Incremental btrfs send/receive fails if file is unlinked and cloned afterwards
Hi,
the commit "Btrfs: incremental send, check if orphanized dir inode needs delayed rename" causes incremental send/receive to fail if a file is unlinked and then reflinked to the same location from the parent snapshot. An xfstest reproducing the issue is attached.
Regards,
Martin
From ebc5e8721264823a0df92b31e5fb1381f7f5e6f8 Mon Sep 17 00:00:00 2001
From: Martin Raiber <[email protected]>
Date: Fri, 25 Sep 2015 13:24:13 +0200
Subject: [PATCH 1/1] btrfs: test for incremental send after file unlink and
then cloning
Creating a snapshot, then removing a file and cloning it back to its original location, causes btrfs send/receive to fail, because it doesn't use the correct path for the file unlink.
--- tests/btrfs/104 | 94 +++++++++++++++++++++++++++++++++++++++++++++++++++++ tests/btrfs/104.out | 7 ++++ tests/btrfs/group | 1 + 3 files changed, 102 insertions(+) create mode 100755 tests/btrfs/104 create mode 100644 tests/btrfs/104.out
tests/btrfs/104 new file +94 -0
@@ -0,0 +1,94 @@ +#! /bin/bash +# FS QA Test No. btrfs/104 +# +# Test that an incremental send works after a files gets unlinked +# and then cloned back from the previous snapshot. +# +#----------------------------------------------------------------------- +# Copyright (C) 2015 SUSE Linux Products GmbH. All Rights Reserved. +# Copyright (C) 2015 Martin Raiber <[email protected]> +# +# 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 -fr $send_files_dir + rm -f $tmp.* +} + +# get standard environment, filters and checks +. ./common/rc +. ./common/filter + +# real QA test starts here +_supported_fs btrfs +_supported_os Linux +_require_scratch +_require_cloner +_need_to_be_root +_require_cp_reflink + +send_files_dir=$TEST_DIR/btrfs-test-$seq + +rm -f $seqres.full +rm -fr $send_files_dir +mkdir $send_files_dir + +_scratch_mkfs >>$seqres.full 2>&1 +_scratch_mount + +# Create our test file with a single extent of 64K starting at file offset 128K. +mkdir -p $SCRATCH_MNT/foo +$XFS_IO_PROG -f -c "pwrite -S 0xaa 128K 64K" $SCRATCH_MNT/foo/bar | _filter_xfs_io + +_run_btrfs_util_prog subvolume snapshot -r $SCRATCH_MNT $SCRATCH_MNT/mysnap1 +_run_btrfs_util_prog subvolume snapshot $SCRATCH_MNT $SCRATCH_MNT/mysnap2 + +#Remove the file and then reflink it back from the original snapshot +rm $SCRATCH_MNT/mysnap2/foo/bar +cp --reflink=always $SCRATCH_MNT/foo/bar $SCRATCH_MNT/mysnap2/foo/bar + +_run_btrfs_util_prog subvolume snapshot -r $SCRATCH_MNT/mysnap2 $SCRATCH_MNT/mysnap2_ro + +_run_btrfs_util_prog send $SCRATCH_MNT/mysnap1 -f $send_files_dir/1.snap +_run_btrfs_util_prog send -p $SCRATCH_MNT/mysnap1 $SCRATCH_MNT/mysnap2_ro \ + -f $send_files_dir/2.snap + +echo "File digest in the original filesystem:" +md5sum $SCRATCH_MNT/mysnap2_ro/foo/bar | _filter_scratch + +# Now recreate the filesystem by receiving both send streams and verify we get +# the same file contents that the original filesystem had. +_scratch_unmount +_scratch_mkfs >>$seqres.full 2>&1 +_scratch_mount + +_run_btrfs_util_prog receive -vv $SCRATCH_MNT -f $send_files_dir/1.snap +_run_btrfs_util_prog receive -vv $SCRATCH_MNT -f $send_files_dir/2.snap + +echo "File digest in the new filesystem:" +md5sum $SCRATCH_MNT/mysnap2_ro/foo/bar | _filter_scratch + +status=0 +exit
tests/btrfs/104.out new file +7 -0
@@ -0,0 +1,7 @@ +QA output created by 104 +wrote 65536/65536 bytes at offset 131072 +XXX Bytes, X ops; XX:XX:XX.X (XXX YYY/sec and XXX ops/sec) +File digest in the original filesystem: +eb4de91c30abc45b27bc4c00a653d84e SCRATCH_MNT/mysnap2_ro/foo/bar +File digest in the new filesystem: +eb4de91c30abc45b27bc4c00a653d84e SCRATCH_MNT/mysnap2_ro/foo/bar
tests/btrfs/group +1 -0
@@ -106,3 +106,4 @@ 101 auto quick replace 102 auto quick metadata enospc 103 auto quick clone compress +104 auto quick send clone
Signature · 1 line
2.5.1
Quoted 6 lines · Martin Raiber
Hi,
the commit "Btrfs: incremental send, check if orphanized dir inode needs delayed rename" causes incremental send/receive to fail if a file is unlinked and then reflinked to the same location from the parent snapshot. An xfstest reproducing the issue is attached.
Thanks for the report and test.
The problem is not related at all to reflinked files (or extent cloning/dedup). You run into this problem by just creating a file with the same name and at the same location, and it needs to be the file with highest inode number in the snapshot.
I've sent a fix and modified your test too, as it didn't conform to the xfstests coding style, had unnecessary requirements (cloner program, cp with reflink support) and added a description of the problem.
thanks
Quoted 3 lines
Regards,
Martin
Signature · 5 lines
Filipe David Manana, "Reasonable men adapt themselves to the world. Unreasonable men adapt the world to themselves. That's why all progress depends on unreasonable men."