No more AP manager IP address remain

brock0150 · ‎03-25-2008

I have a few AP's that were working fine, had some power issues and they went offline. Now they will not reassociate with the controllers. I keep getting this error message.

*Mar 1 00:00:38.955: LWAPP_CLIENT_ERROR: lwapp_name_lookup - Could Not resolve CISCO-LWAPP-CONTROLL

ER

*Mar 1 00:00:48.959: %LWAPP-5-CHANGED: LWAPP changed state to JOIN

*Mar 1 00:00:48.959: %LINK-5-CHANGED: Interface Dot11Radio0, changed state to administratively down

*Mar 1 00:00:48.960: %LINK-5-CHANGED: Interface Dot11Radio1, changed state to administratively down

*Mar 1 00:00:55.959: LWAPP_CLIENT_ERROR_DEBUG: spamHandleJoinTimer: Did not recieve the Join respon

se

*Mar 1 00:00:55.959: LWAPP_CLIENT_ERROR_DEBUG: No more AP manager IP addresses remain.

*Mar 1 00:01:00.963: LWAPP_CLIENT_ERROR_DEBUG: spamHandleJoinTimer: Did not recieve the Join respon

se

*Mar 1 00:01:00.963: LWAPP_CLIENT_ERROR_DEBUG: No more AP manager IP addresses remain.

*Mar 1 00:01:01.025: %SYS-5-RELOAD: Reload requested by LWAPP CLIENT. Reload Reason: DID NOT GET JO

IN RESPONSE.

*Mar 1 00:01:01.025: %LWAPP-5-CHANGED: LWAPP changed state to DOWN

Any suggestions or ideas would be appreciated.

svanguilder · ‎03-25-2008

I a similar issue after a power outage. I hope someone has a solution to this.

brock0150 · ‎03-25-2008

I can get it to work when i set the ip of one of the controllers manually on the ap. lwapp ap controller ip address command, it works fine. But I use to not have to do that. It would be a pain to have to track down the AP everytime it lost power and set a controller address, but at least it works that way.

jakew · ‎03-25-2008

A normal LWAPP discovery/join exchange between a WLC and AP looks like this:

1. AP sends LWAPP Discovery Request to WLC

2. WLC sends LWAPP Discovery Response to AP

3. AP sends LWAPP Join Request to WLC

4. WLC sends LWAPP Join Response to AP

If you follow the log messages, you can see the AP never got the LWAPP Join Response from the WLC. The log message "No more AP manager IP addresses remain" indicates the AP has exhausted it's candidate controller list and received no LWAPP Join Responses.

So, now you have to figure out why. You might take a swag and check the date/time on your controller is current. If it's defaulted, your AP certificates will be out of the valid date/time range and so the WLC won't validate them.

I'd start by running the following debug on your AP:

debug lwapp client event

That'll tell you that the AP is sending out the LWAPP Join Request to the right controller. If it's not, then you have to investigate what's going with the discovery phase.

Next, check that the LWAPP Join Request is arriving at your controller. At your controller console:

debug lwapp events enable

You should see a message that indicates the LWAPP Join Request has arrived from your AP. It'll also tell you if it sends an LWAPP Join Response.

The point is just walk through the LWAPP state machine and find the point where it's failing.

HTH...

megathumpzilla · ‎04-09-2008

I was getting that error until I turned the switchports the controller connected to to pure trunk, no native vlan.

It was a basic configuration blunder on the switch.

l.mourits · ‎04-10-2008

Could be anything, you will have to check eventlog on the controller(s) to see if any particular reason is mentioned there.

One possible cause could be that you have one ap-manager interface and no LAG configured. Since each ap-manager interface can only support up to 48 LAPs it will not send a JOIN ACCEPT message for the 49th LAP.

HTH,

Leo

Scott Fella · ‎04-10-2008

How are you staging the ap... DNS, Option 43, same subnet as the management ip?

Here is a thread regarding MS DHCP and using option 43:

http://forum.cisco.com/eforum/servlet/NetProf?page=netprof&forum=Wireless%20-%20Mobility&topic=Getting%20Started%20with%20Wireless&topicID=.ee7c7c3&CommCmd=MB%3Fcmd%3Dpass_through%26location%3Doutline%40%5E1%40%40.1de0085e/17#selected_message

-Scott
*** Please rate helpful posts ***

ericgarnel · ‎04-10-2008

Sounds like your APs are not releasing addresses, bounce them and it should clear up the issue. I had something similar when we were swapping out our 1020 radios for 1252 radios. When the old radio was disconnected, it did not gracefully release the address. I also opened up the scope to a 23 bit network range and that helped as well and provided for future growth.

beabrams · ‎05-08-2008

Also rememebr if your doing LAG that all the ports have to be plugged into the same switch and that the switches etherchannel load balance has to be src-dest-ip.... since its the only thing the WLC understands.