Failure of AC800M Primary and Backup CPU's
After analyzing the Controller logs, it was discovered that both PM891 CPUs have the same IP Addresess setup that resulted in faulting both CPUs.
The extract from the Controller log:
"E 2012-04-24 12:08:35.707 NIS:Sup Primary and Backup CPU is configured with same IP address".
"0x19c0848 (tNetTask): duplicate IP address 172.17.80.173 sent from ethernet address 00:00:23:13:5e:a3 to target ethernet address 00:00:23:13:5e:bf"
Note that same IP Address 172.17.80.173 (that is being used for Primary Controller) has been setup explicitly using a IP Config Tool on Backup Controller Backup redundant network address. It should be "172.17.82.173" using default rule.
I 2012-04-18 14:23:27.106 Network Interface 172.16.80.173 (255.255.252.0 ) : 00-00-23-13-5e-be
I 2012-04-18 14:23:27.113 Network Interface 172.17.80.173 (255.255.252.0 ) : 00-00-23-13-5e-bf
I 2012-04-18 14:23:27.128 Backup Interface 172.16.82.173 (255.255.252.0 ) (default rule)
I 2012-04-18 14:23:27.130 Backup Interface 172.17.80.173 (255.255.252.0 ) (explicit set)
IP address must be setup only on Primary CPU, keeping Backup CPU in power off condition. IP Config should be performed by connecting TK212 cable to CN4 port of Primary CPU and No IP Config should be performed on the backup CPU.
When we setup the IP addresses on Primary CPU, we also define the IP Address of Backup CPU using "Default Rule". Again during this process we keep the Backup CPU in Power Off condition.
Once we set the IP Address on Primary CPU, then the IP address for backup CPU gets transferred as soon as Backup CPU gets powered ON.
If both CPUs have same IP Address setup then there will be an IP Conflict that will cause the CPUs to crash. Once the duplicate IP was resolved, Duplicate IP Address line was not present in the controller log and controllers worked without any issues. Therefore it appears that duplicate IP Address conflict was the cause of CPUs crashing.