* [PATCH V4] xfstests: don't remove the two first devices from SCRATCH_DEV_POOL
@ 2013-08-23 13:07 ` Stefan Behrens
0 siblings, 0 replies; 24+ messages in thread
From: Stefan Behrens @ 2013-08-23 13:07 UTC (permalink / raw)
To: xfs; +Cc: linux-btrfs
Since common/config is executed twice, if SCRATCH_DEV_POOL is configured
via the environment, the current code removes the first device entry twice
which means that you lose the second device for the test.
The fix is to not remove anything from SCRATCH_DEV_POOL anymore.
That used to be done (I can only guess) to allow to pass the
SCRATCH_DEV_POOL as an argument to _scratch_mkfs. Since _scratch_mkfs adds
the SCRATCH_DEV, the pool mustn't contain that device anymore.
A new function _scratch_pool_mkfs is introduced that does the expected
thing.
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
---
V1 -> V2:
- rebased.
- fixed a missing adaption to the new way to deal with SCRATCH_DEV_POOL
in _test_replace() in tests/btrfs/003.
V2 -> V3:
- V2 was the result of a git rebase mistake (a user error), forget it,
sorry. V3 now contains the correct fix for a missing adaption to the
new way to deal with SCRATCH_DEV_POOL in _test_replace() in
tests/btrfs/003.
V3 -> V4:
- rebased, because it didn't apply cleanly anymore today.
common/config | 1 -
common/rc | 12 ++++++++++++
tests/btrfs/002 | 3 ++-
tests/btrfs/003 | 16 ++++++++--------
tests/btrfs/006 | 4 ++--
5 files changed, 24 insertions(+), 12 deletions(-)
diff --git a/common/config b/common/config
index 39dd469..586870b 100644
--- a/common/config
+++ b/common/config
@@ -267,7 +267,6 @@ get_next_config() {
exit 1
fi
SCRATCH_DEV=`echo $SCRATCH_DEV_POOL | awk '{print $1}'`
- SCRATCH_DEV_POOL=`echo $SCRATCH_DEV_POOL | awk '{ ORS=" "; for (i = 2; i <= NF; i++) print $i}'`
fi
echo $SCRATCH_DEV | grep -q ":" > /dev/null 2>&1
diff --git a/common/rc b/common/rc
index ae80b12..77e96c4 100644
--- a/common/rc
+++ b/common/rc
@@ -550,6 +550,18 @@ _scratch_mkfs()
esac
}
+_scratch_pool_mkfs()
+{
+ case $FSTYP in
+ btrfs)
+ $MKFS_BTRFS_PROG $MKFS_OPTIONS $* $SCRATCH_DEV_POOL > /dev/null
+ ;;
+ *)
+ echo "_scratch_pool_mkfs is not implemented for $FSTYP" 1>&2
+ ;;
+ esac
+}
+
# Create fs of certain size on scratch device
# _scratch_mkfs_sized <size in bytes> [optional blocksize]
_scratch_mkfs_sized()
diff --git a/tests/btrfs/002 b/tests/btrfs/002
index 03e9137..f4389ae 100755
--- a/tests/btrfs/002
+++ b/tests/btrfs/002
@@ -45,8 +45,9 @@ _need_to_be_root
_supported_fs btrfs
_supported_os Linux
_require_scratch
+_require_scratch_dev_pool
-_scratch_mkfs $SCRATCH_DEV_POOL > /dev/null 2>&1 || _fail "mkfs failed"
+_scratch_pool_mkfs > /dev/null 2>&1 || _fail "mkfs failed"
_scratch_mount
# Create and save sha256sum
diff --git a/tests/btrfs/003 b/tests/btrfs/003
index 5c88651..262b1d5 100755
--- a/tests/btrfs/003
+++ b/tests/btrfs/003
@@ -58,7 +58,7 @@ rm -f $seqres.full
_test_raid0()
{
export MKFS_OPTIONS="-m raid0 -d raid0"
- _scratch_mkfs $SCRATCH_DEV_POOL >> $seqres.full 2>&1 || _fail "mkfs failed"
+ _scratch_pool_mkfs >> $seqres.full 2>&1 || _fail "mkfs failed"
_scratch_mount
dirp=`mktemp -duq $SCRATCH_MNT/dir.XXXXXX`
_populate_fs -n 1 -f 20 -d 10 -r $dirp -s 10
@@ -68,7 +68,7 @@ _test_raid0()
_test_raid1()
{
export MKFS_OPTIONS="-m raid1 -d raid1"
- _scratch_mkfs $SCRATCH_DEV_POOL >> $seqres.full 2>&1 || _fail "mkfs failed"
+ _scratch_pool_mkfs >> $seqres.full 2>&1 || _fail "mkfs failed"
_scratch_mount
dirp=`mktemp -duq $SCRATCH_MNT/dir.XXXXXX`
_populate_fs -n 1 -f 20 -d 10 -r $dirp -s 10
@@ -78,7 +78,7 @@ _test_raid1()
_test_raid10()
{
export MKFS_OPTIONS="-m raid10 -d raid10"
- _scratch_mkfs $SCRATCH_DEV_POOL >> $seqres.full 2>&1 || _fail "mkfs failed"
+ _scratch_pool_mkfs >> $seqres.full 2>&1 || _fail "mkfs failed"
_scratch_mount
dirp=`mktemp -duq $SCRATCH_MNT/dir.XXXXXX`
_populate_fs -n 1 -f 20 -d 10 -r $dirp -s 10
@@ -88,7 +88,7 @@ _test_raid10()
_test_single()
{
export MKFS_OPTIONS="-m single -d single"
- _scratch_mkfs $SCRATCH_DEV_POOL >> $seqres.full 2>&1 || _fail "mkfs failed"
+ _scratch_pool_mkfs >> $seqres.full 2>&1 || _fail "mkfs failed"
_scratch_mount
dirp=`mktemp -duq $SCRATCH_MNT/dir.XXXXXX`
_populate_fs -n 1 -f 20 -d 10 -r $dirp -s 10
@@ -108,7 +108,7 @@ _test_add()
_scratch_mount
dirp=`mktemp -duq $SCRATCH_MNT/dir.XXXXXX`
_populate_fs -n 1 -f 20 -d 10 -r $dirp -s 10
- for i in `seq 1 $n`; do
+ for i in `seq 2 $n`; do
$BTRFS_UTIL_PROG device add ${devs[$i]} $SCRATCH_MNT >> $seqres.full 2>&1 || _fail "device add failed"
done
$BTRFS_UTIL_PROG filesystem balance $SCRATCH_MNT >> $seqres.full 2>&1 || _fail "balance failed"
@@ -124,9 +124,9 @@ _test_replace()
local d
local DEVHTL=""
- # exclude the last disk in the disk pool
+ # exclude the first and the last disk in the disk pool
n=$(($n-1))
- ds=${devs[@]:0:$n}
+ ds=${devs[@]:1:$(($n-1))}
export MKFS_OPTIONS="-m raid1 -d raid1"
_scratch_mkfs "$ds" >> $seqres.full 2>&1 || _fail "tr: mkfs failed"
@@ -164,7 +164,7 @@ _test_replace()
_test_remove()
{
- _scratch_mkfs "$SCRATCH_DEV_POOL" >> $seqres.full 2>&1 || _fail "mkfs failed"
+ _scratch_pool_mkfs >> $seqres.full 2>&1 || _fail "mkfs failed"
_scratch_mount
dirp=`mktemp -duq $SCRATCH_MNT/dir.XXXXXX`
_populate_fs -n 1 -f 20 -d 10 -r $dirp -s 10
diff --git a/tests/btrfs/006 b/tests/btrfs/006
index 9f7beff..715fd80 100755
--- a/tests/btrfs/006
+++ b/tests/btrfs/006
@@ -53,12 +53,12 @@ rm -f $seqres.full
FIRST_POOL_DEV=`echo $SCRATCH_DEV_POOL | awk '{print $1}'`
LAST_POOL_DEV=`echo $SCRATCH_DEV_POOL | awk '{print $NF}'`
-TOTAL_DEVS=`echo $SCRATCH_DEV $SCRATCH_DEV_POOL | wc -w`
+TOTAL_DEVS=`echo $SCRATCH_DEV_POOL | wc -w`
LABEL=TestLabel.$seq
echo "Scratch $SCRATCH_DEV First $FIRST_POOL_DEV last $LAST_POOL_DEV Total $TOTAL_DEVS" > $seqres.full
-_scratch_mkfs $SCRATCH_DEV_POOL >> $seqres.full 2>&1 || _fail "mkfs failed"
+_scratch_pool_mkfs >> $seqres.full 2>&1 || _fail "mkfs failed"
