From: Eryu Guan <eguan@redhat.com>
To: "zhangyi (F)" <yi.zhang@huawei.com>
Cc: fstests@vger.kernel.org, linux-unionfs@vger.kernel.org,
miklos@szeredi.hu, amir73il@gmail.com, miaoxie@huawei.com,
yangerkun@huawei.com
Subject: Re: [xfstests PATCH v2 2/4] overlay: add fsck.overlay whiteout test
Date: Tue, 6 Feb 2018 12:22:38 +0800 [thread overview]
Message-ID: <20180206042238.GO18267@eguan.usersys.redhat.com> (raw)
In-Reply-To: <20180131102604.40701-3-yi.zhang@huawei.com>
On Wed, Jan 31, 2018 at 06:26:02PM +0800, zhangyi (F) wrote:
> Add fsck.overlay test case to test it how to deal with orphan/valid
> whiteouts in underlying directories of overlayfs.
>
> Signed-off-by: zhangyi (F) <yi.zhang@huawei.com>
> Reviewed-by: Amir Goldstein <amir73il@gmail.com>
So sorry for the late review! All four patches look fine to me overall,
just some minor issues inline.
> ---
> tests/overlay/201 | 232 ++++++++++++++++++++++++++++++++++++++++++++++++++
> tests/overlay/201.out | 10 +++
> tests/overlay/group | 1 +
> 3 files changed, 243 insertions(+)
> create mode 100755 tests/overlay/201
> create mode 100644 tests/overlay/201.out
>
> diff --git a/tests/overlay/201 b/tests/overlay/201
> new file mode 100755
> index 0000000..219fad4
> --- /dev/null
> +++ b/tests/overlay/201
> @@ -0,0 +1,232 @@
> +#! /bin/bash
> +# FS QA Test 201
> +#
> +# Test fsck.overlay how to deal with whiteouts in overlayfs.
> +#
> +#-----------------------------------------------------------------------
> +# Copyright (c) 2017 Huawei. All Rights Reserved.
^^^^ 2018? Other tests also have 2017.
> +# Author: zhangyi (F) <yi.zhang@huawei.com>
> +#
> +# 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
> +
> +# remove previous $seqres.full before test
> +rm -f $seqres.full
> +
> +# real QA test starts here
> +_supported_fs overlay
> +_supported_os Linux
> +_require_scratch_nocheck
> +_require_command "$FSCK_OVERLAY_PROG" fsck.overlay
I noticed that you sourced common/attr and added _require_attrs in patch
3/4, but not 2/4 and 4/4, maybe we need it in all three tests.
> +
> +# remove all files from previous tests
> +_scratch_mkfs
> +
> +OVL_REDIRECT_XATTR="trusted.overlay.redirect"
> +OVL_OPAQUE_XATTR="trusted.overlay.opaque"
I think we can put these common ovl xattr keys (along with
OVL_IMPURE_XATTR)to common/overlay
> +OVL_OPAQUE_XATTR_VAL="y"
> +
> +# Check whiteout
> +check_whiteout()
> +{
> + for arg in $*
> + do
We prefer the format in fstests as
for arg in $*; do
...
done
> + local ttype=`stat -c "%F:%t,%T" $arg`
> +
> + [[ $ttype == "character special file:0,0" ]] || \
I'd quote $ttype (and other check_* functions in other tests) in case
stat returned nothing, then it runs into a syntax error
[[ == "character special file:0,0" ]] # bash syntax error
and with $ttype quoted
[[ "" == "character special file:0,0" ]]
it's a valid test.
