[PATCH 1/3] overlay/017: Remove constant st_ino/d_ino test for hardlinks

[PATCH 1/3] overlay/017: Remove constant st_ino/d_ino test for hardlinks

[Chandan Rajendra]

Constant st_ino/d_ino for hardlinks requires "index" config
feature/mount option to be enabled. overlay/018 already tests constant
st_ino feature for hardlinks. A future commit will add constant d_ino
test to overlay/018 test.

Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
---
 tests/overlay/017 | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/tests/overlay/017 b/tests/overlay/017
index e9c8525..03955d0 100755
--- a/tests/overlay/017
+++ b/tests/overlay/017
@@ -72,10 +72,8 @@ mknod $lowerdir/chrdev c 1 1
 mknod $lowerdir/blkdev b 1 1
 mknod $lowerdir/fifo p
 $here/src/af_unix $lowerdir/socket
-touch $lowerdir/hardlink1
-ln $lowerdir/hardlink1 $lowerdir/hardlink2
 
-FILES="dir file symlink chrdev blkdev fifo socket hardlink1 hardlink2"
+FILES="dir file symlink chrdev blkdev fifo socket"
 
 # Record inode numbers in format <ino> <basename>
 function record_inode_numbers()
-- 
2.9.5

[PATCH 2/3] overlay/018: Add constant d_ino test

[Chandan Rajendra]

Constant d_ino for hardlinks is possible only when "index" config
feature/mount option is enabled.

Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
---
 tests/overlay/018 | 21 +++++++++++++++------
 1 file changed, 15 insertions(+), 6 deletions(-)

diff --git a/tests/overlay/018 b/tests/overlay/018
index a4f67d1..6e026c1 100755
--- a/tests/overlay/018
+++ b/tests/overlay/018
@@ -50,6 +50,7 @@ _supported_fs overlay
 _supported_os Linux
 _require_scratch
 _require_scratch_feature index
+_require_test_program "t_dir_type"
 
 rm -f $seqres.full
 
@@ -65,14 +66,15 @@ ln $lowerdir/foo $lowerdir/bar
 # Record inode numbers in format <ino> <nlink>
 function record_ino_nlink()
 {
-	ls -li $FILES | awk '{ print $1, $3 }' > $1
+	ls -li $FILES | awk '{ print $1, $3, $10}' > $1
 }
 
 # Check inode numbers match recorded inode numbers
 function check_ino_nlink()
 {
-	before=$1
-	after=$2
+	dir=$1
+	before=$2
+	after=$3
 
 	record_ino_nlink $after
 
@@ -80,6 +82,13 @@ function check_ino_nlink()
 	#   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 nlink f; do
+		$here/src/t_dir_type $dir $ino | grep -q $(basename $f) || \
+			echo "$(basename $f) not found by ino $ino (from $before)"
+	done
 }
 
 # Enable overlay index feature to prevent breaking hardlinks on copy up
@@ -104,14 +113,14 @@ echo "one" >> $bar
 
 echo "== After write one =="
 cat $FILES
-check_ino_nlink $tmp.before $tmp.after_one
+check_ino_nlink $SCRATCH_MNT $tmp.before $tmp.after_one
 
 # Verify that the hardlinks survive a mount cycle
 _scratch_cycle_mount index=on
 
 echo "== After mount cycle =="
 cat $FILES
-check_ino_nlink $tmp.after_one $tmp.after_cycle
+check_ino_nlink $SCRATCH_MNT $tmp.after_one $tmp.after_cycle
 
 # Drop caches to get the copied up hardlink out of cache
 echo 3 > /proc/sys/vm/drop_caches
@@ -121,7 +130,7 @@ echo "two" >> $foo
 
 echo "== After write two =="
 cat $FILES
-check_ino_nlink $tmp.after_one $tmp.after_two
+check_ino_nlink $SCRATCH_MNT $tmp.after_one $tmp.after_two
 
 status=0
 exit
-- 
2.9.5

[PATCH 3/3] overlay: Test hardlink breakage on non-samefs setup

[Chandan Rajendra]

This commit adds a test to check hardlink breakage when the overlayfs
instance is composed of two different underlying filesystem instances.

