public inbox for fstests@vger.kernel.org
 help / color / mirror / Atom feed
From: Qu Wenruo <quwenruo.btrfs@gmx.com>
To: Justin Maggard <jmaggard10@gmail.com>,
	fstests@vger.kernel.org, linux-btrfs@vger.kernel.org
Cc: Justin Maggard <jmaggard@netgear.com>
Subject: Re: [PATCH] btrfs: test if receive with qgroups corrupts metadata
Date: Fri, 21 Jul 2017 07:44:52 +0800	[thread overview]
Message-ID: <73a8c8a4-a9b3-18d5-35cd-9fbc7e4a1429@gmx.com> (raw)
In-Reply-To: <20170720230333.15930-1-jmaggard@netgear.com>



On 2017年07月21日 07:03, 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().

This is quite interesting.
Despite subvolume/snapshot creation, receiving is done mostly by doing 
normal read/write/utime/... operations, which are handled quite well.

BTW, the test case is doing send/receive concurrently, is the bug still 
reproducible without concurrency ?

Thanks,
Qu

> 
> Bisecting points to 6426c7ad697d (btrfs: qgroup: Fix qgroup accounting
> when creating snapshot) as the culprit.
> ---
>   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
> +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
> 

  reply	other threads:[~2017-07-20 23:45 UTC|newest]

Thread overview: 7+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-07-20 23:03 [PATCH] btrfs: test if receive with qgroups corrupts metadata Justin Maggard
2017-07-20 23:44 ` Qu Wenruo [this message]
2017-07-21  1:24   ` Justin Maggard
2017-07-21  6:52 ` Eryu Guan
  -- strict thread matches above, loose matches on Subject: below --
2017-10-30 22:32 Justin Maggard
2017-10-31  0:21 ` Qu Wenruo
2017-10-31  0:25   ` Justin Maggard

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=73a8c8a4-a9b3-18d5-35cd-9fbc7e4a1429@gmx.com \
    --to=quwenruo.btrfs@gmx.com \
    --cc=fstests@vger.kernel.org \
    --cc=jmaggard10@gmail.com \
    --cc=jmaggard@netgear.com \
    --cc=linux-btrfs@vger.kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox