APs suddenly lose connection to WLC 5508

[rponte]

Hello,

I have a WLC 5508 with 50 APs and there are 2 APs 3501 that suddenly lose connection to the WLC. When they lose connection i can ping them and i see them in the Switch with cdp.

They are in two differents branches and in those branches are others APs that work perfectly.

I Validated it is not a license problem.

When i shutdown the port in the Switch connected to the AP and turn it on again it registers in the WLC.

What do you think about the problem?

Thanks in advance.

[Leo Laohoo]

1.  Are the APs in FlexConnect?

2.  What firmware is/are the controllers running on?

[rponte]

1.  Are the APs in FlexConnect?

No, all the APs are in Local Mode. The problem is just with 1 AP in each branch because i have other that does not loose connection

2.  What firmware is/are the controllers running on?

8.0.121.0

[Leo Laohoo]

8.0.121.0????

Console or remote into the affected AP when they get disconnected from the controller.  

Post the output to the following AP commands: 

1.  dir; 

2.  sh ip interface brief; and 

3.  sh logs

[rponte]

1. dir:

Directory of flash:/

2 -rwx 402 Jan 18 2017 19:17:22 +00:00 env_vars
3 -rwx 56278 Mar 12 2016 13:47:57 +00:00 event.log
4 -rwx 64 Mar 12 2016 13:47:48 +00:00 sensord_CSPRNG0
87 drwx 1152 Mar 12 2016 13:47:34 +00:00 ap3g1-k9w8-mx.153-3.JA5
40 -rwx 27576 Dec 9 2011 16:44:46 +00:00 lwapp_reap.cfg
41 -rwx 280 Nov 20 2011 15:49:41 +00:00 lwapp_officeextend.cfg
5 -rwx 7192 Jan 18 2017 20:42:11 +00:00 private-multiple-fs
6 -rwx 75 Jan 18 2017 19:17:22 +00:00 capwap-saved-config
45 drwx 0 Mar 1 2002 00:10:56 +00:00 configs
11 drwx 128 Mar 1 2002 00:09:46 +00:00 ap3g1-rcvk9w8-mx
8 drwx 1152 Mar 12 2016 14:25:49 +00:00 ap3g1-k9w8-mx.153-3.JA4

2. sh ip int brief

Interface IP-Address OK? Method Status Protocol
BVI1 10.122.142.32 YES TFTP up up
Dot11Radio0 unassigned NO unset up up
GigabitEthernet0 unassigned NO unset up up

3. sh log

*Jan 19 00:29:57.001: %DTLS-5-SEND_ALERT: Send FATAL : Close notify Alert to 10.122.138.100:5246
*Jan 19 00:30:05.000: %CAPWAP-5-DTLSREQSEND: DTLS connection request sent peer_ip: 10.122.138.100 peer_port: 5246
*Jan 19 00:30:35.003: DTLS_CLIENT_ERROR: ../capwap/base_capwap/dtls/base_capwap_dtls_connection_db.c:2214 Max retransmission count reached for Connection 0x51658D8!

[rponte]

On the WLC:

debug capwap errors enable

*spamApTask2: Jan 19 15:51:54.186: Unable to get AP Bundled Version. Using Controller Version!!!

*spamApTask7: Jan 19 15:52:03.850: Unknown AP type. Using Controller Version!!!

[Leo Laohoo]

Can you please upgrade the firmware of the controller?

[rponte]

is there any other possible solution for this issue? 

It is very difficult to do a maintenance windows for the client.

[Leo Laohoo]

Can you attach the the AP file "event.log" into this thread?  I want to see the contents.

[rponte]

Yes, i can. I attached it

[Leo Laohoo]

What model of switch is this AP connected to?  

Is this AP powered using PoE or injector?

[rponte]

Is this AP powered using PoE or injector?

Yes, it is. But I validated that the PoE was enough to the AP. I checked into the switch the PoE given to the AP.

I don't have the model right know, I'm sorry about that 

[Leo Laohoo]

Yes, it is. But I validated that the PoE was enough to the AP. I checked into the switch the PoE given to the AP.

Ok, so if the APs are on PoE, then can you check the interface of the AP for any line errors. 

Next, check to see if there are any issues with the switch uplink port.  

[rponte]

I did it, and the interface doesn't have errors.

And there is no issues with the uplink port too.

[Leo Laohoo]

Upgrade the firmware.