> + echo "Valid whiteout removed incorrectly"
> + done
> +}
> +
> +# Create a whiteout
> +make_whiteout()
> +{
> + for arg in $*
> + do
> + mknod $arg c 0 0
> + done
> +}
> +
> +# Create an opaque directory
> +make_opaque_dir()
> +{
> + local target=$1
> +
> + mkdir -p $target
> + $SETFATTR_PROG -n $OVL_OPAQUE_XATTR -v $OVL_OPAQUE_XATTR_VAL $target
> +}
> +
> +# Create a redirect directory
> +make_redirect_dir()
> +{
> + local target=$1
> + local value=$2
> +
> + mkdir -p $target
> + $SETFATTR_PROG -n $OVL_REDIRECT_XATTR -v $value $target
> +}
> +
> +# Create test directories
> +lowerdir=$OVL_BASE_SCRATCH_MNT/lower
> +lowerdir2=$OVL_BASE_SCRATCH_MNT/lower2
> +upperdir=$OVL_BASE_SCRATCH_MNT/upper
> +workdir=$OVL_BASE_SCRATCH_MNT/workdir
> +
> +make_test_dirs()
> +{
> + rm -rf $lowerdir $lowerdir2 $upperdir $workdir
> + mkdir -p $lowerdir $lowerdir2 $upperdir $workdir
> +}
> +
> +# Test orphan whiteout in lower and upper layer, should remove
> +echo "+ Orphan whiteout"
> +make_test_dirs
> +make_whiteout $lowerdir/foo $upperdir/{foo,bar}
> +
> +_overlay_fsck_dirs $lowerdir $upperdir $workdir -p >> $seqres.full 2>&1 || \
> + _fail "fsck should not fail"
No need to _fail, we could let test to continue, so just echo. The same
is true for all other tests.
> +ls $lowerdir
> +ls $upperdir
> +
> +# Test valid whiteout covering lower target, should not remove
> +echo "+ Valid whiteout"
> +make_test_dirs
> +touch $lowerdir2/{foo,bar}
> +make_whiteout $upperdir/foo $lowerdir/bar
> +
> +_overlay_fsck_dirs "$lowerdir:$lowerdir2" $upperdir $workdir -p >> \
> + $seqres.full 2>&1 || _fail "fsck should not fail"
> +check_whiteout $upperdir/foo $lowerdir/bar
> +
> +# Test orphan whiteout in opaque directory, should remove
> +echo "+ Orphan whiteout(2)"
> +make_test_dirs
> +mkdir $lowerdir/testdir
> +touch $lowerdir/testdir/foo
> +make_opaque_dir $upperdir/testdir
> +make_whiteout $upperdir/testdir/foo
> +
> +_overlay_fsck_dirs $lowerdir $upperdir $workdir -p >> $seqres.full 2>&1 || \
> + _fail "fsck should not fail"
> +ls $upperdir/testdir
> +
> +# Test orphan whiteout whose parent path is not an merged directory,
> +# should remove
> +echo "+ Orphan whiteout(3)"
> +make_test_dirs
> +mkdir $lowerdir2/{testdir1,testdir2,testdir3}
> +touch $lowerdir2/{testdir1/foo,testdir2/foo,testdir3/foo}
> +mkdir $upperdir/{testdir1,testdir2,testdir3,testdir4}
> +touch $lowerdir/testdir1
> +make_whiteout $lowerdir/testdir2
> +make_opaque_dir $lowerdir/testdir3
> +make_whiteout $upperdir/{testdir1/foo,/testdir2/foo,testdir3/foo,testdir4/foo}
> +
> +_overlay_fsck_dirs "$lowerdir:$lowerdir2" $upperdir $workdir -p >> \
> + $seqres.full 2>&1 || _fail "fsck should not fail"
> +ls $upperdir/testdir1
> +ls $upperdir/testdir2
> +ls $upperdir/testdir3
> +ls $upperdir/testdir4
> +
> +# Test orphan whiteout in redirect directory, should remove
> +echo "+ Orphan whiteout(4)"
> +make_test_dirs
> +mkdir $lowerdir/{testdir,origin}
> +touch $lowerdir/testdir/foo
> +make_redirect_dir $upperdir/testdir "origin"
> +make_whiteout $upperdir/testdir/foo
> +
> +_overlay_fsck_dirs $lowerdir $upperdir $workdir -p >> $seqres.full 2>&1 || \
> + _fail "fsck should not fail"
> +ls $upperdir/testdir
> +
> +# Test valid whiteout in redirect directory cover file in lower
> +# redirect origin directory, should not remove
> +echo "+ Valid whiteout(2)"
> +make_test_dirs
> +mkdir $lowerdir/origin
> +touch $lowerdir/origin/foo
> +make_redirect_dir $upperdir/testdir "origin"
> +make_whiteout $upperdir/testdir/foo
> +
> +_overlay_fsck_dirs $lowerdir $upperdir $workdir -p >> $seqres.full 2>&1 || \
> + _fail "fsck should not fail"
> +check_whiteout $upperdir/testdir/foo
> +
> +# Test valid whiteout covering lower target whose parent directory
> +# merge with a redirect directory in the middle layer, should not remove.
