Isabelle,



Yes I have experienced this before. Typically what I'll do is "shut" the port the Secondary CAS prior to powering up the Primary CAS. The ensures the Primary starts up properly and doesn't try to fight with secondary for control.



I would "shut" the port to the Secondary and restart the Primary. Once the Primary is up and settled out, "no shut" the port for the Secondary and restart it. You may want to flush the ARP cache on the switches they're connect to also. 



CH



-----Original Message-----

From: Cisco Clean Access Users and Administrators [mailto:[log in to unmask]] On Behalf Of Isabelle Graham

Sent: Tuesday, June 23, 2009 09:58 AM

To: [log in to unmask]

Subject: Problems upgrading to 4.1.3.1



We run HA fail-over pairs for our CAM and CASs and we have just upgraded from 4.1.2.1 to 4.1.3.1. 

The CAM pair upgraded with no problems, but after upgrading each CAS pair (and bringing the pair 

back online) we ran into a problem: neither CAS can see the other (both show active, peer node dead) 

so both try to act as active. This causes all sorts of problems as both servers are then fighting 

over the service IP. So far as we can tel no settings were lost in the upgrade. Has anyone else seen 

this (or something similar) who would be able to share insight?



Thanks!



-- 

Isabelle Graham

Information Security

American University