Re: [PATCH v3 2/2] xfs: test xfsdump when an inode < root inode is present

["Darrick J. Wong" <djwong@kernel.org>]

On Tue, Apr 19, 2022 at 01:03:26AM +0800, Zorro Lang wrote:
> From: Eric Sandeen <sandeen@redhat.com>
> 
> This tests a longstanding bug where xfsdumps are not properly
> created when an inode is present on the filesytsem which has
> a lower number than the root inode.
> 
> Signed-off-by: Eric Sandeen <sandeen@redhat.com>
> Signed-off-by: Zorro Lang <zlang@redhat.com>

LGTM
Reviewed-by: Darrick J. Wong <djwong@kernel.org>

--D

> ---
>  tests/xfs/999     | 62 +++++++++++++++++++++++++++++++++++++++++++++++
>  tests/xfs/999.out | 47 +++++++++++++++++++++++++++++++++++
>  2 files changed, 109 insertions(+)
>  create mode 100755 tests/xfs/999
>  create mode 100644 tests/xfs/999.out
> 
> diff --git a/tests/xfs/999 b/tests/xfs/999
> new file mode 100755
> index 00000000..9b1f7f5b
> --- /dev/null
> +++ b/tests/xfs/999
> @@ -0,0 +1,62 @@
> +#! /bin/bash
> +# SPDX-License-Identifier: GPL-2.0
> +# Copyright (c) 2022 Red Hat, Inc. All Rights Reserved.
> +#
> +# FS QA Test No. 999
> +#
> +# Create a filesystem which contains an inode with a lower number
> +# than the root inode. Ensure that xfsdump/xfsrestore handles this.
> +#
> +. ./common/preamble
> +_begin_fstest auto quick dump
> +
> +# Import common functions.
> +. ./common/dump
> +
> +_supported_fs xfs
> +_require_scratch
> +
> +# A large stripe unit will put the root inode out quite far
> +# due to alignment, leaving free blocks ahead of it.
> +_scratch_mkfs_xfs -d sunit=1024,swidth=1024 > $seqres.full 2>&1
> +
> +# Mounting /without/ a stripe should allow inodes to be allocated
> +# in lower free blocks, without the stripe alignment.
> +_scratch_mount -o sunit=0,swidth=0
> +
> +root_inum=$(stat -c %i $SCRATCH_MNT)
> +
> +# Consume space after the root inode so that the blocks before
> +# root look "close" for the next inode chunk allocation
> +$XFS_IO_PROG -f -c "falloc 0 16m" $SCRATCH_MNT/fillfile
> +
> +# And make a bunch of inodes until we (hopefully) get one lower
> +# than root, in a new inode chunk.
> +echo "root_inum: $root_inum" >> $seqres.full
> +for i in $(seq 0 4096) ; do
> +	fname=$SCRATCH_MNT/$(printf "FILE_%03d" $i)
> +	touch $fname
> +	inum=$(stat -c "%i" $fname)
> +	[[ $inum -lt $root_inum ]] && break
> +done
> +
> +echo "created: $inum" >> $seqres.full
> +
> +[[ $inum -lt $root_inum ]] || _notrun "Could not set up test"
> +
> +# Now try a dump and restore. Cribbed from xfs/068
> +_create_dumpdir_stress
> +
> +echo -n "Before: " >> $seqres.full
> +_count_dumpdir_files | tee $tmp.before >> $seqres.full
> +
> +# filter out the file count, it changes as fsstress adds new operations
> +_do_dump_restore | sed -e "/entries processed$/s/[0-9][0-9]*/NUM/g"
> +
> +echo -n "After: " >> $seqres.full
> +_count_restoredir_files | tee $tmp.after >> $seqres.full
> +diff -u $tmp.before $tmp.after
> +
> +# success, all done
> +status=0
> +exit
> diff --git a/tests/xfs/999.out b/tests/xfs/999.out
> new file mode 100644
> index 00000000..f6ab75d2
> --- /dev/null
> +++ b/tests/xfs/999.out
> @@ -0,0 +1,47 @@
> +QA output created by 999
> +Creating directory system to dump using fsstress.
> +
> +-----------------------------------------------
> +fsstress : -f link=10 -f creat=10 -f mkdir=10 -f truncate=5 -f symlink=10
> +-----------------------------------------------
> +xfsdump|xfsrestore ...
> +xfsdump  -s DUMP_SUBDIR - SCRATCH_MNT | xfsrestore  - RESTORE_DIR
> +xfsrestore: using file dump (drive_simple) strategy
> +xfsrestore: searching media for dump
> +xfsrestore: examining media file 0
> +xfsrestore: dump description: 
> +xfsrestore: hostname: HOSTNAME
> +xfsrestore: mount point: SCRATCH_MNT
> +xfsrestore: volume: SCRATCH_DEV
> +xfsrestore: session time: TIME
> +xfsrestore: level: 0
> +xfsrestore: session label: ""
> +xfsrestore: media label: ""
> +xfsrestore: file system ID: ID
> +xfsrestore: session id: ID
> +xfsrestore: media ID: ID
> +xfsrestore: searching media for directory dump
> +xfsrestore: reading directories
> +xfsrestore: NUM directories and NUM entries processed
> +xfsrestore: directory post-processing
> +xfsrestore: restoring non-directory files
> +xfsrestore: restore complete: SECS seconds elapsed
> +xfsrestore: Restore Status: SUCCESS
> +xfsdump: using file dump (drive_simple) strategy
> +xfsdump: level 0 dump of HOSTNAME:SCRATCH_MNT
> +xfsdump: dump date: DATE
> +xfsdump: session id: ID
> +xfsdump: session label: ""
> +xfsdump: ino map <PHASES>
> +xfsdump: ino map construction complete
> +xfsdump: estimated dump size: NUM bytes
> +xfsdump: /var/xfsdump/inventory created
> +xfsdump: creating dump session media file 0 (media 0, file 0)
> +xfsdump: dumping ino map
> +xfsdump: dumping directories
> +xfsdump: dumping non-directory files
> +xfsdump: ending media file
> +xfsdump: media file size NUM bytes
> +xfsdump: dump size (non-dir files) : NUM bytes
> +xfsdump: dump complete: SECS seconds elapsed
> +xfsdump: Dump Status: SUCCESS
> -- 
> 2.31.1
>