Re: [PATCH] xfs/432: fix this test when external devices are in use

["Darrick J. Wong" <djwong@kernel.org>]

On Wed, Jul 27, 2022 at 08:21:42PM +0800, Zorro Lang wrote:
> On Tue, Jul 26, 2022 at 12:51:31PM -0700, Darrick J. Wong wrote:
> > From: Darrick J. Wong <djwong@kernel.org>
> > 
> > This program exercises metadump and mdrestore being run against the
> > scratch device.  Therefore, the test must pass external log / rt device
> > arguments to xfs_repair -n to check the "restored" filesystem.  Fix the
> > incorrect usage, and report repair failures, since this test has been
> > silently failing for a while now.
> > 
> > Signed-off-by: Darrick J. Wong <djwong@kernel.org>
> > ---
> >  tests/xfs/432 |   11 ++++++++++-
> >  1 file changed, 10 insertions(+), 1 deletion(-)
> > 
> > diff --git a/tests/xfs/432 b/tests/xfs/432
> > index 86012f0b..5c6744ce 100755
> > --- a/tests/xfs/432
> > +++ b/tests/xfs/432
> > @@ -89,7 +89,16 @@ _scratch_xfs_metadump $metadump_file -w
> >  xfs_mdrestore $metadump_file $metadump_img
> >  
> >  echo "Check restored metadump image"
> > -$XFS_REPAIR_PROG -n $metadump_img >> $seqres.full 2>&1
> > +repair_args=('-n')
> > +[ "$USE_EXTERNAL" = yes -a ! -z "$SCRATCH_LOGDEV" ] && \
> > +	repair_args+=('-l' "$SCRATCH_LOGDEV")
> > +
> > +[ "$USE_EXTERNAL" = yes -a ! -z "$SCRATCH_RTDEV" ] && \
> > +	repair_args+=('-r' "$SCRATCH_RTDEV")
> > +
> > +$XFS_REPAIR_PROG "${repair_args[@]}" $metadump_img >> $seqres.full 2>&1
> > +res=$?
> > +test $res -ne 0 && echo "xfs_repair on restored fs returned $res?"
> 
> Make sense to me, I don't have better idea. One question, is xfs_metadump
> and xfs_mdrestore support rtdev? Due to I didn't find xfs_metadump have
> a "-r" option, although it has "-l logdev" :)

Oops, no it doesn't, so I'll remove that.

> About the "$res", I don't know why we need this extra variable, as it's
> not used in other place.

If you don't pass the correct arguments to xfs_repair or the metadump
trashes the fs, it'll exit with a nonzero code.  All the output goes to
$seqres.full, which means the test runner has no idea anything went
wrong and marks the test passed even though repair failed.

--D

> Thanks,
> Zorro
> 
> >  
> >  # success, all done
> >  status=0
> > 
>