Network Connect & Active/Passive Cluster Routing Problem

[bcampbell3_]

Using Network Connect, users are having problems routing in a certain situation. Our 6.3R2 SA
6000 devices are configured in an active/passive cluster. If the active node (access1)
is selected, users are able to use Network Connect and route just fine. If I choose
"fail-over" and the passive node (access2) is selected, routing for our Network Connect
users breaks.

On a side note, I'd like to know why our active/passive cluster is failing over by
itself. I get no error messages, just on rare occasion the active/passive cluster will
switch fail-over node... that is how we discovered the Network Connect routing issue.

I've created a case with tech support, but I was looking for any input from the community as well.

Thanks!

[firewall72_]

Hi,

I would check Maintenance, Troubleshooting, System Snapshot and check for Watchdog and Process snapshot logs. These can be uploaded to JTAC for analysis. With regards to your NC routing. Are you using IP Pools? The reason I ask is because I've come across return routing issues in this scenario. I would check to make sure your NC return route is pointing to the right IP, as well as the correct arp entry for the passive node once it becomes active.

-John