> +echo "+ Valid whiteout(3)"
> +make_test_dirs
> +mkdir -p $lowerdir2/origin/subdir
> +touch $lowerdir2/origin/subdir/foo
> +make_redirect_dir $lowerdir/testdir "origin"
> +mkdir -p $upperdir/testdir/subdir
> +make_whiteout $upperdir/testdir/subdir/foo
> +
> +_overlay_fsck_dirs "$lowerdir:$lowerdir2" $upperdir $workdir -p \
> + >> $seqres.full 2>&1 || _fail "fsck should not fail"
> +check_whiteout $upperdir/testdir/subdir/foo
> +
> +# Test invalid whiteout in opaque subdirectory in a redirect directory,
> +# should remove
> +echo "+ Orphan whiteout(5)"
> +make_test_dirs
> +mkdir -p $lowerdir/origin/subdir
> +touch $lowerdir/origin/subdir/foo
> +make_redirect_dir $upperdir/testdir "origin"
> +make_opaque_dir $upperdir/testdir/subdir
> +make_whiteout $upperdir/testdir/subdir/foo
> +
> +_overlay_fsck_dirs $lowerdir $upperdir $workdir -p >> $seqres.full 2>&1 || \
> + _fail "fsck should not fail"
> +ls $upperdir/testdir/subdir
> +
> +# Test valid whiteout in reidrect subdirectory in a opaque directory
> +# covering lower target, should not remove
> +echo "+ Valid whiteout(4)"
> +make_test_dirs
> +mkdir $lowerdir/origin
> +touch $lowerdir/origin/foo
> +make_opaque_dir $upperdir/testdir
> +make_redirect_dir $upperdir/testdir/subdir "/origin"
> +make_whiteout $upperdir/testdir/subdir/foo
> +
> +_overlay_fsck_dirs $lowerdir $upperdir $workdir -p >> $seqres.full 2>&1 || \
> + _fail "fsck should not fail"
> +check_whiteout $upperdir/testdir/subdir/foo
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/overlay/201.out b/tests/overlay/201.out
> new file mode 100644
> index 0000000..157bb85
> --- /dev/null
> +++ b/tests/overlay/201.out
> @@ -0,0 +1,10 @@
> +QA output created by 201
> ++ Orphan whiteout
> ++ Valid whiteout
> ++ Orphan whiteout(2)
> ++ Orphan whiteout(3)
> ++ Orphan whiteout(4)
> ++ Valid whiteout(2)
> ++ Valid whiteout(3)
> ++ Orphan whiteout(5)
> ++ Valid whiteout(4)
> diff --git a/tests/overlay/group b/tests/overlay/group
> index 7e541e4..7c5fcbb 100644
> --- a/tests/overlay/group
> +++ b/tests/overlay/group
> @@ -49,3 +49,4 @@
> 044 auto quick copyup hardlink nonsamefs
> 047 auto quick copyup hardlink
> 048 auto quick copyup hardlink
> +201 auto quick fsck
It'd be great if you could help rebase the patchset against latest
master in next version. Thanks a lot!
All the comments here apply to the other two tests.
Thanks,
Eryu
next prev parent reply other threads:[~2018-02-06 4:22 UTC|newest]
Thread overview: 8+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-01-31 10:26 [xfstests PATCH v2 0/4] overlay: add fsck.overlay basic tests zhangyi (F)
2018-01-31 10:25 ` Eryu Guan
2018-01-31 10:26 ` [xfstests PATCH v2 1/4] overlay: add filesystem check helper zhangyi (F)
2018-01-31 10:26 ` [xfstests PATCH v2 2/4] overlay: add fsck.overlay whiteout test zhangyi (F)
2018-02-06 4:22 ` Eryu Guan [this message]
2018-02-07 4:13 ` zhangyi (F)
2018-01-31 10:26 ` [xfstests PATCH v2 3/4] overlay: add fsck.overlay redirect directory test zhangyi (F)
2018-01-31 10:26 ` [xfstests PATCH v2 4/4] overlay: add fsck.overlay impure xattr test zhangyi (F)
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=20180206042238.GO18267@eguan.usersys.redhat.com \
--to=eguan@redhat.com \
--cc=amir73il@gmail.com \
--cc=fstests@vger.kernel.org \
--cc=linux-unionfs@vger.kernel.org \
--cc=miaoxie@huawei.com \
--cc=miklos@szeredi.hu \
--cc=yangerkun@huawei.com \
--cc=yi.zhang@huawei.com \
/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