# These have to be done unmounted...?
echo "== Set filesystem label to $LABEL"
--
1.8.3.4
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply related [flat|nested] 24+ messages in thread* [PATCH][RESEND] xfstests: add a test for btrfs device replace operation
2013-08-23 13:07 ` Stefan Behrens
@ 2013-08-23 13:07 ` Stefan Behrens
-1 siblings, 0 replies; 24+ messages in thread
From: Stefan Behrens @ 2013-08-23 13:07 UTC (permalink / raw)
To: xfs; +Cc: linux-btrfs
This test performs btrfs device replace tests with all possible profiles
(single/dup/mixed/raid0/raid1/raid10), one round with the '-r' option
to 'btrfs replace start' and one round without this option. The
cancelation is tested only once and with the dup/single profile for
metadata/data.
This test takes 181 seconds on my SSD equiped test box and 237s on
spinning disks. Almost all the time is spent when the filesystem is
populated with test data. The replace operation itself takes less than
a second for all the tests, except for the test that is marked as
'thorough' which will run for about 8 seconds on my test box.
The amount of tests done depends on the number of devices in the
SCRATCH_DEV_POOL. For full test coverage, at least 5 devices should
be available (e.g. 5 partitions). With less than 2 entries in
SCRATCH_DEV_POOL, the test is not executed.
The source and target devices for the replace operation are arbitrarily
chosen out of SCRATCH_DEV_POOl. Since the target device mustn't be
smaller than the source device, the requirement for this test is that
all devices have _exactly_ the same size. If this is not the case, the
test terminates with _notrun.
To check the filesystems after replacing a device, a scrub run is
performed, a btrfsck run, and finally the filesystem is remounted.
This commit depends on my other commit:
"xfstest: don't remove the two first devices from SCRATCH_DEV_POOL"
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
---
V1 -> V2:
Major reworking in order to address all the comments from Eric Sandeen's
review.
V2 -> V3:
- Rebased.
- The check that the devices have the same size was missing the target
device.
V3 -> V4:
Somehow I managed to add code that used a hardcoded path instead of
$SCRATCH_MNT. Fixed this issue.
common/config | 1 +
tests/btrfs/010 | 277 ++++++++++++++++++++++++++++++++++++++++++++++++++++
tests/btrfs/010.out | 3 +
tests/btrfs/group | 1 +
4 files changed, 282 insertions(+)
diff --git a/common/config b/common/config
index 586870b..db086fb 100644
--- a/common/config
+++ b/common/config
@@ -207,6 +207,7 @@ case "$HOSTOS" in
export MKFS_UDF_PROG="`set_prog_path mkudffs`"
export MKFS_BTRFS_PROG="`set_btrfs_mkfs_prog_path_with_opts`"
export BTRFS_UTIL_PROG="`set_prog_path btrfs`"
+ export BTRFS_SHOW_SUPER_PROG="`set_prog_path btrfs-show-super`"
export XFS_FSR_PROG="`set_prog_path xfs_fsr`"
export MKFS_NFS_PROG="false"
;;
diff --git a/tests/btrfs/010 b/tests/btrfs/010
new file mode 100755
index 0000000..36800ff
--- /dev/null
+++ b/tests/btrfs/010
@@ -0,0 +1,277 @@
+#! /bin/bash
+# FSQA Test No. btrfs/010
+#
+# Test of the btrfs replace operation.
+#
+# The amount of tests done depends on the number of devices in the
+# SCRATCH_DEV_POOL. For full test coverage, at least 5 devices should
+# be available (e.g. 5 partitions).
+#
+# The source and target devices for the replace operation are
+# arbitrarily chosen out of SCRATCH_DEV_POOl. Since the target device
+# mustn't be smaller than the source device, the requirement for this
+# test is that all devices have _exactly_ the same size. If this is
+# not the case, this test is not run.
+#
+# To check the filesystems after replacing a device, a scrub run is
+# performed, a btrfsck run, and finally the filesystem is remounted.
+#
+#-----------------------------------------------------------------------
+# Copyright (C) 2013 STRATO. 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
+noise_pid=0
+
+_cleanup()
+{
+ if [ $noise_pid -ne 0 ] && ps -p $noise_pid | grep -q $noise_pid; then
+ kill -TERM $noise_pid
+ fi
+ wait
+ rm -f $tmp.tmp
+}
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# real QA test starts here
+_need_to_be_root
+_supported_fs btrfs
+_require_scratch
+_require_scratch_dev_pool
+_require_command $BTRFS_SHOW_SUPER_PROG btrfs-show-super
+
+rm -f $seqres.full
+rm -f $tmp.tmp
+
+echo "*** test btrfs replace"
+
+workout()
+{
+ local mkfs_options="$1"
+ local num_devs4raid="$2"
+ local with_cancel="$3"
+ local quick="$4"
+ local source_dev="`echo ${SCRATCH_DEV_POOL} | awk '{print $1}'`"
+ local target_dev="`echo ${SCRATCH_DEV_POOL} | awk '{print $NF}'`"
+
+ if [ "`echo $SCRATCH_DEV_POOL | wc -w`" -lt `expr $num_devs4raid + 1` ]; then
+ echo "Skip workout $1 $2 $3 $4" >> $seqres.full
+ echo "Too few devices in SCRATCH_DEV_POOL $SCRATCH_DEV_POOL, required: " `expr $num_devs4raid + 1` >> $seqres.full
+ return 0
+ fi
+
+ # use min number of disks in order to fill up the disk to replace
+ # as much as possible
+ local used_devs_without_1st="`echo $SCRATCH_DEV_POOL | \
+ awk '{ORS=\" \"; for (i = 2; i <= (NF - 1 < '$num_devs4raid' ? NF - 1 : '$num_devs4raid'); i++) print $i}'`"
+
+ # _scratch_mkfs adds the 1st device again (which is $SCRATCH_DEV)
+ _scratch_mkfs $mkfs_options $used_devs_without_1st >> $seqres.full 2>&1 || _fail "mkfs failed"
+
+ # create a filesystem on the target device just for the sake of
+ # being able to query its size with btrfs-show-super
+ $MKFS_BTRFS_PROG $MKFS_OPTIONS $target_dev >> $seqres.full 2>&1 || _fail "mkfs target_dev failed"
+
+ # The source and target devices for the replace operation are
+ # arbitrarily chosen out of the pool. Since the target device mustn't
+ # be smaller than the source device, the requirement for this test is
+ # that all devices have _exactly_ the same size. If this is not the
+ # case, this test is not run.
+ local num_lines=`$BTRFS_SHOW_SUPER_PROG $SCRATCH_DEV $used_devs_without_1st $target_dev | grep dev_item.total_bytes | uniq | wc -l`
+ if [ $num_lines -gt 1 ]; then
+ _notrun "Different device sizes detected"
+ fi
+
+ _scratch_mount
+
+ # Generate 500 times 20K extents in the data chunk and fill up
+ # metadata with inline extents. Ignore ENOSPC.
+ for i in `seq 1 500`; do
+ dd if=/dev/urandom of=$SCRATCH_MNT/l$i bs=16385 count=1
+ dd if=/dev/urandom of=$SCRATCH_MNT/s$i bs=3800 count=1
+ done > /dev/null 2>&1
+
+ if [ "${quick}Q" = "thoroughQ" ]; then
+ # The intention of this "thorough" test is to increase
+ # the probability of random errors, in particular in
+ # conjunction with the background noise generator and
+ # a sync call while the replace operation in ongoing.
+ # Unfortunately it takes quite some time to generate
+ # the test filesystem, therefore most data consists out
+ # of zeros although this data is not very useful for
+ # detecting misplaced read/write requests.
+ # Ignore ENOSPC, it's not a problem..
+ dd if=/dev/urandom of=$SCRATCH_MNT/r bs=1M count=200 >> $seqres.full 2>&1 &
+ dd if=/dev/zero of=$SCRATCH_MNT/0 bs=1M count=2000 >> $seqres.full 2>&1
+ wait
+ elif [ "${with_cancel}Q" = "cancelQ" ]; then
+ # produce some data to prevent that the replace operation
+ # finishes before the cancel request is started
+ dd if=/dev/zero of=$SCRATCH_MNT/0 bs=1M count=1000 >> $seqres.full 2>&1
+ fi
+ sync; sync
+
+ btrfs_replace_test $source_dev $target_dev "" $with_cancel $quick
+ umount $SCRATCH_MNT > /dev/null 2>&1
+
+ if echo $mkfs_options | egrep -qv "raid1|raid5|raid6|raid10" || \
+ [ "${with_cancel}Q" = "cancelQ" ]; then
+ # the -r option has no effect without mirrors, skip -r test
+ # in this case, and if only the canceling should be tested
+ # as well
+ return 0
+ fi
+
+ # One more time with the '-r' option this time. Instead of wasting
+ # time to populate the filesystem with data again, use the
+ # existing filesystem in the state as it is after the previous
+ # replace operation.
+ # If possible, use a strategy to select the source and target
+ # device so that we really change bits on the target disk, see
+ # below.
+
+ # The default: For the 2nd run, the new target drive is the old
+ # source drive, and the new source drive is the old target drive.
+ # Since except for the noise data, the copied data is already on
+ # the new target disk (which is the old source disk), this is not
+ # optimal to check whether data is copied correctly.
+ local tmp_dev="$source_dev"
+ source_dev="$target_dev"
+ target_dev="$tmp_dev"
+
+ # If we have at least one more device in the SCRATCH_DEV_POOL than
+ # used so far, use one of those for the new target devive.
+ if [ "`echo $SCRATCH_DEV_POOL | wc -w`" -gt `expr $num_devs4raid + 1` ]; then
+ target_dev="`echo ${SCRATCH_DEV_POOL} | awk '{print $(NF-1)}'`"
+ fi
+
+ # If the filesystem is built out of more than one devices, use a
+ # different source device for this round.
+ if [ $num_devs4raid -gt 1 ]; then
+ source_dev="`echo ${SCRATCH_DEV_POOL} | awk '{print $2}'`"
+ fi
+
+ # Mount similar to _scratch_mount, but since the SCRATCH_DEV (the
+ # 1st device in SCRATCH_DEV_POOL) was replaced by the previous
+ # btrfs replace operation, substitute SCRATCH_DEV with a device
+ # that is known to be part of the SCRATCH_MNT filesystem.
+ _mount -t $FSTYP `_scratch_mount_options | sed "s&${SCRATCH_DEV}&${source_dev}&"`
+ if [ $? -ne 0 ]; then
+ echo "mount failed"
+ return 1
+ fi
+
+ btrfs_replace_test $source_dev $target_dev "-r" $with_cancel $quick
+ umount $SCRATCH_MNT > /dev/null 2>&1
+}
+
+btrfs_replace_test()
+{
+ local source_dev="$1"
+ local target_dev="$2"
+ local replace_options="$3"
+ local with_cancel="$4"
+ local quick="$5"
+
+ # generate some (slow) background traffic in parallel to the
+ # replace operation. It is not a problem if cat fails early
+ # with ENOSPC.
+ cat /dev/urandom > $SCRATCH_MNT/noise 2>> $seqres.full &
+ noise_pid=$!
+
+ if [ "${with_cancel}Q" = "cancelQ" ]; then
+ # background the replace operation (no '-B' option given)
+ $BTRFS_UTIL_PROG replace start -f $source_dev $target_dev $SCRATCH_MNT >> $seqres.full 2>&1 || _fail "btrfs replace start failed"
+ sleep 1
+ $BTRFS_UTIL_PROG replace cancel $SCRATCH_MNT >> $seqres.full 2>&1 || _fail "btrfs replace cancel failed"
+
+ # 'replace status' waits for the replace operation to finish
+ # before the status is printed
+ $BTRFS_UTIL_PROG replace status $SCRATCH_MNT > $tmp.tmp 2>&1
+ cat $tmp.tmp >> $seqres.full
+ grep -q canceled $tmp.tmp || _fail "btrfs replace status failed"
+ else
+ if [ "${quick}Q" = "thoroughQ" ]; then
+ # On current hardware, the thorough test runs
+ # more than a second. This is a chance to force
+ # a sync in the middle of the replace operation.
+ (sleep 1; sync) > /dev/null 2>&1 &
+ fi
+ $BTRFS_UTIL_PROG replace start -Bf $source_dev $target_dev $SCRATCH_MNT >> $seqres.full 2>&1 || _fail "btrfs replace start failed"
+
+ $BTRFS_UTIL_PROG replace status $SCRATCH_MNT > $tmp.tmp 2>&1
+ cat $tmp.tmp >> $seqres.full
+ grep -q finished $tmp.tmp || _fail "btrfs replace status failed"
+ fi
+
+ if ps -p $noise_pid | grep -q $noise_pid; then
+ kill -TERM $noise_pid 2> /dev/null
+ fi
+ noise_pid=0
+ wait
+
+ # scrub tests on-disk data, that's the reason for the sync.
+ # With the '-B' option (don't background), any type of error causes
+ # exit values != 0, including detected correctable and uncorrectable
+ # errors on the device.
+ sync; sync
+ $BTRFS_UTIL_PROG scrub start -B $SCRATCH_MNT >> $seqres.full 2>&1 || _fail "btrfs scrub failed"
+
+ # Two tests are performed, the 1st is to btrfsck the filesystem,
+ # and the 2nd test is to mount the filesystem.
+ # Usually _check_btrfs_filesystem would perform the mount test,
+ # but it gets confused by the mount output that shows SCRATCH_MNT
+ # mounted but not being mounted to SCRATCH_DEV. This happens
+ # because in /proc/mounts the 2nd device of the filesystem is
+ # shown after the replace operation. Let's just do the mount
+ # test manually after _check_btrfs_filesystem is finished.
