Because file creates happen slowly on ZFS and the number of files
required to run the test is 100K, this test cannot run in a
reasonable amount of time.
Also bail out of test if createmany fails (possible if MDS or OST is
too small), this prevents the test from just timing out instead.
This patch is back-ported from the following one:
Lustre-commit:
eb38c458c868d5389e2641189218f22ad1272aef
Lustre-change: http://review.whamcloud.com/8841
Test-Parameters: envdefinitions=SLOW=yes testlist=conf-sanity
Test-Parameters: envdefinitions=SLOW=yes testlist=conf-sanity \
mdsfilesystemtype=zfs mdtfilesystemtype=zfs ostfilesystemtype=zfs
Signed-off-by: Nathaniel Clark <nathaniel.l.clark@intel.com>
Change-Id: I6d9daad3239b576935190a121a2aa818441ec97b
Signed-off-by: Jian Yu <jian.yu@intel.com>
Reviewed-on: http://review.whamcloud.com/8854
Tested-by: Jenkins
Reviewed-by: Andreas Dilger <andreas.dilger@intel.com>
Tested-by: Maloo <hpdd-maloo@intel.com>
fi
[ $(facet_fstype $SINGLEMDS) = "zfs" ] &&
-# bug number for skipped test: LU-2778 LU-2059
- ALWAYS_EXCEPT="$ALWAYS_EXCEPT 57b 50h"
+# bug number for skipped test: LU-2778 LU-2059 LU-4444
+ ALWAYS_EXCEPT="$ALWAYS_EXCEPT 57b 50h 69"
init_logging
mkdir -p $DIR/$tdir
$LFS setstripe -i 0 $DIR/$tdir
- createmany -o $DIR/$tdir/$tfile- $num_create
+ createmany -o $DIR/$tdir/$tfile- $num_create ||
+ error "createmany: failed to create $num_create files: $?"
# delete all of the files with objects on OST0 so the
# filesystem is not inconsistent later on
$LFS find $MOUNT --ost 0 | xargs rm