Message ID | 161836232608.2754991.16417283237743979525.stgit@magnolia (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
Series | fstests: random fixes | expand |
On Tue, Apr 13, 2021 at 06:05:26PM -0700, Darrick J. Wong wrote: > From: Darrick J. Wong <djwong@kernel.org> > > Greatly increase the number of fs ops that fsstress is supposed to run > in in this test so that we can ensure that it's still running when the > quotaoff gets run. 1000 might have been sufficient in 2013, but it > isn't now. How long does this now run?
On Wed, Apr 14, 2021 at 07:21:35AM +0100, Christoph Hellwig wrote: > On Tue, Apr 13, 2021 at 06:05:26PM -0700, Darrick J. Wong wrote: > > From: Darrick J. Wong <djwong@kernel.org> > > > > Greatly increase the number of fs ops that fsstress is supposed to run > > in in this test so that we can ensure that it's still running when the > > quotaoff gets run. 1000 might have been sufficient in 2013, but it > > isn't now. > > How long does this now run? The same amount of time (~15s per _exercise) as before -- we start fsstress in the background, wait 10 seconds, run quotaoff, wait 5 more seconds, and then kill the fsstress process. Bumping nrops to 1 million gives fsstress enough work to do that it probably won't exit on its own before the quotaoff completes. FWIW the fastest storage I have can run about ~100000 fsstress ops per minute, so I figured that 1 million ought to do for now. --D
Looks good,
Reviewed-by: Christoph Hellwig <hch@lst.de>
diff --git a/tests/xfs/305 b/tests/xfs/305 index fdb48a1d..9a3f0e03 100755 --- a/tests/xfs/305 +++ b/tests/xfs/305 @@ -50,7 +50,7 @@ _exercise() _qmount mkdir -p $QUOTA_DIR - $FSSTRESS_PROG -d $QUOTA_DIR -n 1000 -p 100 $FSSTRESS_AVOID >/dev/null 2>&1 & + $FSSTRESS_PROG -d $QUOTA_DIR -n 1000000 -p 100 $FSSTRESS_AVOID >/dev/null 2>&1 & sleep 10 xfs_quota -x -c "off -$type" $SCRATCH_DEV sleep 5