Re: [PATCH] Do not check ocfs2

From: Joel Becker <jlbec@evilplan.org>
To: Dave Chinner <david@fromorbit.com>
Cc: linux-fsdevel@vger.kernel.org,
	Richard Weinberger <richard@nod.at>,
	Eric Sandeen <sandeen@sandeen.net>,
	xfs@oss.sgi.com
Subject: Re: [PATCH] Do not check ocfs2
Date: Mon, 4 Mar 2013 13:05:23 -0800	[thread overview]
Message-ID: <20130304210522.GA7783@localhost> (raw)
In-Reply-To: <20130304004229.GK23616@dastard>

On Mon, Mar 04, 2013 at 11:42:29AM +1100, Dave Chinner wrote:
> On Sun, Mar 03, 2013 at 04:57:55PM -0600, Eric Sandeen wrote:
> > On 3/3/13 4:53 PM, Richard Weinberger wrote:
> > 
> > >> Anyway, what if you did something more along the lines of [pseudocode]
> > >>
> > >> ocfs2)
> > >> 	if mounted.ocfs2 -f $TEST-DEV | frob_as_necessary[1]
> > >> 		;
> > >> 	else
> > >> 		fsck.ocfs2 $TEST-DEV
> > >> 	fi
> > >> 	;;
> > >>
> > >> so that *if* it's mounted on some other node, the fsck won't run.
> > >> That has downsides as Dave mentioned, but for the case where the
> > >> xfstests node is the only one with it in use, it'll still do the
> > >> beneficial consistency check.
> > >>
> > >> Just tweaking the fsck action bsed on *if* it's mounted (or,
> > >> maybe, if the node is in a cluster?) might be a more generic solution
> > >> that is widely applicable to all ocfs2 test environments.
> > > 
> > > Good point. mounted.ocfs2 really makes sense. I'll implement this on my
> > > test suite and submit a new patch.
> > 
> > Sounds good to me.
> > 
> > It'd be most preferable to do a cluster-wide unmount and fsck,
> 
> Which makes no sense to me, because ./check will then do a cluster
> wide unmount as it runs the test harness initialisation. Hence all
> the subsequent tests will run with the filesystem only mounted on
> the local node....
> 
> Really, xfstests is not designed for testing cluster filesystems in
> clustered environments. If we really want to support clustered
> filesystems and cluster wide operations, then we need to think
> about how to architect multi-host support into xfstests sanely.
> Clustered filesystems are not the only people that could make use of
> such functionality (NFS and CIFS come to mind).... ;)

I'm much happier with the "check for other cluster mounts" approach
rather than avoiding fsck entirely.  We like xfstests in the local or
single-node cases to behave exactly as one would expect.

We do run xfstests on multiple nodes in a cluster, but by hand and with
our own cleanup.  Obviously support for automating that would be awesome
:-)

Joel

> 
> Cheers,
> 
> Dave/
> -- 
> Dave Chinner
> david@fromorbit.com
> --
> To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
> the body of a message to majordomo@vger.kernel.org
> More majordomo info at  http://vger.kernel.org/majordomo-info.html

-- 

"Born under a bad sign.
 I been down since I began to crawl.
 If it wasn't for bad luck,
 I wouldn't have no luck at all."

			http://www.jlbec.org/
			jlbec@evilplan.org

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs