From: Eryu Guan <eguan@redhat.com>
To: Amir Goldstein <amir73il@gmail.com>
Cc: Josef Bacik <jbacik@fb.com>,
"Darrick J . Wong" <darrick.wong@oracle.com>,
Christoph Hellwig <hch@lst.de>,
fstests@vger.kernel.org, linux-xfs@vger.kernel.org
Subject: Re: [PATCH v2 14/14] fstests: add crash consistency fsx test using dm-log-writes
Date: Tue, 5 Sep 2017 19:28:42 +0800 [thread overview]
Message-ID: <20170905112842.GF8034@eguan.usersys.redhat.com> (raw)
In-Reply-To: <1504104706-11965-15-git-send-email-amir73il@gmail.com>
On Wed, Aug 30, 2017 at 05:51:46PM +0300, Amir Goldstein wrote:
> DO NOT MERGE!!! this test fails most likely due to test bug.
>
> The random seed values in this patch fail the test consistently on ext4
> always with the same fsck error ((end of extent exceeds allowed value).
> btrfs also fails, but with slightly different fsck errors each run.
> xfs fails sometimes on file checksum error.
>
> Cherry-picked the test from commit 70d41e17164b
> in Josef Bacik's fstests tree (https://github.com/josefbacik/fstests).
> Quoting from Josef's commit message:
>
> The test just runs some ops and exits, then finds all of the good buffers
> in the directory we provided and:
> - replays up to the mark given
> - mounts the file system and compares the md5sum
> - unmounts and fsck's to check for metadata integrity
>
> dm-log-writes will pretend to do discard and the replay-log tool will
> replay it properly depending on the underlying device, either by writing
> 0's or actually calling the discard ioctl, so I've enabled discard in the
> test for maximum fun.
>
> [Amir:]
> - Removed unneeded _test_falloc_support dynamic FSX_OPTS
> - Added place holders for using constant random seeds
> - Add test to new 'replay' group
Perhaps replace it with 'log' group?
>
> Cc: Josef Bacik <jbacik@fb.com>
> Signed-off-by: Amir Goldstein <amir73il@gmail.com>
> ---
> tests/generic/500 | 138 ++++++++++++++++++++++++++++++++++++++++++++++++++
> tests/generic/500.out | 2 +
> tests/generic/group | 1 +
> 3 files changed, 141 insertions(+)
> create mode 100755 tests/generic/500
> create mode 100644 tests/generic/500.out
>
> diff --git a/tests/generic/500 b/tests/generic/500
> new file mode 100755
> index 0000000..81d45ef
> --- /dev/null
> +++ b/tests/generic/500
> @@ -0,0 +1,138 @@
> +#! /bin/bash
> +# FS QA Test No. 500
> +#
> +# Run fsx with log writes to verify power fail safeness.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2015 Facebook. 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`
> +status=1 # failure is the default!
> +
> +_cleanup()
> +{
> + _cleanup_log_writes
> +}
> +trap "_cleanup; exit \$status" 0 1 2 3 15
> +
> +# get standard environment, filters and checks
> +. ./common/rc
> +. ./common/filter
> +. ./common/dmlogwrites
> +
> +# real QA test starts here
> +_supported_fs generic
> +_supported_os Linux
> +_require_test
> +_require_scratch_nocheck
> +_require_log_writes
> +
> +rm -f $seqres.full
> +rm -rf $TEST_DIR/fsxtests
Remove it after SANITY_DIR is defined?
> +
> +check_files()
> +{
> + local _name=$1
Don't need underscore for a local var.
> +
> + # Now look for our files
> + for i in $(find $SANITY_DIR -type f | grep $_name | grep mark)
> + do
> + local filename=$(basename $i)
> + local mark="${filename##*.}"
> + local expected_size=$(_ls_l -h $i | awk '{ print $5 }')
> + echo "checking $filename ($expected_size)" >> $seqres.full
> + _replay_log $filename
> + _scratch_mount
> + local expected_md5=$(md5sum $i | cut -f 1 -d ' ')
> + local md5=$(md5sum $SCRATCH_MNT/$_name | cut -f 1 -d ' ')
> + local size=$(_ls_l -h $SCRATCH_MNT/$_name | awk '{ print $5 }')
I see this md5 and size patterns are repeating several times in this
test, write new helpers?
