WLC WAN Backup

[Mike Farnschlaeder]

WLC WAN Backup

My customer would like to have a Backup-WLC-Controller at their main site.
On the remote site he has a 5508 with actually 6 accesspoints.
On the mainsite he has also a 5508 which acts as backup only. They have a 2Mbit WAN between this two sites.

I configured in "Wireless - Global Configuration" the IP of the Backup-controller and a heartbeat.
When pulling the plug of the Remote-WLC, all accesspoint are moving to the backup-controller of the main site. This works perfect.
When I plug in again the controller of the remote site, the accesspoints to not jump back.
They only jump back when I configure on each AP under "High Avalability" the Primary Controller. Ok, the customer only has 6 APs, but what if they had more the 100 or so. Do I have to put in the IP of the primary controller on each AP or is there another way???

Next question: The client on the remote site is logged in to the WLAN and has an DHCP-IP and Gateway of the local site.
In case of WLC-failure the client gets a new IP via the DHCP of the Backup-Controller-site. That works fine.
But when the controller comes back and all APs are also jumping back, the client fails.
I have to shutdown its network card and activate it again. After that the client finds the WLAN-SSID and gets a DHCP address again.
How can I solve this problem without always reloading the network card of the client? Is this a windows problem. Tried with XP and 7.

[Andrew Betz]

Hi Mike,

Great job on running through the AP failover and fallback capabilities on the WLCs.  It is important to note that the AP only performs AP fallback from an       unconfigured controller to a configured controller       (Primary/Secondary/Tertiary).  So, if an AP without a primary WLC specified joins WLC-A, and fails over to WLC-B, the AP will not fall back. 

As you mentioned, if the Primary WLC is set, and the AP fails over to another unconfigured WLC, the AP will fallback automatically (once the primary is back online).  This can pose a problem as you noted when clients are currently connected.  To avoid such network inturruption you have a few options:

1) Disable AP Fallback on WLC

2) Add both WLCs to the High Availablility settings to prevent fallback.

This will allow for you to move APs back to the primary during a maintence window to migitate client inturruption. 

In larger deployments the High Availablility settings are typically pushed down via a Lighteight AP Template within WCS.

Here's a great doc that explains AP failover and fallback on the WLC:
http://www.cisco.com/en/US/tech/tk722/tk721/technologies_tech_note09186a00807a85b8.shtml#ap-fall

Hope this helps.

Cheers.

Drew

[Leo Laohoo]

Try H-REAP.