For example,
$ mount -t xfs /dev/loop0 /mnt/test
$ mount -t xfs /dev/loop1 /mnt/scratch
$ mkdir /mnt/scratch/upper
$ mkdir /mnt/scratch/work
$ mount -t overlay overlay -o lowerdir=/mnt/test \
        -o upperdir=/mnt/scratch/upper \
        -o workdir=/mnt/scratch/work /mnt/merge

Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>
---
 tests/overlay/044     | 145 ++++++++++++++++++++++++++++++++++++++++++++++++++
 tests/overlay/044.out |  21 ++++++++
 tests/overlay/group   |   1 +
 3 files changed, 167 insertions(+)
 create mode 100755 tests/overlay/044
 create mode 100644 tests/overlay/044.out

diff --git a/tests/overlay/044 b/tests/overlay/044
new file mode 100755
index 0000000..9c0ff04
--- /dev/null
+++ b/tests/overlay/044
@@ -0,0 +1,145 @@
+#! /bin/bash
+# FSQA Test No. 044
+#
+# Test hardlink breakage on non-samefs setup
+# This is a variant of overlay/018 to test.
+#
+# This simple test demonstrates a known issue with overlayfs:
+# - file A and B are hardlinked in lower
+# - modify A to trigger copy up
+# - file A is no longer a hardlink of file B
+#
+#-----------------------------------------------------------------------
+#
+# Copyright (C) 2017 IBM Corporation. All Rights Reserved.
+# Author: Chandan Rajendra <chandan@linux.vnet.ibm.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"
+
+tmp=/tmp/$$
+status=1	# failure is the default!
+trap "_cleanup; exit \$status" 0 1 2 3 15
+
+_cleanup()
+{
+	rm -f $tmp.*
+}
+
+# get standard environment, filters and checks
+. ./common/rc
+. ./common/filter
+
+# real QA test starts here
+_supported_fs overlay
+_supported_os Linux
+_require_scratch
+_require_test
+_require_scratch_feature index
+_require_test_program "t_dir_type"
+
+rm -f $seqres.full
+
+# Record inode numbers in format <ino> <nlink>
+function record_ino_nlink()
+{
+	ls -li $FILES | awk '{ print $1, $3, $10}' > $1
+}
+
+# Check inode numbers match recorded inode numbers
+function check_ino_nlink()
+{
+	dir=$1
+	before=$2
+	after=$3
+
+	record_ino_nlink $after
+
+	# Test constant stat(2) st_ino/st_nlink -
+	#   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 nlink f; do
+		$here/src/t_dir_type $dir $ino | grep -q $(basename $f) || \
+			echo "$(basename $f) not found by ino $ino (from $before)"
+	done
+}
+
+lowerdir=$OVL_BASE_TEST_DIR/$seq-ovl-lower
+rm -rf $lowerdir
+mkdir $lowerdir
+
+# Create 2 hardlinked files in lower
+echo "zero" >> $lowerdir/foo
+ln $lowerdir/foo $lowerdir/bar
+
+_scratch_mkfs >>$seqres.full 2>&1
+
+upperdir=$OVL_BASE_SCRATCH_MNT/$OVL_UPPER
+workdir=$OVL_BASE_SCRATCH_MNT/$OVL_WORK
+
+# Enable overlay index feature to prevent breaking hardlinks on copy up
+OVERLAY_MOUNT_OPTIONS="${OVERLAY_MOUNT_OPTIONS} -o index=on"
+_overlay_scratch_mount_dirs $lowerdir $upperdir $workdir \
+			    $OVERLAY_MOUNT_OPTIONS
+
+rm -f $tmp.*
+
+foo=$SCRATCH_MNT/foo
+bar=$SCRATCH_MNT/bar
+
+FILES="$foo $bar"
+
+echo "== Before copy up =="
+cat $FILES
+record_ino_nlink $tmp.before
+
+# Modify content of one of the hardlinks
+# Intentionally modify the last hardlink in $FILES, so after mount cycle
+# when reading the first file in $FILES, last file won't be in inode/dcache
+echo "one" >> $bar
+
+echo "== After write one =="
+cat $FILES
+check_ino_nlink $SCRATCH_MNT $tmp.before $tmp.after_one
+
+# Verify that the hardlinks survive a mount cycle
+$UMOUNT_PROG $SCRATCH_MNT
+_overlay_scratch_mount_dirs $lowerdir $upperdir $workdir \
+			    $OVERLAY_MOUNT_OPTIONS
+
+echo "== After mount cycle =="
+cat $FILES
+check_ino_nlink $SCRATCH_MNT $tmp.after_one $tmp.after_cycle
+
+# Drop caches to get the copied up hardlink out of cache
+echo 3 > /proc/sys/vm/drop_caches
+
+# Modify content of the other hardlink
+echo "two" >> $foo
+
+echo "== After write two =="
+cat $FILES
+check_ino_nlink $SCRATCH_MNT $tmp.after_one $tmp.after_two
+
+status=0
+exit
diff --git a/tests/overlay/044.out b/tests/overlay/044.out
new file mode 100644
index 0000000..8e519e8
--- /dev/null
+++ b/tests/overlay/044.out
@@ -0,0 +1,21 @@
+QA output created by 044
+== Before copy up ==
+zero
+zero
+== After write one ==
+zero
+one
+zero
+one
+== After mount cycle ==
+zero
+one
+zero
+one
+== After write two ==
+zero
+one
+two
+zero
+one
+two
diff --git a/tests/overlay/group b/tests/overlay/group
index a99ff07..b32c6cd 100644
--- a/tests/overlay/group
+++ b/tests/overlay/group
@@ -46,3 +46,4 @@
 041 auto quick copyup nonsamefs
 042 auto quick copyup hardlink
 043 auto quick copyup nonsamefs
+044 auto quick copyup hardlink nonsamefs
-- 
2.9.5

