btrfs: incremental send after moving directories around

Test that an incremental send operation works after doing radical
changes in the directory hierarchy that involve switching the inode
that directory entries point to.

This test exercises scenarios used to fail in btrfs and are fixed by
the following patches for the linux kernel:

 "Btrfs: send, fix failure to move directories with the same name around"
 "Btrfs: incremental send, fix invalid paths for rename operations"

Signed-off-by: Filipe Manana <fdmanana@suse.com>
Reviewed-by: Eryu Guan <eguan@redhat.com>
Signed-off-by: Eryu Guan <eguan@redhat.com>
This commit is contained in:
Filipe Manana
2016-07-02 13:32:08 +01:00
committed by Eryu Guan
parent 20abd130a8
commit baa017a14b
3 changed files with 266 additions and 0 deletions
+263
View File
@@ -0,0 +1,263 @@
#! /bin/bash
# FS QA Test No. btrfs/127
#
# Test that an incremental send operation works after doing radical changes
# in the directory hierarchy that involve switching the inode that directory
# entries point to.
#
#-----------------------------------------------------------------------
# Copyright (C) 2016 SUSE Linux Products GmbH. All Rights Reserved.
# Author: Filipe Manana <fdmanana@suse.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()
{
cd /
rm -fr $send_files_dir
rm -f $tmp.*
}
# get standard environment, filters and checks
. ./common/rc
. ./common/filter
# real QA test starts here
_supported_fs btrfs
_supported_os Linux
_require_test
_require_scratch
_require_fssum
send_files_dir=$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
# case 1
mkdir -p $SCRATCH_MNT/case_1/d/p1
mkdir $SCRATCH_MNT/case_1/p1
# case 2
mkdir -p $SCRATCH_MNT/case_2/a
mkdir $SCRATCH_MNT/case_2/d
mkdir $SCRATCH_MNT/case_2/e
mkdir $SCRATCH_MNT/case_2/f
mkdir $SCRATCH_MNT/case_2/ance
mkdir $SCRATCH_MNT/case_2/d/ance
mkdir $SCRATCH_MNT/case_2/a/c
mv $SCRATCH_MNT/case_2/e $SCRATCH_MNT/case_2/d/ance
mv $SCRATCH_MNT/case_2/f $SCRATCH_MNT/case_2/d/ance
mv $SCRATCH_MNT/case_2/ance $SCRATCH_MNT/case_2/d/ance
# case 3
mkdir -p $SCRATCH_MNT/case_3/d
mkdir $SCRATCH_MNT/case_3/a
mkdir $SCRATCH_MNT/case_3/waiting_dir
mkdir -p $SCRATCH_MNT/case_3/pre/ance
mkdir $SCRATCH_MNT/case_3/d/ance
mkdir $SCRATCH_MNT/case_3/a/c
mv $SCRATCH_MNT/case_3/waiting_dir $SCRATCH_MNT/case_3/d/ance
# case 4
mkdir -p $SCRATCH_MNT/case_4/tmp
mkdir $SCRATCH_MNT/case_4/below_ance
mkdir -p $SCRATCH_MNT/case_4/pre/wait_dir
mkdir $SCRATCH_MNT/case_4/desc
mkdir $SCRATCH_MNT/case_4/ance
mv $SCRATCH_MNT/case_4/below_ance $SCRATCH_MNT/case_4/ance
mkdir $SCRATCH_MNT/case_4/other_dir
# Filesystem looks like:
#
# . (ino 256)
# |--- case_1/ (ino 257)
# | |---- d/ (ino 258)
# | | |--- p1/ (ino 259)
# | |
# | |---- p1/ (ino 260)
# |
# |--- case_2/ (ino 261)
# | |---- a/ (ino 262)
# | | |---- c/ (ino 268)
# | |
# | |---- d/ (ino 263)
# | |---- ance/ (ino 267)
# | |---- e/ (ino 264)
# | |---- f/ (ino 265)
# | |---- ance/ (ino 266)
# |
# |--- case_3/ (ino 269)
# | |---- a/ (ino 271)
# | | |---- c/ (ino 276)
# | |
# | |---- d/ (ino 270)
# | | |---- ance/ (ino 275)
# | | |---- waiting_dir/ (ino 272)
# | |
# | |---- pre/ (ino 273)
# | |---- ance/ (ino 274)
# |
# |--- case_4/ (ino 277)
# |---- tmp/ (ino 278)
# |---- pre/ (ino 280)
# | |---- wait_dir/ (ino 281)
# |
# |---- desc/ (ino 282)
# |---- ance/ (ino 283)
# | |---- below_ance/ (ino 279)
# |
# |---- other_dir/ (ino 284)
#
_run_btrfs_util_prog subvolume snapshot -r $SCRATCH_MNT $SCRATCH_MNT/mysnap1
# case 1
#
# The directory named "d" (inode 257) has in both snapshots an entry with the
# name "p1" but it refers to different inodes in both snapshots (inode 258 in
# the parent snapshot and inode 259 in the send snapshot). When attempting to
# move inode 258, the operation is delayed because its new parent, inode 259,
# was not yet moved/renamed (as the stream is currently processing inode 258).
# Later on when processing inode 259, btrfs' send also ended up delaying the
# move/rename operation for this inode, so that it would happen after inode 258
# was processed, creating a circular dependency that resulted in the send stream
# terminating without issuing a rename operations for the inodes 258 and 259.
#
mv $SCRATCH_MNT/case_1/d/p1 $SCRATCH_MNT/case_1/p1
mv $SCRATCH_MNT/case_1/p1 $SCRATCH_MNT/case_1/d
# case 2
#
# When the inode 265 is processed, the path for inode 267 is computed, which at
# that time corresponds to "case_2/d/ance", and it was stored in the name cache
# (to avoid recomputing it again later when needed).
# Later on when processing inode 266, btrfs' send end up orphanizing (renaming
# to a name matching the pattern o<ino>-<gen>-<seq>) inode 267 because it has
# the same name as inode 266 and at that time it's a child of the new parent
# directory (inode 263) for inode 266. After the orphanization and while still
# processing inode 266, a rename operation for inode 266 was generated. However
# the source path for that rename operation was incorrect because it ended up
# using the old, pre-orphanization, name of inode 267.
#
mv $SCRATCH_MNT/case_2/a/c $SCRATCH_MNT/case_2
mv $SCRATCH_MNT/case_2/d/ance $SCRATCH_MNT/case_2/c
mv $SCRATCH_MNT/case_2/c/ance/ance $SCRATCH_MNT/case_2/d
mv $SCRATCH_MNT/case_2/c/ance/f $SCRATCH_MNT/case_2
mv $SCRATCH_MNT/case_2/c/ance/e $SCRATCH_MNT/case_2/f
# case 3
#
# This is similar to cases 1 and 2, but adding more complexity just to exercise
# btrfs' incremental send correctness.
#
mv $SCRATCH_MNT/case_3/d/ance $SCRATCH_MNT/case_3/a
mv $SCRATCH_MNT/case_3/a/c $SCRATCH_MNT/case_3
mv $SCRATCH_MNT/case_3/a/ance/waiting_dir $SCRATCH_MNT/case_3/c
mv $SCRATCH_MNT/case_3/pre/ance $SCRATCH_MNT/case_3/d
mv $SCRATCH_MNT/case_3/pre $SCRATCH_MNT/case_3/c/waiting_dir
# case 4
#
# When attempting to rename inode 283, the incremental send stream included an
# invalid destination path for the rename command. This was due to a missing
# path loop detection in the send code that made the rename of inode 283 happen
# without waiting for the rename of inode 284 to happen first (since it's an
# ancestor in the send snapshot that is beyond the current progress and it was
# also renamed/moved).
#
mv $SCRATCH_MNT/case_4/other_dir $SCRATCH_MNT/case_4/tmp
mv $SCRATCH_MNT/case_4/ance/below_ance $SCRATCH_MNT/case_4/tmp/other_dir
mv $SCRATCH_MNT/case_4/pre/wait_dir $SCRATCH_MNT/case_4/tmp/other_dir
mv $SCRATCH_MNT/case_4/pre $SCRATCH_MNT/case_4/tmp/other_dir/below_ance
mv $SCRATCH_MNT/case_4/desc $SCRATCH_MNT/case_4/tmp/other_dir/wait_dir
mv $SCRATCH_MNT/case_4/ance $SCRATCH_MNT/case_4/tmp/other_dir/wait_dir/desc
# Filesystem now looks like:
#
# . (ino 256)
# |--- case_1/ (ino 257)
# | |--- d/ (ino 258)
# | |--- p1/ (ino 260)
# | |--- p1/ (ino 259)
# |
# |--- case_2/ (ino 261)
# | |---- a/ (ino 262)
# | |---- c/ (ino 268)
# | | |---- ance/ (ino 267)
# | |
# | |---- d/ (ino 263)
# | | |---- ance/ (ino 266)
# | |
# | |---- f/ (ino 265)
# | |---- e/ (ino 264)
# |
# |--- case_3/ (ino 269)
# | |---- a/ (ino 271)
# | | |---- ance/ (ino 275)
# | |
# | |---- c/ (ino 276)
# | | |---- waiting_dir/ (ino 272)
# | | |---- pre/ (ino 273)
# | |
# | |---- d/ (ino 270)
# | |---- ance/ (ino 274)
# |
# |--- case_4/ (ino 277)
# |---- tmp/ (ino 278)
# |---- other_dir/ (ino 284)
# |---- below_ance/ (ino 279)
# | |---- pre/ (ino 280)
# |
# |---- wait_dir/ (ino 281)
# |---- desc/ (ino 282)
# |---- ance/ (ino 283)
#
_run_btrfs_util_prog subvolume snapshot -r $SCRATCH_MNT $SCRATCH_MNT/mysnap2
run_check $FSSUM_PROG -A -f -w $send_files_dir/1.fssum $SCRATCH_MNT/mysnap1
run_check $FSSUM_PROG -A -f -w $send_files_dir/2.fssum \
-x $SCRATCH_MNT/mysnap2/mysnap1 $SCRATCH_MNT/mysnap2
_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 \
-f $send_files_dir/2.snap
# Now recreate the filesystem by receiving both send streams and verify we get
# the same content that the original filesystem had.
_scratch_unmount
_scratch_mkfs >>$seqres.full 2>&1
_scratch_mount
_run_btrfs_util_prog receive $SCRATCH_MNT -f $send_files_dir/1.snap
run_check $FSSUM_PROG -r $send_files_dir/1.fssum $SCRATCH_MNT/mysnap1
_run_btrfs_util_prog receive $SCRATCH_MNT -f $send_files_dir/2.snap
run_check $FSSUM_PROG -r $send_files_dir/2.fssum $SCRATCH_MNT/mysnap2
echo "Silence is golden"
status=0
exit
+2
View File
@@ -0,0 +1,2 @@
QA output created by 127
Silence is golden
+1
View File
@@ -129,3 +129,4 @@
124 auto replace
125 auto replace
126 auto quick qgroup
127 auto quick send