Re: [PATCH] generic: test race between block map change and writeback

[Brian Foster <bfoster@redhat.com>]

On Thu, Aug 31, 2017 at 12:02:37PM +0800, Eryu Guan wrote:
> Run delalloc writes & append writes & non-data-integrity syncs
> concurrently to test the race between block map change vs writeback.
> 
> This is to cover an XFS bug that data could be written to wrong
> block and delay allocated blocks are leaked because the block map
> was changed due to the removal of speculative allocated eofblocks
> when writeback is in progress.
> 
> And this test partially mimics what lustre-racer[1] test does, using
> which this bug was first found.
> 
> [1] https://git.hpdd.intel.com/?p=fs/lustre-release.git;a=tree;f=lustre/tests/racer;hb=HEAD
> 
> Signed-off-by: Eryu Guan <eguan@redhat.com>
> ---
> 
> This may not reproduce the bug on all hosts, but it does reproduce the XFS
> corruption issue reliably on my different test hosts.
> 

Was this problem fixed already or are we still waiting on a fix?

>  tests/generic/451     | 130 ++++++++++++++++++++++++++++++++++++++++++++++++++
>  tests/generic/451.out |   2 +
>  tests/generic/group   |   1 +
>  3 files changed, 133 insertions(+)
>  create mode 100755 tests/generic/451
>  create mode 100644 tests/generic/451.out
> 
> diff --git a/tests/generic/451 b/tests/generic/451
> new file mode 100755
> index 000000000000..72cdd1c01de2
> --- /dev/null
> +++ b/tests/generic/451
> @@ -0,0 +1,130 @@
> +#! /bin/bash
> +# FS QA Test 451
> +#
> +# Run delalloc writes & append writes & non-data-integrity syncs concurrently
> +# to test the race between block map change vs writeback.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2017 Red Hat 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"
> +
> +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
> +
> +MAXFILES=200
> +BLOCK_SZ=65536
> +
> +LOOP_CNT=12

If I skip the failure detection below, the test runs for 100s on my vm.
Otherwise it fails consistently within ~45s (worst case in 5 or 6
tries). Do you observe differently? If not, I'm wondering if we could
speed up the common case and reduce the number of iterations.

> +LOOP_TIME=5
> +PROC_CNT=16
> +
> +stop=$tmp.stop
> +
> +# get a random file to work on
> +getfile()
> +{
> +	echo $SCRATCH_MNT/$((RANDOM % MAXFILES))
> +}
> +
> +# delalloc write a relative big file to get enough dirty pages to be written
> +# back, and XFS needs big enough file to trigger speculative preallocations, so
> +# freeing these eofblocks could change the extent record
> +do_write()
> +{
> +	local blockcount=$((RANDOM % 100))
> +	local filesize=$((blockcount * BLOCK_SZ))
> +	$XFS_IO_PROG -ftc "pwrite -b $BLOCK_SZ 0 $filesize" `getfile` >/dev/null 2>&1

Long line here. Otherwise the rest of the test looks good.

Brian

> +}
> +
> +# append another dirty page to the file, the writeback might pick it up too if
> +# the file is already under writeback
> +do_append()
> +{
> +	echo "test string" >> `getfile`
> +}
> +
> +# issue WB_SYNC_NONE writeback with the '-w' option of sync_range xfs_io
> +# command, so that the last dirty page from append write can be picked up in
> +# this writeback cycle. This is not mandatory but could help reproduce XFS
> +# corruption more easily.
> +do_writeback()
> +{
> +	$XFS_IO_PROG -c "sync_range -w 0 0" `getfile` >/dev/null 2>&1
> +}
> +
> +# remove previous $seqres.full before test
> +rm -f $seqres.full
> +
> +# real QA test starts here
> +_supported_fs generic
> +_supported_os Linux
> +# do fsck after each iteration in test
> +_require_scratch_nocheck
> +_require_xfs_io_command "sync_range"
> +
> +_scratch_mkfs >>$seqres.full 2>&1
> +_scratch_mount
> +
> +# loop for $LOOP_CNT iterations, and each iteration starts $PROC_CNT processes
> +# for each operation and runs for $LOOP_TIME seconds, and check filesystem
> +# consistency after each iteration
> +for i in `seq 1 $LOOP_CNT`; do
> +	rm -f $stop
> +	for j in `seq 1 $PROC_CNT`; do
> +		while [ ! -e $stop ]; do
> +			do_write
> +		done &
> +
> +		while [ ! -e $stop ]; do
> +			do_append
> +		done &
> +
> +		while [ ! -e $stop ]; do
> +			do_writeback
> +		done &
> +	done
> +	sleep $LOOP_TIME
> +	touch $stop
> +	wait
> +
> +	_scratch_unmount
> +	# test exits here if fs is inconsistent
> +	_check_scratch_fs
> +	_scratch_mount
> +done
> +
> +echo "Silence is golden"
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/generic/451.out b/tests/generic/451.out
> new file mode 100644
> index 000000000000..db924411b72f
> --- /dev/null
> +++ b/tests/generic/451.out
> @@ -0,0 +1,2 @@
> +QA output created by 451
> +Silence is golden
> diff --git a/tests/generic/group b/tests/generic/group
> index 044ec3f355ed..b4bd66bc65a9 100644
> --- a/tests/generic/group
> +++ b/tests/generic/group
> @@ -453,3 +453,4 @@
>  448 auto quick rw
>  449 auto quick acl enospc
>  450 auto quick rw
> +451 auto rw
> -- 
> 2.13.5
> 
> --
> 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