Re: [PATCH 1/3] overlay/017: Remove constant st_ino/d_ino test for hardlinks

[Amir Goldstein]

On Sat, Nov 18, 2017 at 7:58 AM, Chandan Rajendra
<chandan@linux.vnet.ibm.com> wrote:
> Constant st_ino/d_ino for hardlinks requires "index" config
> feature/mount option to be enabled. overlay/018 already tests constant
> st_ino feature for hardlinks. A future commit will add constant d_ino
> test to overlay/018 test.
>
> Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>

Looks good and tested.

Thanks,
Amir.

Re: [PATCH 2/3] overlay/018: Add constant d_ino test

[Amir Goldstein]

On Sat, Nov 18, 2017 at 7:58 AM, Chandan Rajendra
<chandan@linux.vnet.ibm.com> wrote:
> Constant d_ino for hardlinks is possible only when "index" config
> feature/mount option is enabled.
>
> Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>

Looks good and tested.

Thanks,
Amir.

Re: [PATCH 3/3] overlay: Test hardlink breakage on non-samefs setup

[Amir Goldstein]

On Sat, Nov 18, 2017 at 7:58 AM, Chandan Rajendra
<chandan@linux.vnet.ibm.com> wrote:
> This commit adds a test to check hardlink breakage when the overlayfs
> instance is composed of two different underlying filesystem instances.
>
> For example,
> $ mount -t xfs /dev/loop0 /mnt/test
> $ mount -t xfs /dev/loop1 /mnt/scratch
> $ mkdir /mnt/scratch/upper
> $ mkdir /mnt/scratch/work
> $ mount -t overlay overlay -o lowerdir=/mnt/test \
>         -o upperdir=/mnt/scratch/upper \
>         -o workdir=/mnt/scratch/work /mnt/merge
>
> Signed-off-by: Chandan Rajendra <chandan@linux.vnet.ibm.com>

Looks good and tested.

Thanks,
Amir.