-----Zoltan Forray wrote: -----

>Please correct me if I am wrong, but I was under the impression that
>if a Windows node does NOT have a NODENAME statement in it's dsm.opt,
>it simply asks the OS for the hostname and uses that.  This value is
>also used for the filespace names.

The documented default is the Windows machine name. This is often the
same as the IP host name, but it is possible for the two names to be
different.

[Material removed]

>Or does the scheduling process use the registry keys, not caring
>about the hostname changed?

Our experience with renamed Windows nodes indicates that the scheduler
service locks in the node name in effect at the time when dsmcutil
was used to create the service. I don't know whether this behavior
is documented. The service presumably uses the registry to store the
node name. When a Windows node is renamed, we tell the system
administrator do remove and recreate the service.