+ umount $SCRATCH_MNT > /dev/null 2>&1
+ if [ "${with_cancel}Q" != "cancelQ" ]; then
+ # after the replace operation, use the target_dev for everything
+ _check_btrfs_filesystem $target_dev
+ _mount -t $FSTYP `_scratch_mount_options | sed "s&${SCRATCH_DEV}&${target_dev}&"`
+ else
+ _check_btrfs_filesystem $source_dev
+ _scratch_mount
+ fi
+}
+
+workout "-m single -d single" 1 no quick
+workout "-m single -d single -M" 1 no quick
+workout "-m dup -d single" 1 no quick
+workout "-m dup -d single" 1 cancel quick
+workout "-m dup -d dup -M" 1 no quick
+workout "-m raid0 -d raid0" 2 no quick
+workout "-m raid1 -d raid1" 2 no thorough
+#workout "-m raid5 -d raid5" 2 no quick # not yet supported for btrfs replace
+#workout "-m raid6 -d raid6" 3 no quick # not yet supported for btrfs replace
+workout "-m raid10 -d raid10" 4 no quick
+
+echo "*** done"
+status=0
+exit
diff --git a/tests/btrfs/010.out b/tests/btrfs/010.out
new file mode 100644
index 0000000..ef04a0e
--- /dev/null
+++ b/tests/btrfs/010.out
@@ -0,0 +1,3 @@
+QA output created by 010
+*** test btrfs replace
+*** done
diff --git a/tests/btrfs/group b/tests/btrfs/group
index 9eabadc..17c8ea7 100644
--- a/tests/btrfs/group
+++ b/tests/btrfs/group
@@ -12,3 +12,4 @@
007 auto rw metadata
008 auto quick
009 auto quick
+010 auto
--
1.8.3.4
^ permalink raw reply related [flat|nested] 24+ messages in thread* [PATCH][RESEND] xfstests: add a test for btrfs device replace operation
@ 2013-08-23 13:07 ` Stefan Behrens
0 siblings, 0 replies; 24+ messages in thread
From: Stefan Behrens @ 2013-08-23 13:07 UTC (permalink / raw)
To: xfs; +Cc: linux-btrfs
This test performs btrfs device replace tests with all possible profiles
(single/dup/mixed/raid0/raid1/raid10), one round with the '-r' option
to 'btrfs replace start' and one round without this option. The
cancelation is tested only once and with the dup/single profile for
metadata/data.
This test takes 181 seconds on my SSD equiped test box and 237s on
spinning disks. Almost all the time is spent when the filesystem is
populated with test data. The replace operation itself takes less than
a second for all the tests, except for the test that is marked as
'thorough' which will run for about 8 seconds on my test box.
The amount of tests done depends on the number of devices in the
SCRATCH_DEV_POOL. For full test coverage, at least 5 devices should
be available (e.g. 5 partitions). With less than 2 entries in
SCRATCH_DEV_POOL, the test is not executed.
The source and target devices for the replace operation are arbitrarily
chosen out of SCRATCH_DEV_POOl. Since the target device mustn't be
smaller than the source device, the requirement for this test is that
all devices have _exactly_ the same size. If this is not the case, the
test terminates with _notrun.
To check the filesystems after replacing a device, a scrub run is
performed, a btrfsck run, and finally the filesystem is remounted.
This commit depends on my other commit:
"xfstest: don't remove the two first devices from SCRATCH_DEV_POOL"
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
---
V1 -> V2:
Major reworking in order to address all the comments from Eric Sandeen's
review.
V2 -> V3:
- Rebased.
- The check that the devices have the same size was missing the target
device.
V3 -> V4:
Somehow I managed to add code that used a hardcoded path instead of
$SCRATCH_MNT. Fixed this issue.
common/config | 1 +
tests/btrfs/010 | 277 ++++++++++++++++++++++++++++++++++++++++++++++++++++
tests/btrfs/010.out | 3 +
tests/btrfs/group | 1 +
4 files changed, 282 insertions(+)
diff --git a/common/config b/common/config
index 586870b..db086fb 100644
--- a/common/config
+++ b/common/config
@@ -207,6 +207,7 @@ case "$HOSTOS" in
export MKFS_UDF_PROG="`set_prog_path mkudffs`"
export MKFS_BTRFS_PROG="`set_btrfs_mkfs_prog_path_with_opts`"
export BTRFS_UTIL_PROG="`set_prog_path btrfs`"
+ export BTRFS_SHOW_SUPER_PROG="`set_prog_path btrfs-show-super`"
export XFS_FSR_PROG="`set_prog_path xfs_fsr`"
export MKFS_NFS_PROG="false"
;;
diff --git a/tests/btrfs/010 b/tests/btrfs/010
new file mode 100755
index 0000000..36800ff
--- /dev/null
+++ b/tests/btrfs/010
@@ -0,0 +1,277 @@
+#! /bin/bash
+# FSQA Test No. btrfs/010
+#
+# Test of the btrfs replace operation.
+#
+# The amount of tests done depends on the number of devices in the
+# SCRATCH_DEV_POOL. For full test coverage, at least 5 devices should
+# be available (e.g. 5 partitions).
+#
+# The source and target devices for the replace operation are
+# arbitrarily chosen out of SCRATCH_DEV_POOl. Since the target device
+# mustn't be smaller than the source device, the requirement for this
+# test is that all devices have _exactly_ the same size. If this is
+# not the case, this test is not run.
+#
+# To check the filesystems after replacing a device, a scrub run is
+# performed, a btrfsck run, and finally the filesystem is remounted.
+#
+#-----------------------------------------------------------------------
+# Copyright (C) 2013 STRATO. 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
+noise_pid=0
+
+_cleanup()
+{
+ if [ $noise_pid -ne 0 ] && ps -p $noise_pid | grep -q $noise_pid; then
+ kill -TERM $noise_pid
+ fi
+ wait
+ rm -f $tmp.tmp
+}
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# real QA test starts here
+_need_to_be_root
+_supported_fs btrfs
+_require_scratch
+_require_scratch_dev_pool
+_require_command $BTRFS_SHOW_SUPER_PROG btrfs-show-super
+
+rm -f $seqres.full
+rm -f $tmp.tmp
+
+echo "*** test btrfs replace"
+
+workout()
+{
+ local mkfs_options="$1"
+ local num_devs4raid="$2"
+ local with_cancel="$3"
+ local quick="$4"
+ local source_dev="`echo ${SCRATCH_DEV_POOL} | awk '{print $1}'`"
+ local target_dev="`echo ${SCRATCH_DEV_POOL} | awk '{print $NF}'`"
+
+ if [ "`echo $SCRATCH_DEV_POOL | wc -w`" -lt `expr $num_devs4raid + 1` ]; then
+ echo "Skip workout $1 $2 $3 $4" >> $seqres.full
+ echo "Too few devices in SCRATCH_DEV_POOL $SCRATCH_DEV_POOL, required: " `expr $num_devs4raid + 1` >> $seqres.full
+ return 0
+ fi
+
+ # use min number of disks in order to fill up the disk to replace
+ # as much as possible
+ local used_devs_without_1st="`echo $SCRATCH_DEV_POOL | \
+ awk '{ORS=\" \"; for (i = 2; i <= (NF - 1 < '$num_devs4raid' ? NF - 1 : '$num_devs4raid'); i++) print $i}'`"
+
+ # _scratch_mkfs adds the 1st device again (which is $SCRATCH_DEV)
+ _scratch_mkfs $mkfs_options $used_devs_without_1st >> $seqres.full 2>&1 || _fail "mkfs failed"
+
+ # create a filesystem on the target device just for the sake of
+ # being able to query its size with btrfs-show-super
+ $MKFS_BTRFS_PROG $MKFS_OPTIONS $target_dev >> $seqres.full 2>&1 || _fail "mkfs target_dev failed"
+
+ # The source and target devices for the replace operation are
+ # arbitrarily chosen out of the pool. Since the target device mustn't
+ # be smaller than the source device, the requirement for this test is
+ # that all devices have _exactly_ the same size. If this is not the
+ # case, this test is not run.
+ local num_lines=`$BTRFS_SHOW_SUPER_PROG $SCRATCH_DEV $used_devs_without_1st $target_dev | grep dev_item.total_bytes | uniq | wc -l`
+ if [ $num_lines -gt 1 ]; then
+ _notrun "Different device sizes detected"
+ fi
+
+ _scratch_mount
+
+ # Generate 500 times 20K extents in the data chunk and fill up
+ # metadata with inline extents. Ignore ENOSPC.
