xfstests: add test for btrfs send issuing premature rmdir operations
diff mbox

Message ID 1392820352-8550-1-git-send-email-fdmanana@gmail.com
State Not Applicable
Headers show

Commit Message

Filipe Manana Feb. 19, 2014, 2:32 p.m. UTC
Regression test for btrfs incremental send issue where a rmdir instruction
is sent against an orphan directory inode which is not empty yet, causing
btrfs receive to fail when it attempts to remove the directory.

This issue is fixed by the following linux kernel btrfs patch:

   Btrfs: fix send attempting to rmdir non-empty directories

Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
---
 tests/btrfs/041     |  153 +++++++++++++++++++++++++++++++++++++++++++++++++++
 tests/btrfs/041.out |    1 +
 tests/btrfs/group   |    1 +
 3 files changed, 155 insertions(+)
 create mode 100755 tests/btrfs/041
 create mode 100644 tests/btrfs/041.out

Comments

Dave Chinner Feb. 24, 2014, 5:25 a.m. UTC | #1
On Wed, Feb 19, 2014 at 02:32:32PM +0000, Filipe David Borba Manana wrote:
> Regression test for btrfs incremental send issue where a rmdir instruction
> is sent against an orphan directory inode which is not empty yet, causing
> btrfs receive to fail when it attempts to remove the directory.
> 
> This issue is fixed by the following linux kernel btrfs patch:
> 
>    Btrfs: fix send attempting to rmdir non-empty directories
> 
> Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
> ---
>  tests/btrfs/041     |  153 +++++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/btrfs/041.out |    1 +
>  tests/btrfs/group   |    1 +
>  3 files changed, 155 insertions(+)
>  create mode 100755 tests/btrfs/041
>  create mode 100644 tests/btrfs/041.out
> 
> diff --git a/tests/btrfs/041 b/tests/btrfs/041
> new file mode 100755
> index 0000000..9de9326
> --- /dev/null
> +++ b/tests/btrfs/041
> @@ -0,0 +1,153 @@
> +#! /bin/bash
> +# FS QA Test No. btrfs/041
> +#
> +# Regression test for btrfs incremental send issue where a rmdir instruction
> +# is sent against an orphan directory inode which is not empty yet, causing
> +# btrfs receive to fail when it attempts to remove the directory.
> +#
> +# This issue is fixed by the following linux kernel btrfs patch:
> +#
> +#   Btrfs: fix send attempting to rmdir non-empty directories
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2014 Filipe Manana.  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=`mktemp -d`
> +status=1	# failure is the default!
> +trap "_cleanup; exit \$status" 0 1 2 3 15
> +
> +_cleanup()
> +{
> +    rm -fr $tmp
> +}
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +
> +# real QA test starts here
> +_supported_fs btrfs
> +_supported_os Linux
> +_require_scratch
> +_need_to_be_root
> +
> +FSSUM_PROG=$here/src/fssum
> +[ -x $FSSUM_PROG ] || _notrun "fssum not built"

This is duplicated across several tests now. Perhaps this should be
factored now into a _requires_fssum helper (separate patch is
fine)?

> +
> +_scratch_unmount
> +_check_btrfs_filesystem $SCRATCH_DEV

you should be able to use _check_scratch_fs() here. I note that the
btrfs path does not unmount the scratch device, so you should update
it to do so (like _check_xfs_filesytem does) and then
_check_scratch_fs() will just Do The Right Thing.

> +_scratch_mkfs >/dev/null 2>&1
> +_scratch_mount
> +
> +_run_btrfs_util_prog receive $SCRATCH_MNT -f $tmp/1.snap
> +run_check $FSSUM_PROG -r $tmp/1.fssum $SCRATCH_MNT/mysnap1 2>> $seqres.full
> +
> +_run_btrfs_util_prog receive $SCRATCH_MNT -f $tmp/2.snap
> +run_check $FSSUM_PROG -r $tmp/2.fssum $SCRATCH_MNT/mysnap2 2>> $seqres.full

Hasn't run_check already redirected everything to $seqres.full?

> +_scratch_unmount
> +_check_btrfs_filesystem $SCRATCH_DEV

_check_scratch_fs() here too.

