| Message ID | 20200917141353.28566-1-johannes.thumshirn@wdc.com (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
| Series | btrfs: remove stale test for alien devices | expand |
The fix is not too far. It got stuck whether to use EUCLEAN or not. Its better to fix the fix rather than killing the messenger in this case. Thanks, Anand On 17/9/20 10:13 pm, Johannes Thumshirn wrote: > btrfs/198 is supposed to be a test for the patch > "btrfs: remove identified alien device in open_fs_devices" but this patch > was never merged in btrfs. > > Remove the test from fstests as it is constantly failing. > > Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> > --- > tests/btrfs/198 | 77 ----------------------------------------------- > tests/btrfs/group | 1 - > 2 files changed, 78 deletions(-) > delete mode 100755 tests/btrfs/198 > > diff --git a/tests/btrfs/198 b/tests/btrfs/198 > deleted file mode 100755 > index 2df075e27134..000000000000 > --- a/tests/btrfs/198 > +++ /dev/null > @@ -1,77 +0,0 @@ > -#! /bin/bash > -# SPDX-License-Identifier: GPL-2.0 > -# Copyright (c) 2019 Oracle. All Rights Reserved. > -# > -# FS QA Test 198 > -# > -# Test stale and alien non-btrfs device in the fs devices list. > -# Bug fixed in: > -# btrfs: remove identified alien device in open_fs_devices > -# > -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.* > -} > - > -# get standard environment, filters and checks > -. ./common/rc > -. ./common/filter > -. ./common/filter.btrfs > - > -# remove previous $seqres.full before test > -rm -f $seqres.full > - > -# real QA test starts here > -_supported_fs btrfs > -_supported_os Linux > -_require_command "$WIPEFS_PROG" wipefs > -_require_scratch > -_require_scratch_dev_pool 4 > - > -workout() > -{ > - raid=$1 > - device_nr=$2 > - > - echo $raid > - _scratch_dev_pool_get $device_nr > - > - _scratch_pool_mkfs "-d$raid -m$raid" >> $seqres.full 2>&1 || \ > - _fail "mkfs failed" > - > - # Make device_1 a free btrfs device for the raid created above by > - # clearing its superblock > - > - # don't test with the first device as auto fs check (_check_scratch_fs) > - # picks the first device > - device_1=$(echo $SCRATCH_DEV_POOL | awk '{print $2}') > - $WIPEFS_PROG -a $device_1 >> $seqres.full 2>&1 > - > - device_2=$(echo $SCRATCH_DEV_POOL | awk '{print $1}') > - _mount -o degraded $device_2 $SCRATCH_MNT > - # Check if missing device is reported as in the 196.out > - $BTRFS_UTIL_PROG filesystem show -m $SCRATCH_MNT | \ > - _filter_btrfs_filesystem_show > - > - _scratch_unmount > - _scratch_dev_pool_put > -} > - > -workout "raid1" "2" > -workout "raid5" "3" > -workout "raid6" "4" > -workout "raid10" "4" > - > -# success, all done > -status=0 > -exit > diff --git a/tests/btrfs/group b/tests/btrfs/group > index 1b5fa695a9f7..d5330cd85cd5 100644 > --- a/tests/btrfs/group > +++ b/tests/btrfs/group > @@ -199,7 +199,6 @@ > 195 auto volume balance > 196 auto metadata log volume > 197 auto quick volume > -198 auto quick volume > 199 auto quick trim > 200 auto quick send clone > 201 auto quick punch log >
On 18/09/2020 02:15, Anand Jain wrote: > The fix is not too far. It got stuck whether to use EUCLEAN or not. > Its better to fix the fix rather than killing the messenger in this case. OK how about removing the test from the auto group then until the fix is merged? It's a constant failure and hiding real regressions. And having to maintain an expunge list doesn't scale either. Thoughts?
