Not that it matters, In my situation, this is what worked.
Update MPROUTE CONFIG Update sysname TCPIP Update directory entry for TCPIP set vswitch vswitch1 disconnect set vswitch vswitch1 rdev 2f00 set vswitch vswitch1 connect force tcpip autolog tcpip set vswitch vswitch2 disconnect set vswitch vswitch2 rdev 2700 set vswitch vswitch2 connect For our configuration, I also had to do a set vswitch vswitch1 discon set vswitch vswitch1 prirouter set vswitch vswitch1 connect set vswitch vswitch2 discon set vswitch vswitch2 prirouter set vswitch vswitch2 connect Assuming everything works cprel a vmlink maint cf1 Update SYSTEM CONFIG to reflect new rdev's and prirouter vmlink maint 193 cpsyntax system config rel z (det rel x (det cpacc maint cf1 a My backout plan consisted of From another VM system def mdisk 1198 8785 9 resvol <== tcpmaint's 198 disk acc 1198 e update mproute config with old names update sysname tcpip with old addresses rel e (det I had previously updated vmsecure config to authorize operator to manage TCPIP. I would use that to update the directory entry for tcpip. Then set vswitch vswitch1 discon set vswitch vswitch1 rdev 1e08 set vswitch vswitch1 connect force tcpip autolog tcpip set vswitch vswitch2 discon set vswitch vswitch2 rdev 1f08 set vswitch vswitch2 connect I didn't have to exercise the backoff plan, but there were connection issues for the Linux systems. Those were resolved by making the vswitche s prirouter (they were not defined that way previously).
