Hi we are running two WLCs (126.96.36.199) in HA SSO in separate data centers; as per manual there must be a layer 2 connection between the WLC nodes. Is it supported to have routing between the WLCs? ..that would also mean that I would have to reconfigure the IPs of one WLC, so it would be in another subnet.. I don't know if this is supported either or of the management IPs have to be in the same subnet?
"To support the active and standby WLCs in different data centers, in release 7.5, back-to-back
redundancy port connectivity between peers is no longer mandatory and the redundancy ports can
be connected via switches such that there is L2 adjacency between the two controllers."
Thanks in advance
As per my knowledge WLC is a Layer 2 device. So it will not support routing.
I think both Mgmt IP must be in same subnet.
Best Practice and recommend way is to use direct connection between WLC to WLC.
Don't forget to rate helpful posts
Those routes are only for the service-port and are mend to route additional out-of-band networks within the service-port network.
HA-SSO is a concept which has been created to act when one of the two controllers physically fail or when there is a problem on the adjacent (core/distribution) switch. To be able to do this without any impact for the end-users there is no other option for the back-up controller than simply "claim" the active unit's IP addresses. This requires the controllers to be layer 2 adjacent.
Depending on your other requirements it might be a better solution to just use both controllers separate and configure them as primary and secondary controller on AP level. If you use local mode AP's your clients probably will need to go through DHCP again because of other client networks as well.
Please rate useful posts... :-)
Can somebody confirm that we don't need direct cable between the WLCs with 188.8.131.52 code in order to set up the SSO HA? We can deploy the WLC in different data centers with stretched VLANs between them, right?
As long as you span the vlan between DC's and there is no concern for latency, then you can. Just remember that it's not just the redundancy port (RP) but the management and all interfaces created on the controller. When the primary fails or there is a failover, the backup takes over all the IP address configured on the primary so make sure you span everything. If this is an issue and it might, then your best design is N+1.
*** Please rate helpful posts ***