Wireless Access

We have 4 6000 controllers. We had to reboot them. Now controller 1 shows all the APs down except from controller 4. We know the APs are up and running because they are being used with no complaints. I just trying to find out why controller 1 shows the APs down even though they are up and running. Any suggestions?

Do you have any redundancy between the controllers?  Could the APs have connected to their backup controller?

Also, if the APs from controller 4 connected to controller 1 in the past, they will show as Down when they swing back to controller 1.  If you want to get rid of them from the DB, do "clear gap-db ap-name <name>".

If this is a master/local setup, the master should know the true status of all APs, whether they terminate on the master or the locals.

Sounds like these are all masters, though.

Controller 1 is the master and the other 3 are locals.

The APs should all show as up in that case, even if they are terminate on one of the local controllers. 

You are saying that on the local controller, the AP shows up and on the master that controls that local, the AP is reported as down?

Do "show switches" on the master and make sure the local controllers are reporting up.  If they are down, that would be why the APs are reporting down.

How the APs are provisioned? Do they first go to the master controller and then get the config and go to the local or are they directly provisioned with local controller's IP address only?

Also can you please post the output of the "show gap-debug" from master and local controller both?

-Alap

Most of the times restarting STM process on the master controller resolves the issue. Please note that if there are APs terminating on master controller, they will go down.

From enable mode, do 

"process restart stm core"

It will generate a crash file also under  flash. You can send crash file to TAC and ask them to find out the root cause. 

-Alap

I just opened at ticket for TAC. The problem seems to be very sporatic. I go in and 210 of the 1085 are showing up. Some are from controller 2 and some from controller 4. I click on an AP to provision it so I can change the group assignment and after about 30 seconds it comes back with "internal error" message. I log off and back in and now 20 of the 1085 are showing up. Not even a full controller. 10 minutes later 0 of 1085 are showing up. This all started since we rebooting the controllers because we had users complaining about slow connections throughout campus. I just installed a new AP yesterday and it's part of the APs showing up and then down.

Please update the thread once this issue gets resolve. It will be good to find out what triggered it and what is the solution.

I have the same issue where the Master Controller shows some of the  APs are down when in reality they are not.

I have opened a TAC but the reply was not very helpfull

Thanks

Well we basically got it fixed. As it turns out we had to create 2 seperate VLANs on the controller. 1 for the controller information and the 2nd for management. Seems strange since thats no how we had it setup in the past but it did solve our controller problems.