well, strange cifs in many way''s,,

short story;
we've cifs enabled a cluster volume, probably had some issues while
doing this since it never seemed to become available through the
"normal" cluster-resource creation as a cifs-resource...

But, on the local server as a cifs-resource it works.
So, workaround is right now;
CIFS resource is a cluster resource,

start script and stop script in the cluster starts and stops cifs
and this makes CIFS work from both servers in the cluster.

BUT,, and here's the BIG problem,
cifsstop last in the cluster unload script works initially,
but directly after the cluster resource have been migrated or stopped,
CIFS.nlm autoloads again...... Always after the unload script is

This then creates 2 CIFS resources in the network with the same name
while only one of them actually holds the resource.

Question in the end;
Why the CIFS.nlm autoload after doing a cifsstop in the unload-script?
Is there anyway to prevent this auto-loading of cifs ?