Hi there.
Theoretically you can do this in NDS Manager & ConsoleOne just by deleting the server object.
However, I've seen problems with both of these. Your best bet is to use DSREPAIR.
Here is the TID for this operation
1. Go to the server that contains the master replica of the partition.
2. Using DSREPAIR, verify that time is synchronized among all servers in the replica ring (Load dsrepair -- Time Synchronization). If time is synchronized, you can proceed with step 3; if not, you need to deal with your time sync problem.
3. Using DSREPAIR, find out which servers are in the replica ring.
4. Go to the console of each server in the replica ring & run DSREPAIR -a. Go to the replica ring & remove the crashed server from it.
5. On the console of the server that holds the master, we will force a sync of DS:
set dstrace=on set dstrace=+s set dstrace=*h
You're done when you see something like
End sync of partition <partition name> All processed = YES
6. Gto ConsoleOne & remove the server object, its volume objects, SAS service, and any other references to it.
------------
Bill
Consultant / Network Engineer
CNE, CCNA