Cheers,

Dave.
Filipe Manana Feb. 24, 2014, 11:58 a.m. UTC | #2
On Mon, Feb 24, 2014 at 5:25 AM, Dave Chinner <david@fromorbit.com> wrote:
> On Wed, Feb 19, 2014 at 02:32:32PM +0000, Filipe David Borba Manana wrote:
>> Regression test for btrfs incremental send issue where a rmdir instruction
>> is sent against an orphan directory inode which is not empty yet, causing
>> btrfs receive to fail when it attempts to remove the directory.
>>
>> This issue is fixed by the following linux kernel btrfs patch:
>>
>>    Btrfs: fix send attempting to rmdir non-empty directories
>>
>> Signed-off-by: Filipe David Borba Manana <fdmanana@gmail.com>
>> ---
>>  tests/btrfs/041     |  153 +++++++++++++++++++++++++++++++++++++++++++++++++++
>>  tests/btrfs/041.out |    1 +
>>  tests/btrfs/group   |    1 +
>>  3 files changed, 155 insertions(+)
>>  create mode 100755 tests/btrfs/041
>>  create mode 100644 tests/btrfs/041.out
>>
>> diff --git a/tests/btrfs/041 b/tests/btrfs/041
>> new file mode 100755
>> index 0000000..9de9326
>> --- /dev/null
>> +++ b/tests/btrfs/041
>> @@ -0,0 +1,153 @@
>> +#! /bin/bash
>> +# FS QA Test No. btrfs/041
>> +#
>> +# Regression test for btrfs incremental send issue where a rmdir instruction
>> +# is sent against an orphan directory inode which is not empty yet, causing
>> +# btrfs receive to fail when it attempts to remove the directory.
>> +#
>> +# This issue is fixed by the following linux kernel btrfs patch:
>> +#
>> +#   Btrfs: fix send attempting to rmdir non-empty directories
>> +#
>> +#-----------------------------------------------------------------------
>> +# Copyright (c) 2014 Filipe Manana.  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=`mktemp -d`
>> +status=1     # failure is the default!
>> +trap "_cleanup; exit \$status" 0 1 2 3 15
>> +
>> +_cleanup()
>> +{
>> +    rm -fr $tmp
>> +}
>> +
>> +# get standard environment, filters and checks
>> +. ./common/rc
>> +. ./common/filter
>> +
>> +# real QA test starts here
>> +_supported_fs btrfs
>> +_supported_os Linux
>> +_require_scratch
>> +_need_to_be_root
>> +
>> +FSSUM_PROG=$here/src/fssum
>> +[ -x $FSSUM_PROG ] || _notrun "fssum not built"
>
> This is duplicated across several tests now. Perhaps this should be
> factored now into a _requires_fssum helper (separate patch is
> fine)?
>
>> +
>> +_scratch_unmount
>> +_check_btrfs_filesystem $SCRATCH_DEV
>
> you should be able to use _check_scratch_fs() here. I note that the
> btrfs path does not unmount the scratch device, so you should update
> it to do so (like _check_xfs_filesytem does) and then
> _check_scratch_fs() will just Do The Right Thing.
>
>> +_scratch_mkfs >/dev/null 2>&1
>> +_scratch_mount
>> +
>> +_run_btrfs_util_prog receive $SCRATCH_MNT -f $tmp/1.snap
>> +run_check $FSSUM_PROG -r $tmp/1.fssum $SCRATCH_MNT/mysnap1 2>> $seqres.full
>> +
>> +_run_btrfs_util_prog receive $SCRATCH_MNT -f $tmp/2.snap
>> +run_check $FSSUM_PROG -r $tmp/2.fssum $SCRATCH_MNT/mysnap2 2>> $seqres.full
>
> Hasn't run_check already redirected everything to $seqres.full?
>
>> +_scratch_unmount
>> +_check_btrfs_filesystem $SCRATCH_DEV
>
> _check_scratch_fs() here too.
>
> Cheers,

All done in V2 (same for the other test).
For the require_fssum() function, I made a patch on top of these 2 so
that it updates all existing btrfs tests that need fssum ("[PATCH]
xfstests: add function _require_fssum()").

Thanks for the good suggestions Dave.

