Wireless Access

I see that the GRE striping IP address is configured in the AP-system profile and would like to know how this address is used. I have seen on a previous post that one radio uses the LMS-IP address to build a GRE tunnel across one port and the GRE striping address is used to build the tunnel endpoint for the other radio across the other port. Is this correct assessment of the process?. In the event of an AP connecting to a local controller how does the controller know to use the GRE striping address (particularly in the event of a failover when the backup LMS-IP address is being used). There doesnt seem to be any documentation on how this really works.

#AP225

The user guide states -

"In the upstream direction when the AP transmits GRE frames to the controller, the bonding driver must be in active-active mode and not in the default active-standby mode to allow link aggregation."

What does this mean in practice? and where can i change/check this setting?

#AP225

This is speaking of the LACP config on the switch. If using dynamic LACP both ports need to be active. If using static simply build the portchannel. As far as the AP needing to use a backup LMS IP, as long as the Controller can communicate on the GRE Striping address you'll be ok. No controller will actually own that address on any interface.
#AP225

If I have two controllers with 50% of the APs on each and using LMS and backup-LMS to load balance them I assume I would need two different GRE striping IP addresses one in each ap system profile?

#AP225