well, it should be quite straight forward is my guess;

Client needs to change their internal network at the mainoffice for
the clients. Probably we will NOT do a complete redesign of the IP net
for the servers;

today; net
change to; net

Servers, switches, etc are planned NOT to change IP,
only clients and access directly toward the Netware resources.

For all "standalone" NW servers, we simply add a 2nd binding and the
set the NCP and SLP to include that network.

On the cluster ( 2 server cluster ) ; we do the same, but also we need
to adjust all cluster-scripts/bindings to reflect the new IP.

Should this be an OK operation and hopefully a stable solution ?
The alternative would be to have a machine or a router/swich that
connects both the IP networks,, though this could become a bottleneck.

Any tips or ideas ?