Nested overlay tests
v2 cover Nested overlay tests
Hi Eryu,
This v2 post includes two more nested overlay tests.
As described in previous post [1] the purpose of these tests is not so much to validate the setup of nested overlayfs, which is a less common setup, but to provide test coverage for some overlayfs fixes that were merge in v5.5-rc2 as well as some fixes that have been posted since.
Specifically, test overlay/071 exercises a "nested xino setup". This setup results in "xino bit overlflow", a corner case that did not have test coverage until now. This test fails on current upstream because of an overlayfs bug [2].
On top of that, both overlay/070 and overlay/071 fail on current upstream, because of a /proc/locks format bug [3].
Thanks,
Amir.
[1] https://lore.kernel.org/fstests/[email protected]/
[2] https://marc.info/?l=linux-unionfs&m=157708323410613&w=2
[3] https://lore.kernel.org/linux-fsdevel/[email protected]/
Amir Goldstein (5):
overlay: create the overlay/nested test group
overlay: test file handles with nested overlay over samefs lower
overlay: test file handles with nested overlay over non-samefs lower
overlay: test constant ino with nested overlay over samefs lower
overlay: test constant ino with nested overlay over non-samefs lower
tests/overlay/068 | 304 ++++++++++++++++++++++++++++++++++++++++ tests/overlay/068.out | 50 +++++++ tests/overlay/069 | 313 ++++++++++++++++++++++++++++++++++++++++++ tests/overlay/069.out | 50 +++++++ tests/overlay/070 | 222 ++++++++++++++++++++++++++++++ tests/overlay/070.out | 2 + tests/overlay/071 | 236 +++++++++++++++++++++++++++++++ tests/overlay/071.out | 2 + tests/overlay/group | 8 +- 9 files changed, 1185 insertions(+), 2 deletions(-) create mode 100755 tests/overlay/068 create mode 100644 tests/overlay/068.out create mode 100755 tests/overlay/069 create mode 100644 tests/overlay/069.out create mode 100755 tests/overlay/070 create mode 100644 tests/overlay/070.out create mode 100755 tests/overlay/071 create mode 100644 tests/overlay/071.out
Signature · 1 line
2.17.1
PATCH v2 1/5 overlay: create the overlay/nested test group
For tests that mount an overlayfs over overlayfs.
--- tests/overlay/group | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
tests/overlay/group +2 -2
@@ -24,14 +24,14 @@ 019 auto stress 020 auto quick copyup perms 021 auto quick copyup -022 auto quick mount +022 auto quick mount nested 023 auto quick attr 024 auto quick mount 025 auto quick attr 026 auto attr quick 027 auto quick perms 028 auto copyup quick -029 auto quick +029 auto quick nested 030 auto quick perms 031 auto quick whiteout 032 auto quick copyup hardlink
Signature · 1 line
2.17.1
PATCH v2 2/5 overlay: test file handles with nested overlay over samefs lower
In reply to Amir GoldsteinThis is a variant of overlay file handles test for an overlayfs that is nested over another lower overlayfs on the same fs.
--- tests/overlay/068 | 304 ++++++++++++++++++++++++++++++++++++++++++ tests/overlay/068.out | 50 +++++++ tests/overlay/group | 1 + 3 files changed, 355 insertions(+) create mode 100755 tests/overlay/068 create mode 100644 tests/overlay/068.out
tests/overlay/068 new file +304 -0
@@ -0,0 +1,304 @@ +#! /bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright (C) 2019 CTERA Networks. All Rights Reserved. +# +# FS QA Test No. 068 +# +# Test encode/decode of nested overlay file handles +# +# This is a variant of overlay file handles test for an overlayfs that is +# nested over another lower overlayfs on the same fs. +# +# - Check encode/write/decode/read content of lower/upper file handles +# - Check encode/decode/write/read content of lower/upper file handles +# - Check decode/read of unlinked lower/upper files and directories +# - Check decode/read of lower file handles after copy up, link and unlink +# - Check decode/read of lower file handles after rename of parent and self +# +# This test does not cover connectable file handles of non-directories, +# because name_to_handle_at() syscall does not support requesting connectable +# file handles. +# +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.* + # Unmount the nested overlay mount + $UMOUNT_PROG $mnt2 2>/dev/null +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter + +# real QA test starts here + +_supported_fs overlay +_supported_os Linux +_require_scratch +# _require_exportfs already requires open_by_handle, but let's not count on it +_require_test_program "open_by_handle" +# We need to require all features together, because nfs_export cannot +# be enabled when index is disabled +_require_scratch_overlay_features index nfs_export redirect_dir + +# Lower dir of nested overlay is the scratch overlay mount at SCRATCH_MNT +upper2=$OVL_BASE_SCRATCH_MNT/$OVL_UPPER.2 +work2=$OVL_BASE_SCRATCH_MNT/$OVL_WORK.2 +mnt2=$OVL_BASE_SCRATCH_MNT/$OVL_MNT.2 + +lowerdir=$OVL_BASE_SCRATCH_MNT/$OVL_LOWER/lowertestdir +upperdir=$OVL_BASE_SCRATCH_MNT/$OVL_UPPER/uppertestdir +lowertestdir=$mnt2/lowertestdir +uppertestdir=$mnt2/uppertestdir +NUMFILES=1 + +# Create test dir and empty test files +create_test_files() +{ + local dir=$1 + local opt=$2 + + src/open_by_handle -cp $opt $dir $NUMFILES +} + +# Test encode/decode file handles on overlay mount +test_file_handles() +{ + local dir=$1 + shift + + echo test_file_handles $dir $* | _filter_scratch | _filter_ovl_dirs | \ + sed -e "s,$tmp\.,,g" + $here/src/open_by_handle $* $dir $NUMFILES +} + +# Re-create lower/upper/work dirs +create_dirs() +{ + # Create the underlying overlay dirs + _scratch_mkfs + + # Create the nested overlay upper dirs + mkdir -p $upper2 $work2 $mnt2 +} + +# Mount a nested overlay with $SCRATCH_MNT as lower