From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from ipmailnode02.adl6.internode.on.net ([150.101.137.148]:43914 "EHLO ipmailnode02.adl6.internode.on.net" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750859AbeEUWvW (ORCPT ); Mon, 21 May 2018 18:51:22 -0400 Date: Tue, 22 May 2018 08:51:20 +1000 From: Dave Chinner Subject: Re: [PATCH] xfs/132: unmount scratch fs after test Message-ID: <20180521225120.GT23861@dastard> References: <20180521160648.GI4910@magnolia> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <20180521160648.GI4910@magnolia> Sender: fstests-owner@vger.kernel.org To: "Darrick J. Wong" Cc: Eryu Guan , fstests List-ID: On Mon, May 21, 2018 at 09:06:48AM -0700, Darrick J. Wong wrote: > From: Darrick J. Wong > > If this test runs successfully, it'll leave a shutdown scratch fs > behind. The test framework doesn't unmount the fs for us (because we > told it not to check the fs) so the test framework tries to "test -d > $SCRATCH_MNT" and blows up on the IO error it receives from the dead fs. > > Signed-off-by: Darrick J. Wong > --- > tests/xfs/132 | 1 + > 1 file changed, 1 insertion(+) > > diff --git a/tests/xfs/132 b/tests/xfs/132 > index 6bb03e79..3fa695a7 100755 > --- a/tests/xfs/132 > +++ b/tests/xfs/132 > @@ -67,6 +67,7 @@ _scratch_mount > > # The corrupt inode should be tripped over during these initial file creates. > touch $SCRATCH_MNT/file{0,1,2,3,4,5}{0,1,2,3,4,5} 2>&1 | _filter_scratch > +_scratch_unmount I've already NAK'd that patch - the test harness should be unmounting scratch in all situations and returning the test harness to a known initial state before the next test runs... Cheers, Dave. -- Dave Chinner david@fromorbit.com