>
> Dave.
> --
> Dave Chinner
> david@fromorbit.com

Patch
diff mbox

diff --git a/tests/btrfs/041 b/tests/btrfs/041
new file mode 100755
index 0000000..9de9326
--- /dev/null
+++ b/tests/btrfs/041
@@ -0,0 +1,153 @@ 
+#! /bin/bash
+# FS QA Test No. btrfs/041
+#
+# Regression test for btrfs incremental send issue where a rmdir instruction
+# is sent against an orphan directory inode which is not empty yet, causing
+# btrfs receive to fail when it attempts to remove the directory.
+#
+# This issue is fixed by the following linux kernel btrfs patch:
+#
+#   Btrfs: fix send attempting to rmdir non-empty directories
+#
+#-----------------------------------------------------------------------
+# Copyright (c) 2014 Filipe Manana.  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=`mktemp -d`
+status=1	# failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+_cleanup()
+{
+    rm -fr $tmp
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# real QA test starts here
+_supported_fs btrfs
+_supported_os Linux
+_require_scratch
+_need_to_be_root
+
+FSSUM_PROG=$here/src/fssum
+[ -x $FSSUM_PROG ] || _notrun "fssum not built"
+
+rm -f $seqres.full
+
+_scratch_mkfs >/dev/null 2>&1
+_scratch_mount
+
+mkdir -p $SCRATCH_MNT/a/b
+mkdir $SCRATCH_MNT/0
+mkdir $SCRATCH_MNT/1
+mkdir $SCRATCH_MNT/a/b/c
+mv $SCRATCH_MNT/0 $SCRATCH_MNT/a/b/c
+mv $SCRATCH_MNT/1 $SCRATCH_MNT/a/b/c
+echo 'ola mundo' > $SCRATCH_MNT/a/b/c/foo.txt
+mkdir $SCRATCH_MNT/a/b/c/x
+mkdir $SCRATCH_MNT/a/b/c/x2
+mkdir $SCRATCH_MNT/a/b/y
+mkdir $SCRATCH_MNT/a/b/z
+mkdir -p $SCRATCH_MNT/a/b/d1/d2/d3
+mkdir $SCRATCH_MNT/a/b/d4
+
+# Filesystem looks like:
+#
+# .                        (ino 256)
+# |-- a/                   (ino 257)
+#     |-- b/               (ino 258)
+#         |-- c/           (ino 261)
+#         |   |-- foo.txt  (ino 262)
+#         |   |-- 0/       (ino 259)
+#         |   |-- 1/       (ino 260)
+#         |   |-- x/       (ino 263)
+#         |   |-- x2/      (ino 264)
+#         |
+#         |-- y/           (ino 265)
+#         |-- z/           (ino 266)
+#         |-- d1/          (ino 267)
+#         |   |-- d2/      (ino 268)
+#         |       |-- d3/  (ino 269)
+#         |
+#         |-- d4/          (ino 270)
+
+_run_btrfs_util_prog subvolume snapshot -r $SCRATCH_MNT $SCRATCH_MNT/mysnap1
+
+rm -f $SCRATCH_MNT/a/b/c/foo.txt
+mv $SCRATCH_MNT/a/b/y $SCRATCH_MNT/a/b/YY
+mv $SCRATCH_MNT/a/b/z $SCRATCH_MNT/a
+mv $SCRATCH_MNT/a/b/c/x $SCRATCH_MNT/a/b/YY
+mv $SCRATCH_MNT/a/b/c/0 $SCRATCH_MNT/a/b/YY/00
+mv $SCRATCH_MNT/a/b/c/x2 $SCRATCH_MNT/a/z/X_2
+mv $SCRATCH_MNT/a/b/c/1 $SCRATCH_MNT/a/z/X_2
+rmdir $SCRATCH_MNT/a/b/c
+mv $SCRATCH_MNT/a/b/d4 $SCRATCH_MNT/a/d44
+mv $SCRATCH_MNT/a/b/d1/d2 $SCRATCH_MNT/a/d44
+rmdir $SCRATCH_MNT/a/b/d1
+
+# Filesystem now looks like:
+#
+# .                        (ino 256)
+# |-- a/                   (ino 257)
+#     |-- b/               (ino 258)
+#     |   |-- YY/          (ino 265)
+#     |        |-- x/      (ino 263)
+#     |        |-- 00/     (ino 259)
+#     |
+#     |-- z/               (ino 266)
+#     |   |-- X_2/         (ino 264)
+#     |        |-- 1/      (ino 260)
+#     |
+#     |-- d44/             (ino 270)
+#          |-- d2/         (ino 268)
+#              |-- d3/     (ino 269)
+
+_run_btrfs_util_prog subvolume snapshot -r $SCRATCH_MNT $SCRATCH_MNT/mysnap2
+
+run_check $FSSUM_PROG -A -f -w $tmp/1.fssum $SCRATCH_MNT/mysnap1
+run_check $FSSUM_PROG -A -f -w $tmp/2.fssum -x $SCRATCH_MNT/mysnap2/mysnap1 \
+	$SCRATCH_MNT/mysnap2
+
+_run_btrfs_util_prog send $SCRATCH_MNT/mysnap1 -f $tmp/1.snap
+_run_btrfs_util_prog send -p $SCRATCH_MNT/mysnap1 $SCRATCH_MNT/mysnap2 \
+	-f $tmp/2.snap
+
+_scratch_unmount
+_check_btrfs_filesystem $SCRATCH_DEV
+
+_scratch_mkfs >/dev/null 2>&1
+_scratch_mount
+
+_run_btrfs_util_prog receive $SCRATCH_MNT -f $tmp/1.snap
+run_check $FSSUM_PROG -r $tmp/1.fssum $SCRATCH_MNT/mysnap1 2>> $seqres.full
+
+_run_btrfs_util_prog receive $SCRATCH_MNT -f $tmp/2.snap
+run_check $FSSUM_PROG -r $tmp/2.fssum $SCRATCH_MNT/mysnap2 2>> $seqres.full
+
+_scratch_unmount
+_check_btrfs_filesystem $SCRATCH_DEV
+
+status=0
+exit
diff --git a/tests/btrfs/041.out b/tests/btrfs/041.out
new file mode 100644
index 0000000..309876c
--- /dev/null
+++ b/tests/btrfs/041.out
@@ -0,0 +1 @@ 
+QA output created by 041
diff --git a/tests/btrfs/group b/tests/btrfs/group
index f5452b5..fd0193c 100644
--- a/tests/btrfs/group
+++ b/tests/btrfs/group
@@ -43,3 +43,4 @@ 
 038 auto quick
 039 auto quick
 040 auto quick
+041 auto quick