layer +mount_dirs() +{ + # Mount the underlying overlay + _scratch_mount -o "index=on,nfs_export=on,redirect_dir=on" + + # Mount the nested overlay + _overlay_mount_dirs $SCRATCH_MNT $upper2 $work2 overlay2 $mnt2 \ + -o "index=on,nfs_export=on,redirect_dir=on" 2>/dev/null || + _notrun "cannot mount nested overlay with nfs_export=on option" + _fs_options overlay2 | grep -q "nfs_export=on" || \ + _notrun "cannot enable nfs_export feature on nested overlay" +} + +# Unmount the nested overlay mount and check underlying overlay layers +unmount_dirs() +{ + # unmount & check nested overlay + $UMOUNT_PROG $mnt2 + _overlay_check_dirs $SCRATCH_MNT $upper2 $work2 \ + -o "index=on,nfs_export=on,redirect_dir=on" + + # unmount & check underlying overlay + _scratch_unmount + _check_scratch_fs +} + +# Check non-stale file handles of lower/upper files and verify +# that handle decoded before copy up is encoded to upper after +# copy up. Verify reading data from file open by file handle +# and verify access_at() with dirfd open by file handle. +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +mount_dirs +# Check encode/decode of upper regular file handles +test_file_handles $uppertestdir +# Check encode/decode of upper dir file handle +test_file_handles $uppertestdir -p +# Check encode/write/decode/read/write of upper file handles +test_file_handles $uppertestdir -wrap +# Check encode/decode of lower regular file handles before copy up +test_file_handles $lowertestdir +# Check encode/decode of lower dir file handles before copy up +test_file_handles $lowertestdir -p +# Check encode/write/decode/read/write of lower file handles across copy up +test_file_handles $lowertestdir -wrap +unmount_dirs + +# Check copy up after encode/decode of lower/upper files +# (copy up of disconnected dentry to index dir) +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +mount_dirs +# Check encode/decode/write/read of upper regular file handles +test_file_handles $uppertestdir -a +test_file_handles $uppertestdir -r +# Check encode/decode/write/read of lower regular file handles +test_file_handles $lowertestdir -a +test_file_handles $lowertestdir -r +unmount_dirs + +# Check non-stale handles to unlinked but open lower/upper files +create_dirs +create_test_files $upperdir +create_test_files $upperdir.rw +create_test_files $lowerdir +create_test_files $lowerdir.rw +mount_dirs +test_file_handles $uppertestdir -dk +# Check encode/write/unlink/decode/read of upper regular file handles +test_file_handles $uppertestdir.rw -rwdk +test_file_handles $lowertestdir -dk +# Check encode/write/unlink/decode/read of lower file handles across copy up +test_file_handles $lowertestdir.rw -rwdk +unmount_dirs + +# Check stale handles of unlinked lower/upper files (nlink = 1,0) +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +mount_dirs +# Check decode of upper file handles after unlink/rmdir (nlink == 0) +test_file_handles $uppertestdir -dp +# Check decode of lower file handles after unlink/rmdir (nlink == 0) +test_file_handles $lowertestdir -dp +unmount_dirs + +# Check non-stale file handles of linked lower/upper files (nlink = 1,2,1) +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +mount_dirs +# Check encode/decode of upper file handles (nlink == 1) +test_file_handles $uppertestdir +# Check decode/read of upper file handles after link (nlink == 2) +test_file_handles $uppertestdir -wlr +# Check decode/read of upper file handles after link + unlink (nlink == 1) +test_file_handles $uppertestdir -ur +# Check encode/decode of lower file handles before copy up (nlink == 1) +test_file_handles $lowertestdir +# Check decode/read of lower file handles after copy up + link (nlink == 2) +test_file_handles $lowertestdir -wlr +# Check decode/read of lower file handles after copy up + link + unlink (nlink == 1) +test_file_handles $lowertestdir -ur +unmount_dirs + +# Check non-stale file handles of linked lower/upper hardlinks (nlink = 2,1) +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +# Create lower/upper hardlinks +test_file_handles $lowerdir -l >/dev/null +test_file_handles $upperdir -l >/dev/null +mount_dirs +# Check encode/decode of upper hardlink file handles (nlink == 2) +test_file_handles $uppertestdir +# Check decode/read of upper hardlink file handles after unlink (nlink == 1) +test_file_handles $uppertestdir -wur +# Check encode/decode of lower hardlink file handles before copy up (nlink == 2) +test_file_handles $lowertestdir +# Check decode/read of lower hardlink file handles after copy up + unlink (nlink == 1) +test_file_handles $lowertestdir -wur +unmount_dirs + +# Check stale file handles of unlinked lower/upper hardlinks (nlink = 2,0) +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +# Create lower/upper hardlinks +test_file_handles $lowerdir -l >/dev/null +test_file_handles $upperdir -l >/dev/null +mount_dirs +# Check encode/decode of upper hardlink file handles (nlink == 2) +test_file_handles $uppertestdir +# Check decode of upper hardlink file handles after 2*unlink (nlink == 0) +test_file_handles $uppertestdir -d +# Check encode/decode of lower hardlink file handles before copy up (nlink == 2) +test_file_handles $lowertestdir +# Check decode of lower hardlink file handles after copy up + 2*unlink (nlink == 0) +test_file_handles $lowertestdir -d +unmount_dirs + +# Check non-stale file handles of lower/upper renamed files +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +mount_dirs +# Check decode/read of upper file handles after rename in same upper parent +test_file_handles $uppertestdir -wmr +# Check decode/read of lower file handles after copy up + rename in same merge parent +test_file_handles $lowertestdir -wmr +unmount_dirs + +# Check non-stale file handles of lower/upper moved files +create_dirs +create_test_files $upperdir -w +create_test_files $lowerdir -w +mkdir -p $lowerdir.lo $lowerdir.up $upperdir.lo $upperdir.up +mount_dirs +# Check encode/decode/read of lower/upper file handles after move to new upper testdir +test_file_handles $uppertestdir -o $tmp.upper_file_handles +test_file_handles $lowertestdir -o $tmp.lower_file_handles +mv $uppertestdir/* $uppertestdir.up/ +mv $lowertestdir/* $uppertestdir.lo/ +# Check open and read from stored file handles +test_file_handles $mnt2 -r -i $tmp.upper_file_handles +test_file_handles $mnt2 -r -i $tmp.lower_file_handles +# Check encode/decode/read of lower/upper file handles after move to new merge testdir +test_file_handles $uppertestdir.up -o $tmp.upper_file_handles +test_file_handles $uppertestdir.lo -o $tmp.lower_file_handles +mv $uppertestdir.up/* $lowertestdir.up/ +mv $uppertestdir.lo/* $lowertestdir.lo/ +# Check open and read from stored file handles +test_file_handles $mnt2 -r -i $tmp.upper_file_handles +test_file_handles $mnt2 -r -i $tmp.lower_file_handles +unmount_dirs + +# Check non-stale file handles of lower/upper renamed dirs +create_dirs +create_test_files $upperdir -w +create_test_files $lowerdir -w +create_test_files $upperdir/subdir -w +create_test_files $lowerdir/subdir -w +mount_dirs +# Check encode/decode/read of lower/upper file handles after rename of testdir +test_file_handles $uppertestdir -p -o $tmp.upper_file_handles +test_file_handles $lowertestdir -p -o $tmp.lower_file_handles +# Check encode/decode/read of lower/upper file handles after rename of testdir's parent +test_file_handles $uppertestdir/subdir -p -o $tmp.upper_subdir_file_handles +test_file_handles $lowertestdir/subdir -p -o $tmp.lower_subdir_file_handles +# Rename pure upper dir +mv $uppertestdir $uppertestdir.new/ +# Copy up lower dir, index and rename +mv $lowertestdir $lowertestdir.new/ +# Check open, read and readdir from stored file handles +# (testdir argument is the mount point and NOT the dir +# we are trying to open by stored file handle) +test_file_handles $mnt2 -rp -i $tmp.upper_file_handles +test_file_handles $mnt2 -rp -i $tmp.lower_file_handles +test_file_handles $mnt2 -rp -i $tmp.upper_subdir_file_handles +test_file_handles $mnt2 -rp -i $tmp.lower_subdir_file_handles +# Retry decoding lower subdir file handle when indexed testdir is in dcache +# (providing renamed testdir argument pins the indexed testdir to dcache) +test_file_handles $lowertestdir.new -rp -i $tmp.lower_subdir_file_handles +unmount_dirs + +status=0 +exit
tests/overlay/068.out new file +50 -0
50 lines not on this page · raw message
tests/overlay/group +1 -0
@@ -70,3 +70,4 @@ 065 auto quick mount 066 auto quick copyup 067 auto quick copyup nonsamefs +068 auto quick copyup hardlink exportfs nested
Signature · 1 line
2.17.1
PATCH v2 3/5 overlay: test file handles with nested overlay over non-samefs lower
In reply to Amir GoldsteinThis is a variant of overlay file handles test for an overlayfs that is nested over another lower overlayfs on non-samefs.
--- tests/overlay/069 | 313 ++++++++++++++++++++++++++++++++++++++++++ tests/overlay/069.out | 50 +++++++ tests/overlay/group | 1 + 3 files changed, 364 insertions(+) create mode 100755 tests/overlay/069 create mode 100644 tests/overlay/069.out
tests/overlay/069 new file +313 -0
@@ -0,0 +1,313 @@ +#! /bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright (C) 2019 CTERA Networks. All Rights Reserved. +# +# FS QA Test No. 069 +# +# Test encode/decode of nested overlay file handles +# +# This is a variant of overlay file handles test for an overlayfs that is +# nested over another lower overlayfs on non-samefs. +# +# - Check encode/write/decode/read content of lower/upper file handles +# - Check encode/decode/write/read content of lower/upper file handles +# - Check decode/read of unlinked lower/upper files and directories +# - Check decode/read of lower file handles after copy up, link and unlink +# - Check decode/read of lower file handles after rename of parent and self +# +# This test does not cover connectable file handles of non-directories, +# because name_to_handle_at() syscall does not support requesting connectable +# file handles. +# +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.* + # Unmount the nested overlay mount + $UMOUNT_PROG $mnt2 2>/dev/null +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter + +# real QA test starts here + +_supported_fs overlay +_supported_os Linux +_require_test +_require_scratch_nocheck +# _require_exportfs already requires open_by_handle, but let's not count on it +_require_test_program "open_by_handle" +# We need to require all features together, because nfs_export cannot +# be enabled when index is disabled +_require_scratch_overlay_features index nfs_export redirect_dir + +# Lower overlay lower layer is on test fs, upper is on scratch fs +lower=$OVL_BASE_TEST_MNT/$OVL_LOWER-$seq +upper=$OVL_BASE_SCRATCH_MNT/$OVL_UPPER +work=$OVL_BASE_SCRATCH_MNT/$OVL_WORK + +# Lower dir of nested overlay is the scratch overlay mount at SCRATCH_MNT +upper2=$OVL_BASE_TEST_DIR/$OVL_UPPER.2 +work2=$OVL_BASE_TEST_DIR/$OVL_WORK.2 +mnt2=$OVL_BASE_TEST_DIR/$OVL_MNT.2 + +lowerdir=$lower/lowertestdir +upperdir=$upper/uppertestdir +lowertestdir=$mnt2/lowertestdir +uppertestdir=$mnt2/uppertestdir +NUMFILES=1 + +# Create test dir and empty test files +create_test_files() +{ + local dir=$1 + local opt=$2 + + src/open_by_handle -cp $opt $dir $NUMFILES +} + +# Test encode/decode file handles on overlay mount +test_file_handles() +{ + local