On Wed, Mar 21, 2018 at 02:22:29PM +0200, Amir Goldstein wrote: > > + > > +_log_writes_mount > > +$FSSTRESS_PROG $fsstress_args > /dev/null 2>&1 > > You should run fsstress with run_check() so output will go to $seqres.full > this way if you are able to catch a bug, you can take the random seed > from fsstress output and repeat the same event sequence, which > doesn't guaranty, but can increase the chances of reproducing the bug. I suggested dropping run_check, as I don't think we care about the fsstress return value here (and I always try to avoid new run_check usage), but I agree that it might be useful to save the fsstress output to $seqres.full. Thanks, Eryu
