Hi, I'm hoping someone here has a similar experience who can advise me
on how to solve the problem.

Right now I have a 2-node cluster which seems to be running fairly
smoothly. I've got three LUNs, one for SBD, one for iPrint and one for

My problem is with particularly on migrating the iPrint resource - I
have not brought the iFolder resource online yet.

Now the resource is working fine and dandy. I can bring it up and down
on either of the node with no problem. But if I choose the 'Migrate'
option, it goes into a comatose state.

E.g. on node 1 I've got the resource iprint_service running. I can
bring it offline and bring it back online to node 2. However, if I
migrate the resource directly from node 1 to node 2 (i.e. without going
offline), it becomes comatose.

I've narrowed the issue to being with the EVMS, I think. After the
resource is comatose, on the destination server (node 2 in the example
above), there is a process "evms -b". This one accepts the command for
deporting and activating the container. On the other node (node 1),
there is a worker process for EVMS.

I know that these processes have to be invoked in order for the
migration to work. The problem is that after the resource becomes
comatose, these processes are stuck there. I remember reading somewhere
that killing the processes will not solve the problem. The only way is
to reboot.

Does anyone know why the resource can be brought online/offline with no
problem but cannot be migrated??