xfstests: very basic defragmentation testing for xfs & ext4
PATCH 1/1 xfstests: very basic defragmentation testing for xfs & ext4
Test a few very basic defragmentation scenarios.
This test creates some files, runs defrag on them, and compares the before/after fragmentation as well as file md5sums and timestamps.
(md5sums may just be reading buffered data though, suggestions?)
The test currently expects to find e4defrag in /usr/bin
It should be relatively easy to add more interestingly fragmented files to the tests, as well as to test that memory-mapped files aren't touched, etc - but this gives us a framework.
---
group +1 -0
@@ -326,3 +326,4 @@ prealloc 215 auto metadata quick 216 log metadata auto quick 217 log metadata auto +218 auto fsr quick
218 new file +131 -0
131 lines not on this page · raw message
218.out new file +19 -0
@@ -0,0 +1,19 @@ +QA output created by 218 +Empty file: +Before: 0 +After: 0 +Sparse file (no blocks): +Before: 0 +After: 0 +Contiguous file: +Before: 1 +After: 1 +Write backwards sync, but contiguous - should defrag to 1 extent +Before: 10 +After: 1 +Write backwards sync leaving holes - defrag should do nothing +Before: 16 +After: 16 +Write forwards sync leaving holes - defrag should do nothing +Before: 16 +After: 16
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 6 lines · Eric Sandeen
This test creates some files, runs defrag on them, and compares the before/after fragmentation as well as file md5sums and timestamps.
(md5sums may just be reading buffered data though, suggestions?)
(umount; mount)?
Cheers, Andreas
--
Andreas Dilger
Sr. Staff Engineer, Lustre Group
Sun Microsystems of Canada, Inc.
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 9 lines · Andreas Dilger
Quoted 6 lines · Eric Sandeen
This test creates some files, runs defrag on them, and compares the before/after fragmentation as well as file md5sums and timestamps.
(md5sums may just be reading buffered data though, suggestions?)
(umount; mount)?
I suppose that'd work :)
-Eric
Quoted 1 line
Cheers, Andreas
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 8 lines · Eric Sandeen
Test a few very basic defragmentation scenarios.
This test creates some files, runs defrag on them, and compares the before/after fragmentation as well as file md5sums and timestamps.
(md5sums may just be reading buffered data though, suggestions?)
As Andreas said unmount it before.
Quoted 5 lines
The test currently expects to find e4defrag in
/usr/bin218 new file +1
-0+_supported_fs xfs ext4
218 line 64
Quoted 10 lines
218 new file +10
-0+fragfile=$SCRATCH_MNT/fragfile.$$ + +if [ "$FSTYP" == "xfs" ]; then + DEFRAG=/usr/sbin/xfs_fsr +else + DEFRAG=/usr/bin/e4defrag +fi + +_require_command $DEFRAG +_require_command /usr/sbin/filefrag
I think we're much better off adding a helpers for this in a new common.defrag (or adding it somwhere).
Then set a DEFRAF_PROG variable there, and add a _require_defrag ala
_require_defrag()
{
case "$FSTYP" in
xfs)
DEFRAG_PROG=/usr/sbin/xfs_fsr
;;
ext4)
DEFRAG_PROG=/usr/bin/e4defrag
;;
*)
_notrun "defragmentation not supported for fstype \"$FSTYP\""
;;
esac
_require_command $DEFRAG
_require_command /usr/sbin/filefrag
}_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Quoted 1 line · Christoph Hellwig
On Wed, Jul 15, 2009 at 03:21:11PM -0500, Eric Sandeen wrote:
...
Quoted 24 lines
Quoted 2 lines
(md5sums may just be reading buffered data though, suggestions?)
As Andreas said unmount it before.
Quoted 5 lines
The test currently expects to find e4defrag in
/usr/bin218 new file +1
-0+_supported_fs xfs ext4218 line 64
Quoted 10 lines
218 new file +10
-0+fragfile=$SCRATCH_MNT/fragfile.$$ + +if [ "$FSTYP" == "xfs" ]; then + DEFRAG=/usr/sbin/xfs_fsr +else + DEFRAG=/usr/bin/e4defrag +fi + +_require_command $DEFRAG +_require_command /usr/sbin/filefragI think we're much better off adding a helpers for this in a new common.defrag (or adding it somwhere).
Quoted 2 lines
Then set a
DEFRAF_PROGvariable there, and add a_require_defragala
Both ideas sound good, will send a V2.
-Eric
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
PATCH v2 1/1 xfstests 218: very basic defragmentation testing for xfs & ext4
In reply to Eric SandeenTest a few very basic defragmentation scenarios.
This test creates some files, runs defrag on them, and compares the before/after fragmentation as well as file md5sums and timestamps.
The test currently expects to find e4defrag in /usr/bin
It should be relatively easy to add more interestingly fragmented files to the tests, as well as to test that memory-mapped files aren't touched, etc - but this gives us a framework.