dir=$1 + shift + + echo test_file_handles $dir $* | _filter_test_dir | _filter_ovl_dirs | \ + sed -e "s,$tmp\.,,g" + $here/src/open_by_handle $* $dir $NUMFILES +} + +# Re-create lower/upper/work dirs +create_dirs() +{ + # Create the underlying overlay dirs + _scratch_mkfs + + # Re-create the nested overlay upper dirs + rm -rf $lower $upper2 $work2 $mnt2 + mkdir $lower $upper2 $work2 $mnt2 +} + +# Mount a nested overlay with $SCRATCH_MNT as lower layer +mount_dirs() +{ + # Mount the underlying non-samefs overlay + _overlay_mount_dirs $lower $upper $work overlay1 $SCRATCH_MNT \ + -o "index=on,nfs_export=on,redirect_dir=on" + + # Mount the nested overlay + _overlay_mount_dirs $SCRATCH_MNT $upper2 $work2 overlay2 $mnt2 \ + -o "index=on,nfs_export=on,redirect_dir=on" 2>/dev/null || + _notrun "cannot mount nested overlay with nfs_export=on option" + _fs_options overlay2 | grep -q "nfs_export=on" || \ + _notrun "cannot enable nfs_export feature on nested overlay" +} + +# Unmount the nested overlay mount and check underlying overlay layers +unmount_dirs() +{ + # unmount & check nested overlay + $UMOUNT_PROG $mnt2 + _overlay_check_dirs $SCRATCH_MNT $upper2 $work2 \ + -o "index=on,nfs_export=on,redirect_dir=on" + + # unmount & check underlying overlay + $UMOUNT_PROG $SCRATCH_MNT + _overlay_check_dirs $lower $upper $work \ + -o "index=on,nfs_export=on,redirect_dir=on" +} + +# Check non-stale file handles of lower/upper files and verify +# that handle decoded before copy up is encoded to upper after +# copy up. Verify reading data from file open by file handle +# and verify access_at() with dirfd open by file handle. +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +mount_dirs +# Check encode/decode of upper regular file handles +test_file_handles $uppertestdir +# Check encode/decode of upper dir file handle +test_file_handles $uppertestdir -p +# Check encode/write/decode/read/write of upper file handles +test_file_handles $uppertestdir -wrap +# Check encode/decode of lower regular file handles before copy up +test_file_handles $lowertestdir +# Check encode/decode of lower dir file handles before copy up +test_file_handles $lowertestdir -p +# Check encode/write/decode/read/write of lower file handles across copy up +test_file_handles $lowertestdir -wrap +unmount_dirs + +# Check copy up after encode/decode of lower/upper files +# (copy up of disconnected dentry to index dir) +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +mount_dirs +# Check encode/decode/write/read of upper regular file handles +test_file_handles $uppertestdir -a +test_file_handles $uppertestdir -r +# Check encode/decode/write/read of lower regular file handles +test_file_handles $lowertestdir -a +test_file_handles $lowertestdir -r +unmount_dirs + +# Check non-stale handles to unlinked but open lower/upper files +create_dirs +create_test_files $upperdir +create_test_files $upperdir.rw +create_test_files $lowerdir +create_test_files $lowerdir.rw +mount_dirs +test_file_handles $uppertestdir -dk +# Check encode/write/unlink/decode/read of upper regular file handles +test_file_handles $uppertestdir.rw -rwdk +test_file_handles $lowertestdir -dk +# Check encode/write/unlink/decode/read of lower file handles across copy up +test_file_handles $lowertestdir.rw -rwdk +unmount_dirs + +# Check stale handles of unlinked lower/upper files (nlink = 1,0) +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +mount_dirs +# Check decode of upper file handles after unlink/rmdir (nlink == 0) +test_file_handles $uppertestdir -dp +# Check decode of lower file handles after unlink/rmdir (nlink == 0) +test_file_handles $lowertestdir -dp +unmount_dirs + +# Check non-stale file handles of linked lower/upper files (nlink = 1,2,1) +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +mount_dirs +# Check encode/decode of upper file handles (nlink == 1) +test_file_handles $uppertestdir +# Check decode/read of upper file handles after link (nlink == 2) +test_file_handles $uppertestdir -wlr +# Check decode/read of upper file handles after link + unlink (nlink == 1) +test_file_handles $uppertestdir -ur +# Check encode/decode of lower file handles before copy up (nlink == 1) +test_file_handles $lowertestdir +# Check decode/read of lower file handles after copy up + link (nlink == 2) +test_file_handles $lowertestdir -wlr +# Check decode/read of lower file handles after copy up + link + unlink (nlink == 1) +test_file_handles $lowertestdir -ur +unmount_dirs + +# Check non-stale file handles of linked lower/upper hardlinks (nlink = 2,1) +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +# Create lower/upper hardlinks +test_file_handles $lowerdir -l >/dev/null +test_file_handles $upperdir -l >/dev/null +mount_dirs +# Check encode/decode of upper hardlink file handles (nlink == 2) +test_file_handles $uppertestdir +# Check decode/read of upper hardlink file handles after unlink (nlink == 1) +test_file_handles $uppertestdir -wur +# Check encode/decode of lower hardlink file handles before copy up (nlink == 2) +test_file_handles $lowertestdir +# Check decode/read of lower hardlink file handles after copy up + unlink (nlink == 1) +test_file_handles $lowertestdir -wur +unmount_dirs + +# Check stale file handles of unlinked lower/upper hardlinks (nlink = 2,0) +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +# Create lower/upper hardlinks +test_file_handles $lowerdir -l >/dev/null +test_file_handles $upperdir -l >/dev/null +mount_dirs +# Check encode/decode of upper hardlink file handles (nlink == 2) +test_file_handles $uppertestdir +# Check decode of upper hardlink file handles after 2*unlink (nlink == 0) +test_file_handles $uppertestdir -d +# Check encode/decode of lower hardlink file handles before copy up (nlink == 2) +test_file_handles $lowertestdir +# Check decode of lower hardlink file handles after copy up + 2*unlink (nlink == 0) +test_file_handles $lowertestdir -d +unmount_dirs + +# Check non-stale file handles of lower/upper renamed files +create_dirs +create_test_files $upperdir +create_test_files $lowerdir +mount_dirs +# Check decode/read of upper file handles after rename in same upper parent +test_file_handles $uppertestdir -wmr +# Check decode/read of lower file handles after copy up + rename in same merge parent +test_file_handles $lowertestdir -wmr +unmount_dirs + +# Check non-stale file handles of lower/upper moved files +create_dirs +create_test_files $upperdir -w +create_test_files $lowerdir -w +mkdir -p $lowerdir.lo $lowerdir.up $upperdir.lo $upperdir.up +mount_dirs +# Check encode/decode/read of lower/upper file handles after move to new upper testdir +test_file_handles $uppertestdir -o $tmp.upper_file_handles +test_file_handles $lowertestdir -o $tmp.lower_file_handles +mv $uppertestdir/* $uppertestdir.up/ +mv $lowertestdir/* $uppertestdir.lo/ +# Check open and read from stored file handles +test_file_handles $mnt2 -r -i $tmp.upper_file_handles +test_file_handles $mnt2 -r -i $tmp.lower_file_handles +# Check encode/decode/read of lower/upper file handles after move to new merge testdir +test_file_handles $uppertestdir.up -o $tmp.upper_file_handles +test_file_handles $uppertestdir.lo -o $tmp.lower_file_handles +mv $uppertestdir.up/* $lowertestdir.up/ +mv $uppertestdir.lo/* $lowertestdir.lo/ +# Check open and read from stored file handles +test_file_handles $mnt2 -r -i $tmp.upper_file_handles +test_file_handles $mnt2 -r -i $tmp.lower_file_handles +unmount_dirs + +# Check non-stale file handles of lower/upper renamed dirs +create_dirs +create_test_files $upperdir -w +create_test_files $lowerdir -w +create_test_files $upperdir/subdir -w +create_test_files $lowerdir/subdir -w +mount_dirs +# Check encode/decode/read of lower/upper file handles after rename of testdir +test_file_handles $uppertestdir -p -o $tmp.upper_file_handles +test_file_handles $lowertestdir -p -o $tmp.lower_file_handles +# Check encode/decode/read of lower/upper file handles after rename of testdir's parent +test_file_handles $uppertestdir/subdir -p -o $tmp.upper_subdir_file_handles +test_file_handles $lowertestdir/subdir -p -o $tmp.lower_subdir_file_handles +# Rename pure upper dir +mv $uppertestdir $uppertestdir.new/ +# Copy up lower dir, index and rename +mv $lowertestdir $lowertestdir.new/ +# Check open, read and readdir from stored file handles +# (testdir argument is the mount point and NOT the dir +# we are trying to open by stored file handle) +test_file_handles $mnt2 -rp -i $tmp.upper_file_handles +test_file_handles $mnt2 -rp -i $tmp.lower_file_handles +test_file_handles $mnt2 -rp -i $tmp.upper_subdir_file_handles +test_file_handles $mnt2 -rp -i $tmp.lower_subdir_file_handles +# Retry decoding lower subdir file handle when indexed testdir is in dcache +# (providing renamed testdir argument pins the indexed testdir to dcache) +test_file_handles $lowertestdir.new -rp -i $tmp.lower_subdir_file_handles +unmount_dirs + +status=0 +exit
tests/overlay/069.out new file +50 -0
50 lines not on this page · raw message
tests/overlay/group +1 -0
@@ -71,3 +71,4 @@ 066 auto quick copyup 067 auto quick copyup nonsamefs 068 auto quick copyup hardlink exportfs nested +069 auto quick copyup hardlink exportfs nested nonsamefs
Signature · 1 line
2.17.1
PATCH v2 4/5 overlay: test constant ino with nested overlay over samefs lower
In reply to Amir GoldsteinAlso test that d_ino of readdir entries and i_ino from /proc/locks are consistent with st_ino and that inode numbers persist after rename to new parent, drop caches and mount cycle.
--- tests/overlay/070 | 222 ++++++++++++++++++++++++++++++++++++++++++ tests/overlay/070.out | 2 + tests/overlay/group | 1 + 3 files changed, 225 insertions(+) create mode 100755 tests/overlay/070 create mode 100644 tests/overlay/070.out
tests/overlay/070 new file +222 -0
@@ -0,0 +1,222 @@ +#! /bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright (C) 2019 CTERA Networks. All Rights Reserved. +# +# FSQA Test No. 070 +# +# This is a variant of overlay/017 to test constant st_ino numbers for +# nested overlay setup, where all layers of both overlays are on the +# same fs. +# +# This simple test demonstrates a known issue with overlayfs: +# - stat file A shows inode number X +# - modify A to trigger copy up +# - stat file A shows inode number Y != X +# +# Also test that d_ino of readdir entries and i_ino from /proc/locks are +# consistent with st_ino and that inode numbers persist after rename to +# new parent, drop caches and mount cycle. +# +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.* + # Unmount the nested overlay mount + $UMOUNT_PROG $mnt2 2>/dev/null +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter + +# real QA test starts here +_supported_fs overlay +_supported_os Linux +_require_scratch_nocheck +_require_test_program "af_unix" +_require_test_program "t_dir_type" +# We need to require all features together, because nfs_export cannot +# be enabled when index is disabled +_require_scratch_overlay_features index nfs_export redirect_dir + +lower=$OVL_BASE_SCRATCH_MNT/$OVL_LOWER +upper=$OVL_BASE_SCRATCH_MNT/$OVL_UPPER +work=$OVL_BASE_SCRATCH_MNT/$OVL_WORK +# Lower dir of nested overlay is the scratch overlay mount at SCRATCH_MNT +upper2=$OVL_BASE_SCRATCH_MNT/$OVL_UPPER.2 +work2=$OVL_BASE_SCRATCH_MNT/$OVL_WORK.2 +mnt2=$OVL_BASE_SCRATCH_MNT/$OVL_MNT.2 + +lowerdir=$lower/lowertestdir +upperdir=$upper/uppertestdir +lowertestdir=$mnt2/lowertestdir +uppertestdir=$mnt2/uppertestdir + +create_dirs() +{ + # Create the underlying overlay dirs + _scratch_mkfs + + # Create the nested overlay upper dirs + mkdir -p $upper2 $work2 $mnt2 +} + +# Mount a nested overlay with $SCRATCH_MNT as lower layer +mount_dirs() +{ + # Mount the underlying overlay with file handle support + _scratch_mount -o "index=on,nfs_export=on" + + # Mount the nested overlay + # Enable redirect_dir for renaming a merge directory. + # Enabling xino in this test requires that base filesystem inode numbers will + # not use bit 63 in inode number of the test files, because bit 63 is used by + # overlayfs to indicate the layer. Let's just assume that this is true for all + # tested filesystems and if we are wrong, the test may fail. + _overlay_mount_dirs $SCRATCH_MNT $upper2 $work2 overlay2 $mnt2 \ + -o "redirect_dir=on,index=on,xino=on" || \ + _notrun "cannot mount nested overlay with xino=on option" + _fs_options overlay2 | grep -q "xino=on" || \ + _notrun "cannot enable xino feature on nested overlay" +} + +# Unmount the nested overlay mount and check underlying overlay layers +unmount_dirs() +{ + # unmount & check nested overlay + $UMOUNT_PROG $mnt2 + _overlay_check_dirs $SCRATCH_MNT $upper2 $work2 \ + -o "redirect_dir=on,index=on,xino=on" + + # unmount & check underlying overlay + $UMOUNT_PROG $SCRATCH_MNT + _overlay_check_scratch_dirs $lower $upper $work \ + -o "index=on,nfs_export=on" +} + +FILES="dir file symlink link chrdev blkdev fifo socket" + +create_test_files() +{ + local dir=$1 + + # Create our test files. + mkdir -p $dir/dir + touch $dir/file + ln -s $dir/file $dir/symlink + touch $dir/link + ln $dir/link $dir/link2 + mknod $dir/chrdev c 1 1 + mknod $dir/blkdev b 1 1 + mknod $dir/fifo p + $here/src/af_unix $dir/socket +} + +# Record inode numbers in format <ino> <basename> +record_inode_numbers() +{ + local dir=$1 + local outfile=$2 + + echo "record_inode_numbers $outfile" >> $seqres.full + + for n in $FILES; do + ls -id $dir/$n + done | \ + while read ino file; do + f=`basename $file` + echo $ino $f | tee -a $seqres.full >> $outfile + # /proc/locks exposes i_ino - compare it to st_ino. flock -n + # doesn't follow symlink, blocks on fifo and fails on socket + [[ $f =~ fifo|socket|symlink ]] || \ + flock -n $file cat /proc/locks | tee -a $seqres.full | grep -q ":$ino " || \ + echo "lock for $f not found by ino $ino ($outfile) - see $seqres.full" + done +} + +# Check inode numbers match recorder inode numbers +check_inode_numbers() +{ + local dir=$1 + local before=$2 + local after=$3 + + record_inode_numbers $dir $after + + # Test constant stat(2) st_ino - + # Compare before..after - expect silence + # We use diff -u so out.bad will tell us which stage failed + diff -u $before $after + + # Test constant readdir(3)/getdents(2) d_ino - + # Expect to find file by inode number + cat $before | while read ino f; do + $here/src/t_dir_type $dir $ino | tee -a $seqres.full | grep -q $f || \ + echo "$f not found by ino $ino (from $before) - see $seqres.full" + done +} + +rm -f $seqres.full + +create_dirs + +create_test_files $lowerdir +create_test_files $upperdir + +mount_dirs + +# Record inode numbers in the lower overlay +record_inode_numbers $SCRATCH_MNT/lowertestdir $tmp.lower.lo +record_inode_numbers $SCRATCH_MNT/uppertestdir $tmp.lower.up + +# Record inode numbers before copy up +record_inode_numbers $lowertestdir $tmp.before.lo +record_inode_numbers $uppertestdir $tmp.before.up + +# Copy up all files +for f in $FILES; do + # chown -h modifies all those file types + chown -h 100 $lowertestdir/$f + chown -h 100 $uppertestdir/$f +done + +# Compare inode numbers before/after copy up +check_inode_numbers $lowertestdir $tmp.before.lo $tmp.after_copyup.lo +check_inode_numbers $uppertestdir $tmp.before.up $tmp.after_copyup.up + +# Move all files to another dir +mkdir $lowertestdir.2 $uppertestdir.2 + +for f in $FILES; do + mv $lowertestdir/$f $lowertestdir.2/ + mv $uppertestdir/$f $uppertestdir.2/ +done + +echo 3 > /proc/sys/vm/drop_caches + +# Compare inode numbers before/after rename and drop caches +check_inode_numbers $lowertestdir.2 $tmp.after_copyup.lo $tmp.after_move.lo +check_inode_numbers $uppertestdir.2 $tmp.after_copyup.up $tmp.after_move.up + +# Verify that the inode numbers survive a mount cycle +unmount_dirs +mount_dirs + +# Compare inode numbers before/after mount cycle +check_inode_numbers $lowertestdir.2 $tmp.after_move.lo $tmp.after_cycle.lo +check_inode_numbers $uppertestdir.2 $tmp.after_move.up $tmp.after_cycle.up + +unmount_dirs + +echo "Silence is golden" +status=0 +exit
tests/overlay/070.out new file +2 -0
@@ -0,0 +1,2 @@ +QA output created by 070 +Silence is golden
tests/overlay/group +1 -0
@@ -72,3 +72,4 @@ 067 auto quick copyup nonsamefs 068 auto quick copyup hardlink exportfs nested 069 auto quick copyup hardlink exportfs nested nonsamefs +070 auto quick copyup redirect nested
Signature · 1 line
2.17.1
Re: [PATCH v2 4/5] overlay: test constant ino with nested overlay over samefs lower
Quoted 4 lines · Amir Goldstein
Also test that
d_inoof readdir entries andi_inofrom/proc/locksare consistent withst_inoand that inode numbers persist after rename to new parent, drop caches and mount cycle.
overlay/070 and overlay/071 fail for me like this:
QA output created by 071
+flock: cannot open lock file
/scratch/ovl-mnt/lowertestdir/blkdev: No such device or address
...
I.e. there's no block dev with rdev=1/1.
I don't see any other way to fix this, than to remove the device tests. Why are these needed? Is locking code in any way dependent on file type?