On Fri, Sep 18, 2020 at 07:06:42AM +0000, Johannes Thumshirn wrote: > On 18/09/2020 02:15, Anand Jain wrote: > > The fix is not too far. It got stuck whether to use EUCLEAN or not. > > Its better to fix the fix rather than killing the messenger in this case. > > OK how about removing the test from the auto group then until the fix is merged? > It's a constant failure and hiding real regressions. And having to maintain an > expunge list doesn't scale either. > > Thoughts? Yeah, please remove it from the auto group then. Thanks, Eryu >
On 21/9/20 12:15 am, Eryu Guan wrote: > On Fri, Sep 18, 2020 at 07:06:42AM +0000, Johannes Thumshirn wrote: >> On 18/09/2020 02:15, Anand Jain wrote: >>> The fix is not too far. It got stuck whether to use EUCLEAN or not. >>> Its better to fix the fix rather than killing the messenger in this case. >> >> OK how about removing the test from the auto group then until the fix is merged? >> It's a constant failure and hiding real regressions. And having to maintain an >> expunge list doesn't scale either. >> >> Thoughts? > The patch is in the ML for review. [PATCH] btrfs: free device without BTRFS_MAGIC https://patchwork.kernel.org/patch/11786607/ I am OK if you still think it has to be removed from the auto. But I don't think it is required now. > Yeah, please remove it from the auto group then. > > Thanks, > Eryu >>
On Mon, Sep 21, 2020 at 10:37:49AM +0800, Anand Jain wrote: > > > On 21/9/20 12:15 am, Eryu Guan wrote: > > On Fri, Sep 18, 2020 at 07:06:42AM +0000, Johannes Thumshirn wrote: > >> On 18/09/2020 02:15, Anand Jain wrote: > >>> The fix is not too far. It got stuck whether to use EUCLEAN or not. > >>> Its better to fix the fix rather than killing the messenger in this case. > >> > >> OK how about removing the test from the auto group then until the fix is merged? > >> It's a constant failure and hiding real regressions. And having to maintain an > >> expunge list doesn't scale either. > >> > >> Thoughts? > > > > The patch is in the ML for review. > > [PATCH] btrfs: free device without BTRFS_MAGIC > > https://patchwork.kernel.org/patch/11786607/ > > I am OK if you still think it has to be removed from the auto. > But I don't think it is required now. The patch needs some discussion so the ETA when it'll land in some branch is unclear, removing it from auto still makes sense.
diff --git a/tests/btrfs/198 b/tests/btrfs/198 deleted file mode 100755 index 2df075e27134..000000000000 --- a/tests/btrfs/198 +++ /dev/null @@ -1,77 +0,0 @@ -#! /bin/bash -# SPDX-License-Identifier: GPL-2.0 -# Copyright (c) 2019 Oracle. All Rights Reserved. -# -# FS QA Test 198 -# -# Test stale and alien non-btrfs device in the fs devices list. -# Bug fixed in: -# btrfs: remove identified alien device in open_fs_devices -# -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.* -} - -# get standard environment, filters and checks -. ./common/rc -. ./common/filter -. ./common/filter.btrfs - -# remove previous $seqres.full before test -rm -f $seqres.full - -# real QA test starts here -_supported_fs btrfs -_supported_os Linux -_require_command "$WIPEFS_PROG" wipefs -_require_scratch -_require_scratch_dev_pool 4 - -workout() -{ - raid=$1 - device_nr=$2 - - echo $raid - _scratch_dev_pool_get $device_nr - - _scratch_pool_mkfs "-d$raid -m$raid" >> $seqres.full 2>&1 || \ - _fail "mkfs failed" - - # Make device_1 a free btrfs device for the raid created above by - # clearing its superblock - - # don't test with the first device as auto fs check (_check_scratch_fs) - # picks the first device - device_1=$(echo $SCRATCH_DEV_POOL | awk '{print $2}') - $WIPEFS_PROG -a $device_1 >> $seqres.full 2>&1 - - device_2=$(echo $SCRATCH_DEV_POOL | awk '{print $1}') - _mount -o degraded $device_2 $SCRATCH_MNT - # Check if missing device is reported as in the 196.out - $BTRFS_UTIL_PROG filesystem show -m $SCRATCH_MNT | \ - _filter_btrfs_filesystem_show - - _scratch_unmount - _scratch_dev_pool_put -} - -workout "raid1" "2" -workout "raid5" "3" -workout "raid6" "4" -workout "raid10" "4" - -# success, all done -status=0 -exit diff --git a/tests/btrfs/group b/tests/btrfs/group index 1b5fa695a9f7..d5330cd85cd5 100644 --- a/tests/btrfs/group +++ b/tests/btrfs/group @@ -199,7 +199,6 @@ 195 auto volume balance 196 auto metadata log volume 197 auto quick volume -198 auto quick volume 199 auto quick trim 200 auto quick send clone 201 auto quick punch log
btrfs/198 is supposed to be a test for the patch "btrfs: remove identified alien device in open_fs_devices" but this patch was never merged in btrfs. Remove the test from fstests as it is constantly failing. Signed-off-by: Johannes Thumshirn <johannes.thumshirn@wdc.com> --- tests/btrfs/198 | 77 ----------------------------------------------- tests/btrfs/group | 1 - 2 files changed, 78 deletions(-) delete mode 100755 tests/btrfs/198