V2: remount before checking file contents, and create common.defrag.
---
218 new file +96 -0
@@ -0,0 +1,96 @@ +#! /bin/sh +# FS QA Test No. 218 +# +# Basic defragmentation sanity tests +# +#----------------------------------------------------------------------- +# Copyright (c) 2009 Eric Sandeen. All Rights Reserved. +# +# 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 +#----------------------------------------------------------------------- +# +# creator [email protected] + +seq=`basename $0` +echo "QA output created by $seq" + +here=`pwd` +tmp=tmp/$$ +status=1 # failure is the default! +trap "_cleanup; exit \$status" 0 1 2 3 15 + +_cleanup() +{ + cd / + rm -f $tmp.* + _cleanup_testdir +} + +# get standard environment, filters and checks +. ./common.rc +. ./common.filter +. ./common.defrag + +# real QA test starts here +_supported_fs xfs ext4 +_supported_os Linux + +_setup_testdir +# We require scratch so that we'll have free contiguous space +_require_scratch +_scratch_mkfs >/dev/null 2>&1 +_scratch_mount + +_require_defrag + +fragfile=$SCRATCH_MNT/fragfile.$$ + +rm -f $fragfile + +# Craft some fragmented files, defrag them, check the result. + +echo "zero-length file:" | tee -a $seq.full +touch $fragfile +_defrag $fragfile + +echo "Sparse file (no blocks):" | tee -a $seq.full +xfs_io -F -f -c "truncate 1m" $fragfile +_defrag $fragfile + +echo "Contiguous file:" | tee -a $seq.full +dd if=/dev/zero of=$fragfile bs=4k count=4 &>/dev/null +_defrag $fragfile + +echo "Write backwards sync, but contiguous - should defrag to 1 extent" | tee -a $seq.full +for I in `seq 9 -1 0`; do + dd if=/dev/zero of=$fragfile bs=4k count=1 conv=notrunc seek=$I oflag=sync &>/dev/null +done +_defrag $fragfile + +echo "Write backwards sync leaving holes - defrag should do nothing" | tee -a $seq.full +for I in `seq 31 -2 0`; do + dd if=/dev/zero of=$fragfile bs=4k count=1 conv=notrunc seek=$I oflag=sync &>/dev/null +done +_defrag $fragfile + +echo "Write forwards sync leaving holes - defrag should do nothing" | tee -a $seq.full +for I in `seq 0 2 31`; do + dd if=/dev/zero of=$fragfile bs=4k count=1 conv=notrunc seek=$I oflag=sync &>/dev/null +done +_defrag $fragfile + +rm -f $seq.full +status=0 +exit
218.out new file +19 -0
@@ -0,0 +1,19 @@ +QA output created by 218 +zero-length file: +Before: 0 +After: 0 +Sparse file (no blocks): +Before: 0 +After: 0 +Contiguous file: +Before: 1 +After: 1 +Write backwards sync, but contiguous - should defrag to 1 extent +Before: 10 +After: 1 +Write backwards sync leaving holes - defrag should do nothing +Before: 16 +After: 16 +Write forwards sync leaving holes - defrag should do nothing +Before: 16 +After: 16
common.defrag new file +68 -0
68 lines not on this page · raw message
common.rc +11 -0
@@ -198,6 +198,17 @@ _scratch_mount() _mount -t $FSTYP `_scratch_mount_options $*` } +_scratch_unmount() +{ + $UMOUNT_PROG $SCRATCH_DEV +} + +_scratch_remount() +{ + _scratch_unmount + _scratch_mount +} + _test_mount() { _test_options mount
group +1 -0
@@ -327,3 +327,4 @@ prealloc 215 auto metadata quick 216 log metadata auto quick 217 log metadata auto +218 auto fsr quick
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs
Re: [PATCH V2] xfstests 218: very basic defragmentation testing for xfs & ext4
Quoted 9 lines · Eric Sandeen
Test a few very basic defragmentation scenarios.
This test creates some files, runs defrag on them, and compares the before/after fragmentation as well as file md5sums and timestamps.
The test currently expects to find e4defrag in
/usr/bin
Quoted 7 lines
It should be relatively easy to add more interestingly fragmented files to the tests, as well as to test that memory-mapped files aren't touched, etc - but this gives us a framework.
V2: remount before checking file contents, and create common.defrag.
The patch loks good to me,
Unfortunately it still fails on my Debian testing system with:
QA output created by 218
zero-length file:
-Before: 0
-After: 0
+Before: 1
+After: 1
Sparse file (no blocks):
-Before: 0
-After: 0
+Before: 1
+After: 1
But that's due to a bug in filefrag that I have already reported to the Debian BTS.
_______________________________________________
xfs mailing list
[email protected]
http://oss.sgi.com/mailman/listinfo/xfs