Re: [PATCH] btrfs: test if receive with qgroups corrupts metadata

[Eryu Guan <eguan@redhat.com>]

On Thu, Jul 20, 2017 at 04:03:33PM -0700, Justin Maggard wrote:
> This test case does some concurrent send/receives with qgroups enabled.
> Currently (4.13-rc1) this usually results in btrfs check errors, and
> often also results in a WARN_ON in record_root_in_trans().
> 
> Bisecting points to 6426c7ad697d (btrfs: qgroup: Fix qgroup accounting
> when creating snapshot) as the culprit.

Thanks for the new test! But I'd need some help from btrfs-list to
review if this is a sane test for btrfs.

BTW, you're missing a S-O-B line here :)

> ---
>  tests/btrfs/149     | 101 ++++++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/btrfs/149.out |  17 +++++++++
>  tests/btrfs/group   |   1 +
>  3 files changed, 119 insertions(+)
>  create mode 100755 tests/btrfs/149
>  create mode 100644 tests/btrfs/149.out
> 
> diff --git a/tests/btrfs/149 b/tests/btrfs/149
> new file mode 100755
> index 0000000..5c1912d
> --- /dev/null
> +++ b/tests/btrfs/149
> @@ -0,0 +1,101 @@
> +#! /bin/bash
> +# FS QA Test No. btrfs/149
> +#
> +# Test that incremental send/receive operations don't corrupt metadata when
> +# qgroups are enabled.
> +#
> +#-----------------------------------------------------------------------
> +#
> +# Copyright (c) 2017 NETGEAR, 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
> +#-----------------------------------------------------------------------
> +#
> +
> +seq=`basename $0`
> +seqres=$RESULT_DIR/$seq
> +echo "QA output created by $seq"
> +
> +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 btrfs
> +_supported_os Linux
> +_require_scratch
> +
> +rm -f $seqres.full
> +
> +_scratch_mkfs >>$seqres.full 2>&1
> +_scratch_mount
> +
> +# Enable quotas
> +$BTRFS_UTIL_PROG quota enable $SCRATCH_MNT
> +
> +# Create 2 source and 4 destination subvolumes
> +for subvol in subvol1 subvol2 recv1_1 recv1_2 recv2_1 recv2_2; do
> +	$BTRFS_UTIL_PROG subvolume create $SCRATCH_MNT/$subvol

The only thing I've noticed is that you need to _filter_scratch here and
some other places, so in .out file you don't assume and hardcode
/mnt/scratch as SCRATCH_MNT.

Thanks,
Eryu

> +done
> +mkdir $SCRATCH_MNT/subvol{1,2}/.snapshots
> +
> +# Create base snapshots and send them
> +$BTRFS_UTIL_PROG subvolume snapshot -r $SCRATCH_MNT/subvol1 \
> +	$SCRATCH_MNT/subvol1/.snapshots/1
> +$BTRFS_UTIL_PROG subvolume snapshot -r $SCRATCH_MNT/subvol2 \
> +	$SCRATCH_MNT/subvol2/.snapshots/1
> +for recv in recv1_1 recv1_2 recv2_1 recv2_2; do
> +	$BTRFS_UTIL_PROG send $SCRATCH_MNT/subvol1/.snapshots/1 | \
> +		$BTRFS_UTIL_PROG receive $SCRATCH_MNT/${recv}
> +done
> +
> +# Now do 10 loops of concurrent incremental send/receives
> +for i in `seq 1 10`; do
> +	prev=$i
> +	curr=$((i+1))
> +
> +	$BTRFS_UTIL_PROG subvolume snapshot -r $SCRATCH_MNT/subvol1 \
> +		$SCRATCH_MNT/subvol1/.snapshots/${curr} > /dev/null
> +	($BTRFS_UTIL_PROG send -p $SCRATCH_MNT/subvol1/.snapshots/${prev} \
> +		$SCRATCH_MNT/subvol1/.snapshots/${curr} 2> /dev/null | \
> +		$BTRFS_UTIL_PROG receive $SCRATCH_MNT/recv1_1) > /dev/null &
> +	($BTRFS_UTIL_PROG send -p $SCRATCH_MNT/subvol1/.snapshots/${prev} \
> +		$SCRATCH_MNT/subvol1/.snapshots/${curr} 2> /dev/null | \
> +		$BTRFS_UTIL_PROG receive $SCRATCH_MNT/recv1_2) > /dev/null &
> +
> +	$BTRFS_UTIL_PROG subvolume snapshot -r $SCRATCH_MNT/subvol2 \
> +		$SCRATCH_MNT/subvol2/.snapshots/${curr} > /dev/null
> +	($BTRFS_UTIL_PROG send -p $SCRATCH_MNT/subvol2/.snapshots/${prev} \
> +		$SCRATCH_MNT/subvol2/.snapshots/${curr} 2> /dev/null | \
> +		$BTRFS_UTIL_PROG receive $SCRATCH_MNT/recv2_1) > /dev/null &
> +	($BTRFS_UTIL_PROG send -p $SCRATCH_MNT/subvol2/.snapshots/${prev} \
> +		$SCRATCH_MNT/subvol2/.snapshots/${curr} 2> /dev/null | \
> +		$BTRFS_UTIL_PROG receive $SCRATCH_MNT/recv2_2) > /dev/null &
> +	wait
> +done
> +
> +_scratch_unmount
> +
> +status=0
> +exit
> diff --git a/tests/btrfs/149.out b/tests/btrfs/149.out
> new file mode 100644
> index 0000000..3ea9101
> --- /dev/null
> +++ b/tests/btrfs/149.out
> @@ -0,0 +1,17 @@
> +QA output created by 149
> +Create subvolume '/mnt/scratch/subvol1'
> +Create subvolume '/mnt/scratch/subvol2'
> +Create subvolume '/mnt/scratch/recv1_1'
> +Create subvolume '/mnt/scratch/recv1_2'
> +Create subvolume '/mnt/scratch/recv2_1'
> +Create subvolume '/mnt/scratch/recv2_2'
> +Create a readonly snapshot of '/mnt/scratch/subvol1' in '/mnt/scratch/subvol1/.snapshots/1'
> +Create a readonly snapshot of '/mnt/scratch/subvol2' in '/mnt/scratch/subvol2/.snapshots/1'
> +At subvol /mnt/scratch/subvol1/.snapshots/1
> +At subvol 1
> +At subvol /mnt/scratch/subvol1/.snapshots/1
> +At subvol 1
> +At subvol /mnt/scratch/subvol1/.snapshots/1
> +At subvol 1
> +At subvol /mnt/scratch/subvol1/.snapshots/1
> +At subvol 1
> diff --git a/tests/btrfs/group b/tests/btrfs/group
> index 8240b53..a84a2bd 100644
> --- a/tests/btrfs/group
> +++ b/tests/btrfs/group
> @@ -151,3 +151,4 @@
>  146 auto quick
>  147 auto quick send
>  148 auto quick rw
> +149 auto quick metadata qgroup send
> -- 
> 2.1.4
> 
> --
> 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