+ for i in `seq 1 500`; do
+ dd if=/dev/urandom of=$SCRATCH_MNT/l$i bs=16385 count=1
+ dd if=/dev/urandom of=$SCRATCH_MNT/s$i bs=3800 count=1
+ done > /dev/null 2>&1
+
+ if [ "${quick}Q" = "thoroughQ" ]; then
+ # The intention of this "thorough" test is to increase
+ # the probability of random errors, in particular in
+ # conjunction with the background noise generator and
+ # a sync call while the replace operation in ongoing.
+ # Unfortunately it takes quite some time to generate
+ # the test filesystem, therefore most data consists out
+ # of zeros although this data is not very useful for
+ # detecting misplaced read/write requests.
+ # Ignore ENOSPC, it's not a problem..
+ dd if=/dev/urandom of=$SCRATCH_MNT/r bs=1M count=200 >> $seqres.full 2>&1 &
+ dd if=/dev/zero of=$SCRATCH_MNT/0 bs=1M count=2000 >> $seqres.full 2>&1
+ wait
+ elif [ "${with_cancel}Q" = "cancelQ" ]; then
+ # produce some data to prevent that the replace operation
+ # finishes before the cancel request is started
+ dd if=/dev/zero of=$SCRATCH_MNT/0 bs=1M count=1000 >> $seqres.full 2>&1
+ fi
+ sync; sync
+
+ btrfs_replace_test $source_dev $target_dev "" $with_cancel $quick
+ umount $SCRATCH_MNT > /dev/null 2>&1
+
+ if echo $mkfs_options | egrep -qv "raid1|raid5|raid6|raid10" || \
+ [ "${with_cancel}Q" = "cancelQ" ]; then
+ # the -r option has no effect without mirrors, skip -r test
+ # in this case, and if only the canceling should be tested
+ # as well
+ return 0
+ fi
+
+ # One more time with the '-r' option this time. Instead of wasting
+ # time to populate the filesystem with data again, use the
+ # existing filesystem in the state as it is after the previous
+ # replace operation.
+ # If possible, use a strategy to select the source and target
+ # device so that we really change bits on the target disk, see
+ # below.
+
+ # The default: For the 2nd run, the new target drive is the old
+ # source drive, and the new source drive is the old target drive.
+ # Since except for the noise data, the copied data is already on
+ # the new target disk (which is the old source disk), this is not
+ # optimal to check whether data is copied correctly.
+ local tmp_dev="$source_dev"
+ source_dev="$target_dev"
+ target_dev="$tmp_dev"
+
+ # If we have at least one more device in the SCRATCH_DEV_POOL than
+ # used so far, use one of those for the new target devive.
+ if [ "`echo $SCRATCH_DEV_POOL | wc -w`" -gt `expr $num_devs4raid + 1` ]; then
+ target_dev="`echo ${SCRATCH_DEV_POOL} | awk '{print $(NF-1)}'`"
+ fi
+
+ # If the filesystem is built out of more than one devices, use a
+ # different source device for this round.
+ if [ $num_devs4raid -gt 1 ]; then
+ source_dev="`echo ${SCRATCH_DEV_POOL} | awk '{print $2}'`"
+ fi
+
+ # Mount similar to _scratch_mount, but since the SCRATCH_DEV (the
+ # 1st device in SCRATCH_DEV_POOL) was replaced by the previous
+ # btrfs replace operation, substitute SCRATCH_DEV with a device
+ # that is known to be part of the SCRATCH_MNT filesystem.
+ _mount -t $FSTYP `_scratch_mount_options | sed "s&${SCRATCH_DEV}&${source_dev}&"`
+ if [ $? -ne 0 ]; then
+ echo "mount failed"
+ return 1
+ fi
+
+ btrfs_replace_test $source_dev $target_dev "-r" $with_cancel $quick
+ umount $SCRATCH_MNT > /dev/null 2>&1
+}
+
+btrfs_replace_test()
+{
+ local source_dev="$1"
+ local target_dev="$2"
+ local replace_options="$3"
+ local with_cancel="$4"
+ local quick="$5"
+
+ # generate some (slow) background traffic in parallel to the
+ # replace operation. It is not a problem if cat fails early
+ # with ENOSPC.
+ cat /dev/urandom > $SCRATCH_MNT/noise 2>> $seqres.full &
+ noise_pid=$!
+
+ if [ "${with_cancel}Q" = "cancelQ" ]; then
+ # background the replace operation (no '-B' option given)
+ $BTRFS_UTIL_PROG replace start -f $source_dev $target_dev $SCRATCH_MNT >> $seqres.full 2>&1 || _fail "btrfs replace start failed"
+ sleep 1
+ $BTRFS_UTIL_PROG replace cancel $SCRATCH_MNT >> $seqres.full 2>&1 || _fail "btrfs replace cancel failed"
+
+ # 'replace status' waits for the replace operation to finish
+ # before the status is printed
+ $BTRFS_UTIL_PROG replace status $SCRATCH_MNT > $tmp.tmp 2>&1
+ cat $tmp.tmp >> $seqres.full
+ grep -q canceled $tmp.tmp || _fail "btrfs replace status failed"
+ else
+ if [ "${quick}Q" = "thoroughQ" ]; then
+ # On current hardware, the thorough test runs
+ # more than a second. This is a chance to force
+ # a sync in the middle of the replace operation.
+ (sleep 1; sync) > /dev/null 2>&1 &
+ fi
+ $BTRFS_UTIL_PROG replace start -Bf $source_dev $target_dev $SCRATCH_MNT >> $seqres.full 2>&1 || _fail "btrfs replace start failed"
+
+ $BTRFS_UTIL_PROG replace status $SCRATCH_MNT > $tmp.tmp 2>&1
+ cat $tmp.tmp >> $seqres.full
+ grep -q finished $tmp.tmp || _fail "btrfs replace status failed"
+ fi
+
+ if ps -p $noise_pid | grep -q $noise_pid; then
+ kill -TERM $noise_pid 2> /dev/null
+ fi
+ noise_pid=0
+ wait
+
+ # scrub tests on-disk data, that's the reason for the sync.
+ # With the '-B' option (don't background), any type of error causes
+ # exit values != 0, including detected correctable and uncorrectable
+ # errors on the device.
+ sync; sync
+ $BTRFS_UTIL_PROG scrub start -B $SCRATCH_MNT >> $seqres.full 2>&1 || _fail "btrfs scrub failed"
+
+ # Two tests are performed, the 1st is to btrfsck the filesystem,
+ # and the 2nd test is to mount the filesystem.
+ # Usually _check_btrfs_filesystem would perform the mount test,
+ # but it gets confused by the mount output that shows SCRATCH_MNT
+ # mounted but not being mounted to SCRATCH_DEV. This happens
+ # because in /proc/mounts the 2nd device of the filesystem is
+ # shown after the replace operation. Let's just do the mount
+ # test manually after _check_btrfs_filesystem is finished.
+ umount $SCRATCH_MNT > /dev/null 2>&1
+ if [ "${with_cancel}Q" != "cancelQ" ]; then
+ # after the replace operation, use the target_dev for everything
+ _check_btrfs_filesystem $target_dev
+ _mount -t $FSTYP `_scratch_mount_options | sed "s&${SCRATCH_DEV}&${target_dev}&"`
+ else
+ _check_btrfs_filesystem $source_dev
+ _scratch_mount
+ fi
+}
+
+workout "-m single -d single" 1 no quick
+workout "-m single -d single -M" 1 no quick
+workout "-m dup -d single" 1 no quick
+workout "-m dup -d single" 1 cancel quick
+workout "-m dup -d dup -M" 1 no quick
+workout "-m raid0 -d raid0" 2 no quick
+workout "-m raid1 -d raid1" 2 no thorough
+#workout "-m raid5 -d raid5" 2 no quick # not yet supported for btrfs replace
+#workout "-m raid6 -d raid6" 3 no quick # not yet supported for btrfs replace
+workout "-m raid10 -d raid10" 4 no quick
+
+echo "*** done"
+status=0
+exit
diff --git a/tests/btrfs/010.out b/tests/btrfs/010.out
new file mode 100644
index 0000000..ef04a0e
--- /dev/null
+++ b/tests/btrfs/010.out
@@ -0,0 +1,3 @@
+QA output created by 010
+*** test btrfs replace
+*** done
diff --git a/tests/btrfs/group b/tests/btrfs/group
index 9eabadc..17c8ea7 100644
--- a/tests/btrfs/group
+++ b/tests/btrfs/group
@@ -12,3 +12,4 @@
007 auto rw metadata
008 auto quick
009 auto quick
+010 auto
--
1.8.3.4
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [PATCH][RESEND] xfstests: add a test for btrfs device replace operation
2013-08-23 13:07 ` Stefan Behrens
@ 2013-08-23 18:35 ` Josef Bacik
-1 siblings, 0 replies; 24+ messages in thread
From: Josef Bacik @ 2013-08-23 18:35 UTC (permalink / raw)
To: Stefan Behrens; +Cc: xfs, linux-btrfs
On Fri, Aug 23, 2013 at 03:07:11PM +0200, Stefan Behrens wrote:
> This test performs btrfs device replace tests with all possible profiles
> (single/dup/mixed/raid0/raid1/raid10), one round with the '-r' option
> to 'btrfs replace start' and one round without this option. The
> cancelation is tested only once and with the dup/single profile for
> metadata/data.
>
> This test takes 181 seconds on my SSD equiped test box and 237s on
> spinning disks. Almost all the time is spent when the filesystem is
> populated with test data. The replace operation itself takes less than
> a second for all the tests, except for the test that is marked as
> 'thorough' which will run for about 8 seconds on my test box.
>
> The amount of tests done depends on the number of devices in the
> SCRATCH_DEV_POOL. For full test coverage, at least 5 devices should
> be available (e.g. 5 partitions). With less than 2 entries in
> SCRATCH_DEV_POOL, the test is not executed.
>
> The source and target devices for the replace operation are arbitrarily
> chosen out of SCRATCH_DEV_POOl. Since the target device mustn't be
> smaller than the source device, the requirement for this test is that
> all devices have _exactly_ the same size. If this is not the case, the
> test terminates with _notrun.
>
> To check the filesystems after replacing a device, a scrub run is
> performed, a btrfsck run, and finally the filesystem is remounted.
>
> This commit depends on my other commit:
> "xfstest: don't remove the two first devices from SCRATCH_DEV_POOL"
>
> Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
This worked well, thanks,
Reviewed-by: Josef Bacik <jbacik@fusionio.com>
Thanks,
Josef
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH][RESEND] xfstests: add a test for btrfs device replace operation
@ 2013-08-23 18:35 ` Josef Bacik
0 siblings, 0 replies; 24+ messages in thread
From: Josef Bacik @ 2013-08-23 18:35 UTC (permalink / raw)
To: Stefan Behrens; +Cc: linux-btrfs, xfs
On Fri, Aug 23, 2013 at 03:07:11PM +0200, Stefan Behrens wrote:
> This test performs btrfs device replace tests with all possible profiles
> (single/dup/mixed/raid0/raid1/raid10), one round with the '-r' option
> to 'btrfs replace start' and one round without this option. The
> cancelation is tested only once and with the dup/single profile for
> metadata/data.
>
> This test takes 181 seconds on my SSD equiped test box and 237s on
> spinning disks. Almost all the time is spent when the filesystem is
> populated with test data. The replace operation itself takes less than
> a second for all the tests, except for the test that is marked as
> 'thorough' which will run for about 8 seconds on my test box.
>
> The amount of tests done depends on the number of devices in the
> SCRATCH_DEV_POOL. For full test coverage, at least 5 devices should
> be available (e.g. 5 partitions). With less than 2 entries in
> SCRATCH_DEV_POOL, the test is not executed.
>
> The source and target devices for the replace operation are arbitrarily
> chosen out of SCRATCH_DEV_POOl. Since the target device mustn't be
> smaller than the source device, the requirement for this test is that
> all devices have _exactly_ the same size. If this is not the case, the
> test terminates with _notrun.
>
> To check the filesystems after replacing a device, a scrub run is
> performed, a btrfsck run, and finally the filesystem is remounted.
>
> This commit depends on my other commit:
> "xfstest: don't remove the two first devices from SCRATCH_DEV_POOL"
>
> Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
This worked well, thanks,
Reviewed-by: Josef Bacik <jbacik@fusionio.com>
Thanks,
Josef
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH][RESEND] xfstests: add a test for btrfs device replace operation
2013-08-23 13:07 ` Stefan Behrens
@ 2013-08-28 16:25 ` Rich Johnston
-1 siblings, 0 replies; 24+ messages in thread
From: Rich Johnston @ 2013-08-28 16:25 UTC (permalink / raw)
To: xfstests; +Cc: linux-btrfs
Thanks for the patch Stefan, it has been committed.
--Rich
commit 53b73199db79dcd58622aaac2e9b0f31073bfc44
Author: Stefan Behrens <sbehrens@giantdisaster.de>
Date: Fri Aug 23 13:07:11 2013 +0000
xfstests: add a test for btrfs device replace operation
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH] xfstests: update _filter_size() for Btrfs
2013-08-23 13:07 ` Stefan Behrens
@ 2013-08-23 13:07 ` Stefan Behrens
-1 siblings, 0 replies; 24+ messages in thread
From: Stefan Behrens @ 2013-08-23 13:07 UTC (permalink / raw)
To: xfs; +Cc: linux-btrfs
From: root <root@qvarne.iata>
The btrfs-progs tools changed the output:
- 100GiB instead of 100GB
xfstest btrfs/006 is one that failed due to this change.
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
---
common/filter | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/common/filter b/common/filter
index dbb1674..ee738ca 100644
--- a/common/filter
+++ b/common/filter
@@ -262,7 +262,7 @@ _filter_uuid()
# Filter out sizes like 6.14MB etc
_filter_size()
{
- sed -e "s/[0-9\.]\+\s\?[b|k|m|g|t][b]\?/<SIZE>/ig"
+ sed -e "s/[0-9\.]\+\s\?[b|k|m|g|t][i]\?[b]\?/<SIZE>/ig"
}
# Convert string read from stdin like 128K to bytes and print it to stdout
--
1.8.3.4
^ permalink raw reply related [flat|nested] 24+ messages in thread* [PATCH] xfstests: update _filter_size() for Btrfs
@ 2013-08-23 13:07 ` Stefan Behrens
0 siblings, 0 replies; 24+ messages in thread
From: Stefan Behrens @ 2013-08-23 13:07 UTC (permalink / raw)
To: xfs; +Cc: linux-btrfs
From: root <root@qvarne.iata>
The btrfs-progs tools changed the output:
- 100GiB instead of 100GB
xfstest btrfs/006 is one that failed due to this change.
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
---
common/filter | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/common/filter b/common/filter
index dbb1674..ee738ca 100644
--- a/common/filter
+++ b/common/filter
@@ -262,7 +262,7 @@ _filter_uuid()
# Filter out sizes like 6.14MB etc
_filter_size()
{
- sed -e "s/[0-9\.]\+\s\?[b|k|m|g|t][b]\?/<SIZE>/ig"
+ sed -e "s/[0-9\.]\+\s\?[b|k|m|g|t][i]\?[b]\?/<SIZE>/ig"
}
# Convert string read from stdin like 128K to bytes and print it to stdout
--
1.8.3.4
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [PATCH] xfstests: update _filter_size() for Btrfs
2013-08-23 13:07 ` Stefan Behrens
@ 2013-08-23 18:34 ` Josef Bacik
-1 siblings, 0 replies; 24+ messages in thread
From: Josef Bacik @ 2013-08-23 18:34 UTC (permalink / raw)
To: Stefan Behrens; +Cc: xfs, linux-btrfs
On Fri, Aug 23, 2013 at 03:07:12PM +0200, Stefan Behrens wrote:
> From: root <root@qvarne.iata>
>
> The btrfs-progs tools changed the output:
> - 100GiB instead of 100GB
>
> xfstest btrfs/006 is one that failed due to this change.