> + [ "${md5}" != "${expected_md5}" ] && _fatal "$filename ($size) md5sum mismatched"
Use _fail in this test.
> + _scratch_unmount
> + _check_scratch_fs
> + done
> +}
> +
> +SANITY_DIR=$TEST_DIR/fsxtests
> +mkdir $SANITY_DIR
> +
> +# Create the log
> +_init_log_writes
> +
> +_log_writes_mkfs >> $seqres.full 2>&1
> +
> +# Log writes emulates discard support, turn it on for maximum crying.
> +_mount_log_writes -o discard
> +
> +NUM_FILES=4
> +NUM_OPS=200
> +FSX_OPTS="-N $NUM_OPS -d -P $SANITY_DIR -i $LOGWRITES_DMDEV"
> +# Set random seeds for fsx runs (0 for timestamp + pid)
> +seeds=(- 2885 2886 2887 2888)
^^ meant 0?
> +# Run fsx for a while
> +for j in `seq 1 $NUM_FILES`
> +do
> + run_check $here/ltp/fsx $FSX_OPTS -S ${seeds[$j]} -j $j $SCRATCH_MNT/testfile$j &
> +done
> +wait
> +
> +test_md5=()
> +test_size=()
> +for j in `seq 1 $NUM_FILES`
> +do
> + test_md5[$j]=$(md5sum $SCRATCH_MNT/testfile$j | cut -f 1 -d ' ')
> + test_size[$j]=$(_ls_l -h $SCRATCH_MNT/testfile$j | awk '{ print $5 }')
> +done
> +
> +# Unmount the scratch dir and tear down the log writes target
> +_log_writes_mark last
> +_unmount_log_writes
> +_log_writes_mark end
> +_log_writes_remove
> +_check_scratch_fs
> +
> +# check pre umount
> +_replay_log last
> +_scratch_mount
> +_scratch_unmount
> +_check_scratch_fs
> +
> +for j in `seq 1 $NUM_FILES`
> +do
> + check_files testfile$j
> +done
> +
> +# Check the end
> +_replay_log end
> +_scratch_mount
> +for j in `seq 1 $NUM_FILES`
> +do
> + md5=$(md5sum $SCRATCH_MNT/testfile$j | cut -f 1 -d ' ')
> + size=$(_ls_l -h $SCRATCH_MNT/testfile$j | awk '{ print $5 }')
> + [ "${md5}" != "${test_md5[$j]}" ] && _fatal "testfile$j end md5sum mismatched ($size:${test_size[$j]})"
> +done
> +_scratch_unmount
> +_check_scratch_fs
> +
> +echo "Silence is golden"
> +status=0
> +exit
> +
> diff --git a/tests/generic/500.out b/tests/generic/500.out
> new file mode 100644
> index 0000000..883b2ca
> --- /dev/null
> +++ b/tests/generic/500.out
> @@ -0,0 +1,2 @@
> +QA output created by 500
> +Silence is golden
> diff --git a/tests/generic/group b/tests/generic/group
> index 044ec3f..2396b72 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
> +500 auto log replay
Adding it to auto group seems fine to me.
Thanks,
Eryu
> --
> 2.7.4
>
next prev parent reply other threads:[~2017-09-05 11:28 UTC|newest]
Thread overview: 48+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-08-30 14:51 [PATCH v2 00/14] Crash consistency xfstest using dm-log-writes Amir Goldstein
2017-08-30 14:51 ` [PATCH v2 01/14] common/rc: convert some egrep to grep Amir Goldstein
2017-08-30 15:45 ` Darrick J. Wong
2017-08-30 14:51 ` [PATCH v2 02/14] common/rc: fix _require_xfs_io_command params check Amir Goldstein
2017-08-30 16:17 ` Darrick J. Wong
2017-08-30 14:51 ` [PATCH v2 03/14] fsx: fixes to random seed Amir Goldstein
2017-08-30 14:51 ` [PATCH v2 04/14] fsx: fix path of .fsx* files Amir Goldstein
2017-08-30 14:51 ` [PATCH v2 05/14] fsx: fix compile warnings Amir Goldstein
2017-08-30 14:51 ` [PATCH v2 06/14] fsx: add support for integrity check with dm-log-writes target Amir Goldstein
2017-08-30 14:51 ` [PATCH v2 07/14] fsx: add optional logid prefix to log messages Amir Goldstein
2017-09-05 10:46 ` Eryu Guan
2017-09-05 11:24 ` Amir Goldstein
2017-09-05 11:31 ` Eryu Guan
2017-09-07 7:10 ` Amir Goldstein
2017-08-30 14:51 ` [PATCH v2 08/14] fsx: add support for --record-ops Amir Goldstein
2017-08-30 14:51 ` [PATCH v2 09/14] fsx: add support for -g filldata Amir Goldstein
2017-09-05 10:50 ` Eryu Guan
2017-09-05 11:29 ` Amir Goldstein
2017-09-05 11:33 ` Eryu Guan
2017-08-30 14:51 ` [PATCH v2 10/14] log-writes: add replay-log program to replay dm-log-writes target Amir Goldstein
2017-09-05 11:03 ` Eryu Guan
2017-09-05 13:40 ` Amir Goldstein
2017-08-30 14:51 ` [PATCH v2 11/14] replay-log: output log replay offset in verbose mode Amir Goldstein
2017-08-30 14:51 ` [PATCH v2 12/14] replay-log: add support for replaying ops in target device sector range Amir Goldstein
2017-09-05 11:07 ` Eryu Guan
2017-09-05 11:41 ` Amir Goldstein
2017-08-30 14:51 ` [PATCH v2 13/14] fstests: add support for working with dm-log-writes target Amir Goldstein
2017-09-05 11:22 ` Eryu Guan
2017-09-05 15:15 ` Amir Goldstein
2017-08-30 14:51 ` [PATCH v2 14/14] fstests: add crash consistency fsx test using dm-log-writes Amir Goldstein
2017-09-05 11:28 ` Eryu Guan [this message]
2017-09-05 11:52 ` Amir Goldstein
2017-08-30 15:04 ` [PATCH v2 00/14] Crash consistency xfstest " Amir Goldstein
2017-08-30 15:23 ` Josef Bacik
2017-08-30 18:39 ` Amir Goldstein
2017-08-30 18:55 ` Josef Bacik
2017-08-30 19:43 ` Amir Goldstein
[not found] ` <CAOQ4uxjt-zZ7_iE7ZYUcp8qWYUH=aDLSum70Dmbnth-5smFQ+A@mail.gmail.com>
[not found] ` <20170831134320.lnyu4jibsm3amuk7@destiny>
[not found] ` <CAOQ4uxhgOYDfRxZ74RNd=omOMHxF2MgP+wLe0O6HO7+emnrMfA@mail.gmail.com>
[not found] ` <20170831205403.2tene34ccvw55yo7@destiny>
2017-09-01 6:52 ` Amir Goldstein
2017-09-01 7:03 ` Josef Bacik
2017-09-01 20:07 ` Josef Bacik
2017-09-03 13:39 ` Amir Goldstein
2017-09-04 6:42 ` Dave Chinner
2017-09-04 6:49 ` Amir Goldstein
2018-05-25 8:58 ` Amir Goldstein
2017-08-31 3:38 ` Eryu Guan
2017-08-31 4:29 ` Amir Goldstein
2017-09-01 7:29 ` Amir Goldstein
2017-09-01 7:45 ` Eryu Guan
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=20170905112842.GF8034@eguan.usersys.redhat.com \
--to=eguan@redhat.com \
--cc=amir73il@gmail.com \
--cc=darrick.wong@oracle.com \
--cc=fstests@vger.kernel.org \
--cc=hch@lst.de \
--cc=jbacik@fb.com \
--cc=linux-xfs@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