Hi John,

We did this about 3 years ago to one of our ADSM servers. The approach that we 
took was to duplicate the entire ADSM database  onto the new machine and then 
remove about half of the client from each machine. Once the database was 
duplicated we locked every node defintion where the node should be backing up 
to the other server. Then we deleted the filespaces and node defintions.

Of course this will only work if you use collocation for your tape storage 
pools and you do have to be careful as to which tapes get checked into each 
machine but it was relatively painless and a lot less work than having to do 
exports.

An additional advantage is that to a certain degree, one all of the nodes are 
backing up to their respective servers you have as long as you want to remove 
the redundant data from the database.


regards,

Trevor