Thanks for your response...Is having a mobility group a requirement for inter-controller roaming? I can see a benefit of having a mobility group in controller redundancy, but other than that, what sets it apart?
At a minimum, your 2000 controller is on *THREE* networks: The management interface, the AP manager interface, and *AT LEAST* one interface for an SSID.
Therefore, I don't understand your question. The mobility group will work as long as the controllers can communicate with each other. The more important question is what happens to the wireless client data once it leaves the controller.
In most all cases, multiple controllers *MUST* terminate the client VLAN traffic on the same layer 2 infrastructure. This limits the topology significantly.
Isn't the 2000 controller EOL and doesn't it have significant mobility restrictions anyway?
Thanks for your response. We have 2 controllers that are on the same subnet. Do I need to create a mobility group for clients to roam from an access point joined to one controller to an access point joined to the other controller?
Roaming is an 8 hour conversation in itself, but I will answer your question with a 'yes', you should have a mobility group defined if wireless clients may move between APs associated with different controllers.
Roaming actually has much more to do with the wireless security in use, the config of the client and back-end user database, and the layer 2 connectivity of the multiple controllers.
If you are using WEP or WPA pre-shared-key with the same layer-2 termination on the controllers, then your users really aren't 'roaming' at all, they are constantly re-associating to the different APs anyway.
Roaming, in my mind, means 'fast roaming' meaning less than 100 ms. This would require either Cisco proprietary CCKM, or some of the *sort of* WPA2 fast-reconnect features.