Thanks,
Miklos
Re: [PATCH v2 4/5] overlay: test constant ino with nested overlay over samefs lower
Quoted 14 lines · Miklos Szeredi
Quoted 4 lines · Amir Goldstein
Also test that
d_inoof readdir entries andi_inofrom/proc/locksare consistent withst_inoand that inode numbers persist after rename to new parent, drop caches and mount cycle.overlay/070 and overlay/071 fail for me like this:
QA output created by 071
+flock: cannot open lock file
/scratch/ovl-mnt/lowertestdir/blkdev: No such device or address
...
Quoted 4 lines
I.e. there's no block dev with rdev=1/1.
I don't see any other way to fix this, than to remove the device tests.
I ran into similar complain when I worked on generic/564. Apparently, this is not the first test that uses rdev b/1/1 and c/1/1 so not sure how those tests work for everyone.
In generic/564 I used a loopdev as blockdev and /dev/zero as chardev.
Quoted 3 lines
Why are these needed? Is locking code in any way dependent on file type?
Not strictly needed.
See that they already skip file types fifo|socket|symlink.
But note that we are not testing locking, we are using /proc/locks to get a peek at i_ino, so if we skip also blockdev and chardev, we end up testing no special files at all for i_ino consistency. Not the end of the world, but then again using loop dev and /dev/zero would be quite trivial as well.
If it bothers you, I can post a fix.
Thanks,
Amir.
Re: [PATCH v2 4/5] overlay: test constant ino with nested overlay over samefs lower
Quoted 33 lines · Amir Goldstein
Quoted 14 lines · Miklos Szeredi
Quoted 4 lines · Amir Goldstein
Also test that
d_inoof readdir entries andi_inofrom/proc/locksare consistent withst_inoand that inode numbers persist after rename to new parent, drop caches and mount cycle.overlay/070 and overlay/071 fail for me like this:
QA output created by 071
+flock: cannot open lock file
/scratch/ovl-mnt/lowertestdir/blkdev: No such device or address
...Quoted 4 lines
I.e. there's no block dev with rdev=1/1.
I don't see any other way to fix this, than to remove the device tests.
I ran into similar complain when I worked on generic/564. Apparently, this is not the first test that uses rdev b/1/1 and c/1/1 so not sure how those tests work for everyone.
In generic/564 I used a loopdev as blockdev and/dev/zeroas chardev.Quoted 3 lines
Why are these needed? Is locking code in any way dependent on file type?
Not strictly needed.
See that they already skip file types fifo|socket|symlink.
Quoted 7 lines
But note that we are not testing locking, we are using
/proc/locksto get a peek ati_ino, so if we skip also blockdev and chardev, we end up testing no special files at all fori_inoconsistency. Not the end of the world, but then again using loop dev and/dev/zerowould be quite trivial as well.If it bothers you, I can post a fix.
Using /dev/loop and /dev/zero sounds good.
Thanks,
Miklos
PATCH v2 5/5 overlay: test constant ino with nested overlay over non-samefs lower
In reply to Amir GoldsteinAlso test that d_ino of readdir entries and i_ino from /proc/locks are consistent with st_ino and that inode numbers persist after rename to new parent, drop caches and mount cycle.
With nested xino configuration, directory st_ino is not persistent and its st_ino/d_ino/i_ino values are not consistent, so test only non-dir in this test.
--- tests/overlay/071 | 236 ++++++++++++++++++++++++++++++++++++++++++ tests/overlay/071.out | 2 + tests/overlay/group | 1 + 3 files changed, 239 insertions(+) create mode 100755 tests/overlay/071 create mode 100644 tests/overlay/071.out
tests/overlay/071 new file +236 -0
@@ -0,0 +1,236 @@ +#! /bin/bash +# SPDX-License-Identifier: GPL-2.0 +# Copyright (C) 2019 CTERA Networks. All Rights Reserved. +# +# FSQA Test No. 071 +# +# This is a variant of overlay/017 to test constant st_ino numbers for +# nested overlay setup, where lower overlay layers are not on the same fs. +# +# This simple test demonstrates a known issue with overlayfs: +# - stat file A shows inode number X +# - modify A to trigger copy up +# - stat file A shows inode number Y != X +# +# Also test that d_ino of readdir entries and i_ino from /proc/locks are +# consistent with st_ino and that inode numbers persist after rename to +# new parent, drop caches and mount cycle. +# +# With nested xino configuration, directory st_ino is not persistent and +# its st_ino/d_ino/i_ino values are not consistent, so test only non-dir +# in this test. +# +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.* + # Unmount the nested overlay mount + $UMOUNT_PROG $mnt2 2>/dev/null +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter + +# real QA test starts here +_supported_fs overlay +_supported_os Linux +_require_test +_require_scratch_nocheck +_require_test_program "af_unix" +_require_test_program "t_dir_type" +# We need to require all features together, because nfs_export cannot +# be enabled when index is disabled +_require_scratch_overlay_features index nfs_export redirect_dir + +# Lower overlay lower layer is on test fs, upper is on scratch fs +lower=$OVL_BASE_TEST_MNT/$OVL_LOWER-$seq +upper=$OVL_BASE_SCRATCH_MNT/$OVL_UPPER +work=$OVL_BASE_SCRATCH_MNT/$OVL_WORK +# Lower dir of nested overlay is the scratch overlay mount at SCRATCH_MNT +upper2=$OVL_BASE_SCRATCH_DIR/$OVL_UPPER.2 +work2=$OVL_BASE_SCRATCH_DIR/$OVL_WORK.2 +mnt2=$OVL_BASE_SCRATCH_DIR/$OVL_MNT.2 + +lowerdir=$lower/lowertestdir +upperdir=$upper/uppertestdir +lowertestdir=$mnt2/lowertestdir +uppertestdir=$mnt2/uppertestdir + +create_dirs() +{ + # Create the underlying overlay dirs + _scratch_mkfs + + # Re-create the nested overlay upper dirs + rm -rf $lower $upper2 $work2 $mnt2 + mkdir $lower $upper2 $work2 $mnt2 +} + +# Mount a nested overlay with $SCRATCH_MNT as lower layer +mount_dirs() +{ + # Mount the underlying overlay with file handle support + _overlay_mount_dirs $lower $upper $work overlay1 $SCRATCH_MNT \ + -o "index=on,nfs_export=on,xino=on" || \ + _notrun "cannot mount overlay with xino=on option" + _fs_options overlay1 | grep -q "xino=on" || \ + _notrun "cannot enable xino feature on overlay" + + # Mount the nested overlay + # Enable redirect_dir for renaming a merge directory. + # Enabling xino in this test requires that base filesystem inode numbers will + # not use bit 63 in inode number of the test files, because bit 63 is used by + # overlayfs to indicate the layer. Let's just assume that this is true for all + # tested filesystems and if we are wrong, the test may fail. + _overlay_mount_dirs $SCRATCH_MNT $upper2 $work2 overlay2 $mnt2 \ + -o "redirect_dir=on,index=on,xino=on" || \ + _notrun "cannot mount nested overlay with xino=on option" + _fs_options overlay2 | grep -q "xino=on" || \ + _notrun "cannot enable xino feature on nested overlay" +} + +# Unmount the nested overlay mount and check underlying overlay layers +unmount_dirs() +{ + # unmount & check nested overlay + $UMOUNT_PROG $mnt2 + _overlay_check_dirs $SCRATCH_MNT $upper2 $work2 \ + -o "redirect_dir=on,index=on,xino=on" + + # unmount & check underlying overlay + $UMOUNT_PROG $SCRATCH_MNT + _overlay_check_dirs $lower $upper $work \ + -o "index=on,nfs_export=on" +} + +FILES="file symlink link chrdev blkdev fifo socket" + +create_test_files() +{ + local dir=$1 + + # Create our test files. + mkdir -p $dir + touch $dir/file + ln -s $dir/file $dir/symlink + touch $dir/link + ln $dir/link $dir/link2 + mknod $dir/chrdev c 1 1 + mknod $dir/blkdev b 1 1 + mknod $dir/fifo p + $here/src/af_unix $dir/socket +} + +# Record inode numbers in format <ino> <basename> +record_inode_numbers() +{ + local dir=$1 + local outfile=$2 + + echo "record_inode_numbers $outfile" >> $seqres.full + + for n in $FILES; do + ls -id $dir/$n + done | \ + while read ino file; do + f=`basename $file` + echo $ino $f | tee -a $seqres.full >> $outfile + # /proc/locks exposes i_ino - compare it to st_ino. flock -n + # doesn't follow symlink, blocks on fifo and fails on socket + [[ $f =~ fifo|socket|symlink ]] || \ + flock -n $file cat /proc/locks | tee -a $seqres.full | grep -q ":$ino " || \ + echo "lock for $f not found by ino $ino ($outfile) - see $seqres.full" + done +} + +# Check inode numbers match recorder inode numbers +check_inode_numbers() +{ + local dir=$1 + local before=$2 + local after=$3 + + record_inode_numbers $dir $after + + # Test constant stat(2) st_ino - + # Compare before..after - expect silence + # We use diff -u so out.bad will tell us which stage failed + diff -u $before $after + + # Test constant readdir(3)/getdents(2) d_ino - + # Expect to find file by inode number + cat $before | while read ino f; do + $here/src/t_dir_type $dir $ino | tee -a $seqres.full | grep -q $f || \ + echo "$f not found by ino $ino (from $before) - see $seqres.full" + done +} + +rm -f $seqres.full + +create_dirs + +create_test_files $lowerdir +create_test_files $upperdir + +mount_dirs + +# Record inode numbers in the lower overlay +record_inode_numbers $SCRATCH_MNT/lowertestdir $tmp.lower.lo +record_inode_numbers $SCRATCH_MNT/uppertestdir $tmp.lower.up + +# Compare inode numbers in lower overlay vs. nested overlay +# With nested xino lower/lower, all inode numbers overflow xino bits and +# d_ino/i_ino in nested overlay are the same as in lower overlay. +check_inode_numbers $lowertestdir $tmp.lower.lo $tmp.before.lo + +# Record inode numbers before copy up from nested upper +record_inode_numbers $uppertestdir $tmp.before.up + +# Copy up all files +for f in $FILES; do + # chown -h modifies all those file types + chown -h 100 $lowertestdir/$f + chown -h 100 $uppertestdir/$f +done + +# Compare inode numbers before/after copy up +check_inode_numbers $lowertestdir $tmp.before.lo $tmp.after_copyup.lo +check_inode_numbers $uppertestdir $tmp.before.up $tmp.after_copyup.up + +# Move all files to another dir +mkdir $lowertestdir.2 $uppertestdir.2 + +for f in $FILES; do + mv $lowertestdir/$f $lowertestdir.2/ + mv $uppertestdir/$f $uppertestdir.2/ +done + +echo 3 > /proc/sys/vm/drop_caches + +# Compare inode numbers before/after rename and drop caches +check_inode_numbers $lowertestdir.2 $tmp.after_copyup.lo $tmp.after_move.lo +check_inode_numbers $uppertestdir.2 $tmp.after_copyup.up $tmp.after_move.up + +# Verify that the inode numbers survive a mount cycle +unmount_dirs +mount_dirs + +# Compare inode numbers before/after mount cycle +check_inode_numbers $lowertestdir.2 $tmp.after_move.lo $tmp.after_cycle.lo +check_inode_numbers $uppertestdir.2 $tmp.after_move.up $tmp.after_cycle.up + +unmount_dirs + +echo "Silence is golden" +status=0 +exit
tests/overlay/071.out new file +2 -0
@@ -0,0 +1,2 @@ +QA output created by 071 +Silence is golden
tests/overlay/group +1 -0
@@ -73,3 +73,4 @@ 068 auto quick copyup hardlink exportfs nested 069 auto quick copyup hardlink exportfs nested nonsamefs 070 auto quick copyup redirect nested +071 auto quick copyup redirect nested nonsamefs
Signature · 1 line
2.17.1