>
> Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
Thank you for this, it was super annoying with the new btrfs-progs breaking
this.
Reviewed-by: Josef Bacik <jbacik@fusionio.com>
Thanks,
Josef
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH] xfstests: update _filter_size() for Btrfs
@ 2013-08-23 18:34 ` Josef Bacik
0 siblings, 0 replies; 24+ messages in thread
From: Josef Bacik @ 2013-08-23 18:34 UTC (permalink / raw)
To: Stefan Behrens; +Cc: linux-btrfs, xfs
On Fri, Aug 23, 2013 at 03:07:12PM +0200, Stefan Behrens wrote:
> From: root <root@qvarne.iata>
>
> The btrfs-progs tools changed the output:
> - 100GiB instead of 100GB
>
> xfstest btrfs/006 is one that failed due to this change.
>
> Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
Thank you for this, it was super annoying with the new btrfs-progs breaking
this.
Reviewed-by: Josef Bacik <jbacik@fusionio.com>
Thanks,
Josef
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH] xfstests: update _filter_size() for Btrfs
2013-08-23 13:07 ` Stefan Behrens
@ 2013-08-28 15:38 ` Rich Johnston
-1 siblings, 0 replies; 24+ messages in thread
From: Rich Johnston @ 2013-08-28 15:38 UTC (permalink / raw)
To: xfs; +Cc: linux-btrfs
Thanks for the patch Stefan, it has been committed.
--Rich
commit 8a51dad60a015e0b261e0f78428731b00b992ce9
Author: Stefan Behrens <sbehrens@giantdisaster.de>
Date: Fri Aug 23 13:07:12 2013 +0000
xfstests: update _filter_size() for Btrfs
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH] xfstest: fix btrfs/006 for 10+ devices in SCRATCH_DEV_POOL
2013-08-23 13:07 ` Stefan Behrens
@ 2013-08-23 13:07 ` Stefan Behrens
-1 siblings, 0 replies; 24+ messages in thread
From: Stefan Behrens @ 2013-08-23 13:07 UTC (permalink / raw)
To: xfs; +Cc: linux-btrfs
One problem was the output of "uniq -c" which added spaces depending
on the size of the count value (e.g. one space less for 10+ devices).
The second problem was that "btrfs fi show" was doing the same:
"devid %4llu size %s used %s path %s".
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
---
common/filter.btrfs | 4 ++--
tests/btrfs/006.out | 14 +++++++-------
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/common/filter.btrfs b/common/filter.btrfs
index e9a2bc2..29512cd 100644
--- a/common/filter.btrfs
+++ b/common/filter.btrfs
@@ -10,7 +10,7 @@ _filter_btrfs_version()
_filter_devid()
{
- sed -e "s/\(devid\s\+\)[0-9]\+/\1 <DEVID>/g"
+ sed -e "s/\(devid\)\s\+[0-9]\+/\1 <DEVID>/g"
}
# If passed a number as first arg, filter that number of devices
@@ -53,7 +53,7 @@ _filter_btrfs_device_stats()
_filter_scratch | _filter_scratch_pool | \
sed -e "s/[0-9]\+$/<NUM>/g" | sort | uniq $UNIQ_OPT | \
- sed -e "s/$NUMDEVS /<NUMDEVS> /g"
+ sed -e "s/ *$NUMDEVS /<NUMDEVS> /g"
}
# make sure this script returns success
diff --git a/tests/btrfs/006.out b/tests/btrfs/006.out
index ab33b7e..22bcb77 100644
--- a/tests/btrfs/006.out
+++ b/tests/btrfs/006.out
@@ -6,21 +6,21 @@ TestLabel.006
== Show filesystem by label
Label: 'TestLabel.006' uuid: <UUID>
Total devices <EXACTNUM> FS bytes used <SIZE>
- devid <DEVID> size <SIZE> used <SIZE> path SCRATCH_DEV
+ devid <DEVID> size <SIZE> used <SIZE> path SCRATCH_DEV
== Show filesystem by UUID
Label: 'TestLabel.006' uuid: <EXACTUUID>
Total devices <EXACTNUM> FS bytes used <SIZE>
- devid <DEVID> size <SIZE> used <SIZE> path SCRATCH_DEV
+ devid <DEVID> size <SIZE> used <SIZE> path SCRATCH_DEV
== Sync filesystem
FSSync 'SCRATCH_MNT'
== Show device stats by mountpoint
- <NUMDEVS> [SCRATCH_DEV].corruption_errs <NUM>
- <NUMDEVS> [SCRATCH_DEV].flush_io_errs <NUM>
- <NUMDEVS> [SCRATCH_DEV].generation_errs <NUM>
- <NUMDEVS> [SCRATCH_DEV].read_io_errs <NUM>
- <NUMDEVS> [SCRATCH_DEV].write_io_errs <NUM>
+<NUMDEVS> [SCRATCH_DEV].corruption_errs <NUM>
+<NUMDEVS> [SCRATCH_DEV].flush_io_errs <NUM>
+<NUMDEVS> [SCRATCH_DEV].generation_errs <NUM>
+<NUMDEVS> [SCRATCH_DEV].read_io_errs <NUM>
+<NUMDEVS> [SCRATCH_DEV].write_io_errs <NUM>
== Show device stats by first/scratch dev
[SCRATCH_DEV].corruption_errs <NUM>
[SCRATCH_DEV].flush_io_errs <NUM>
--
1.8.3.4
^ permalink raw reply related [flat|nested] 24+ messages in thread* [PATCH] xfstest: fix btrfs/006 for 10+ devices in SCRATCH_DEV_POOL
@ 2013-08-23 13:07 ` Stefan Behrens
0 siblings, 0 replies; 24+ messages in thread
From: Stefan Behrens @ 2013-08-23 13:07 UTC (permalink / raw)
To: xfs; +Cc: linux-btrfs
One problem was the output of "uniq -c" which added spaces depending
on the size of the count value (e.g. one space less for 10+ devices).
The second problem was that "btrfs fi show" was doing the same:
"devid %4llu size %s used %s path %s".
Signed-off-by: Stefan Behrens <sbehrens@giantdisaster.de>
---
common/filter.btrfs | 4 ++--
tests/btrfs/006.out | 14 +++++++-------
2 files changed, 9 insertions(+), 9 deletions(-)
diff --git a/common/filter.btrfs b/common/filter.btrfs
index e9a2bc2..29512cd 100644
--- a/common/filter.btrfs
+++ b/common/filter.btrfs
@@ -10,7 +10,7 @@ _filter_btrfs_version()
_filter_devid()
{
- sed -e "s/\(devid\s\+\)[0-9]\+/\1 <DEVID>/g"
+ sed -e "s/\(devid\)\s\+[0-9]\+/\1 <DEVID>/g"
}
# If passed a number as first arg, filter that number of devices
@@ -53,7 +53,7 @@ _filter_btrfs_device_stats()
_filter_scratch | _filter_scratch_pool | \
sed -e "s/[0-9]\+$/<NUM>/g" | sort | uniq $UNIQ_OPT | \
- sed -e "s/$NUMDEVS /<NUMDEVS> /g"
+ sed -e "s/ *$NUMDEVS /<NUMDEVS> /g"
}
# make sure this script returns success
diff --git a/tests/btrfs/006.out b/tests/btrfs/006.out
index ab33b7e..22bcb77 100644
--- a/tests/btrfs/006.out
+++ b/tests/btrfs/006.out
@@ -6,21 +6,21 @@ TestLabel.006
== Show filesystem by label
Label: 'TestLabel.006' uuid: <UUID>
Total devices <EXACTNUM> FS bytes used <SIZE>
- devid <DEVID> size <SIZE> used <SIZE> path SCRATCH_DEV
+ devid <DEVID> size <SIZE> used <SIZE> path SCRATCH_DEV
== Show filesystem by UUID
Label: 'TestLabel.006' uuid: <EXACTUUID>
Total devices <EXACTNUM> FS bytes used <SIZE>
- devid <DEVID> size <SIZE> used <SIZE> path SCRATCH_DEV
+ devid <DEVID> size <SIZE> used <SIZE> path SCRATCH_DEV
== Sync filesystem
FSSync 'SCRATCH_MNT'
== Show device stats by mountpoint
- <NUMDEVS> [SCRATCH_DEV].corruption_errs <NUM>
- <NUMDEVS> [SCRATCH_DEV].flush_io_errs <NUM>
- <NUMDEVS> [SCRATCH_DEV].generation_errs <NUM>
- <NUMDEVS> [SCRATCH_DEV].read_io_errs <NUM>
- <NUMDEVS> [SCRATCH_DEV].write_io_errs <NUM>
+<NUMDEVS> [SCRATCH_DEV].corruption_errs <NUM>
+<NUMDEVS> [SCRATCH_DEV].flush_io_errs <NUM>
+<NUMDEVS> [SCRATCH_DEV].generation_errs <NUM>
+<NUMDEVS> [SCRATCH_DEV].read_io_errs <NUM>
+<NUMDEVS> [SCRATCH_DEV].write_io_errs <NUM>
== Show device stats by first/scratch dev
[SCRATCH_DEV].corruption_errs <NUM>
[SCRATCH_DEV].flush_io_errs <NUM>
--
1.8.3.4
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [PATCH] xfstest: fix btrfs/006 for 10+ devices in SCRATCH_DEV_POOL
2013-08-23 13:07 ` Stefan Behrens
@ 2013-08-23 18:33 ` Josef Bacik
-1 siblings, 0 replies; 24+ messages in thread
From: Josef Bacik @ 2013-08-23 18:33 UTC (permalink / raw)
To: Stefan Behrens; +Cc: xfs, linux-btrfs
On Fri, Aug 23, 2013 at 03:07:13PM +0200, Stefan Behrens wrote:
> One problem was the output of "uniq -c" which added spaces depending
> on the size of the count value (e.g. one space less for 10+ devices).
>
> The second problem was that "btrfs fi show" was doing the same:
> "devid %4llu size %s used %s path %s".
>
Got 11 devices together and this works
Reviewed-by: Josef Bacik <jbacik@fusionio.com>
Thanks,
Josef
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH] xfstest: fix btrfs/006 for 10+ devices in SCRATCH_DEV_POOL
@ 2013-08-23 18:33 ` Josef Bacik
0 siblings, 0 replies; 24+ messages in thread
From: Josef Bacik @ 2013-08-23 18:33 UTC (permalink / raw)
To: Stefan Behrens; +Cc: linux-btrfs, xfs
On Fri, Aug 23, 2013 at 03:07:13PM +0200, Stefan Behrens wrote:
> One problem was the output of "uniq -c" which added spaces depending
> on the size of the count value (e.g. one space less for 10+ devices).
>
> The second problem was that "btrfs fi show" was doing the same:
> "devid %4llu size %s used %s path %s".
>
Got 11 devices together and this works
Reviewed-by: Josef Bacik <jbacik@fusionio.com>
Thanks,
Josef
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH] xfstest: fix btrfs/006 for 10+ devices in SCRATCH_DEV_POOL
2013-08-23 13:07 ` Stefan Behrens
@ 2013-08-28 15:30 ` Rich Johnston
-1 siblings, 0 replies; 24+ messages in thread
From: Rich Johnston @ 2013-08-28 15:30 UTC (permalink / raw)
To: xfs; +Cc: linux-btrfs
Thanks for the patch Stefan, it has been committed.
--Rich
commit 0d3bbd18942bfac464d18f88c0a3c100c67a24f2
Author: Stefan Behrens <sbehrens@giantdisaster.de>
Date: Fri Aug 23 13:07:13 2013 +0000
xfstests: fix btrfs/006 for 10+ devices in SCRATCH_DEV_POOL
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH V4] xfstests: don't remove the two first devices from SCRATCH_DEV_POOL
2013-08-23 13:07 ` Stefan Behrens
@ 2013-08-23 18:32 ` Josef Bacik
-1 siblings, 0 replies; 24+ messages in thread
From: Josef Bacik @ 2013-08-23 18:32 UTC (permalink / raw)
To: Stefan Behrens; +Cc: xfs, linux-btrfs
On Fri, Aug 23, 2013 at 03:07:10PM +0200, Stefan Behrens wrote:
> Since common/config is executed twice, if SCRATCH_DEV_POOL is configured
> via the environment, the current code removes the first device entry twice
> which means that you lose the second device for the test.
>
> The fix is to not remove anything from SCRATCH_DEV_POOL anymore.
> That used to be done (I can only guess) to allow to pass the
> SCRATCH_DEV_POOL as an argument to _scratch_mkfs. Since _scratch_mkfs adds
> the SCRATCH_DEV, the pool mustn't contain that device anymore.
>
> A new function _scratch_pool_mkfs is introduced that does the expected
> thing.
>
This didn't break anything and makes sense
Reviewed-by: Josef Bacik <jbacik@fusionio.com>
Thanks,
Josef
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH V4] xfstests: don't remove the two first devices from SCRATCH_DEV_POOL
@ 2013-08-23 18:32 ` Josef Bacik
0 siblings, 0 replies; 24+ messages in thread
From: Josef Bacik @ 2013-08-23 18:32 UTC (permalink / raw)
To: Stefan Behrens; +Cc: linux-btrfs, xfs
On Fri, Aug 23, 2013 at 03:07:10PM +0200, Stefan Behrens wrote:
> Since common/config is executed twice, if SCRATCH_DEV_POOL is configured
> via the environment, the current code removes the first device entry twice
> which means that you lose the second device for the test.
>
> The fix is to not remove anything from SCRATCH_DEV_POOL anymore.
> That used to be done (I can only guess) to allow to pass the
> SCRATCH_DEV_POOL as an argument to _scratch_mkfs. Since _scratch_mkfs adds
> the SCRATCH_DEV, the pool mustn't contain that device anymore.
>
> A new function _scratch_pool_mkfs is introduced that does the expected
> thing.
>
This didn't break anything and makes sense
Reviewed-by: Josef Bacik <jbacik@fusionio.com>
Thanks,
Josef
_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH V4] xfstests: don't remove the two first devices from SCRATCH_DEV_POOL
2013-08-23 13:07 ` Stefan Behrens
@ 2013-08-28 15:18 ` Rich Johnston
-1 siblings, 0 replies; 24+ messages in thread
From: Rich Johnston @ 2013-08-28 15:18 UTC (permalink / raw)
To: xfs; +Cc: linux-btrfs
Thanks for the patch Stefan, it has been committed.
--Rich
commit f1dce456c594a784afb723d1bc7c09056ab3d9d9
Author: Stefan Behrens <sbehrens@giantdisaster.de>
Date: Fri Aug 23 13:07:10 2013 +0000
xfstests: don't remove the two first devices from SCRATCH_DEV_POOL
^ permalink raw reply [flat|nested] 24+ messages in thread