`ls -id $SCRATCH_MNT/dir/subdir |awk '{print $1}'`

it's probably fine as is.

> +_scratch_unmount
> +
> +echo "Silence is goodness..."
> +
> +# Corrupt DIR
> +xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].inumber.i4 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].name 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].offset 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.list[0].namelen 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $DIR_INO" -c "write u.sfdir2.hdr.parent.i4 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $DIR_INO" -c "write core.nlinkv2 0" $SCRATCH_DEV >> $seq.full
> +
> +# Corrupt SUBDIR
> +xfs_db -x -c "inode $SUBDIR_INO" -c "write u.sfdir2.hdr.parent.i4 0" $SCRATCH_DEV >> $seq.full
> +xfs_db -x -c "inode $SUBDIR_INO" -c "write core.nlinkv2 0" $SCRATCH_DEV >> $seq.full
> +
> +echo "===== BEGIN of xfs_repair =====" >> $seq.full
> +echo "" >>$seq.full
> +
> +xfs_repair  $SCRATCH_DEV >> $seq.full 2>&1
> +echo "===== END of xfs_repair =====" >> $seq.full
> +
> +if _check_scratch_fs; then
> +	status=0
> +else
> +	status=1
> +fi
> +exit
> diff --git a/278.out b/278.out
> new file mode 100644
> index 0000000..2acd770
> --- /dev/null
> +++ b/278.out
> @@ -0,0 +1,2 @@
> +QA output created by 278
> +Silence is goodness...
> diff --git a/group b/group
> index 99592d3..6f2d748 100644
> --- a/group
> +++ b/group
> @@ -391,3 +391,4 @@ deprecated
>  275 auto rw
>  276 auto rw metadata
>  277 auto ioctl quick metadata
> +278 repair auto

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs