We're seeing an ongoing issue where several APs accross multiple sites log the error, "%CAPWAP-3-ERRORLOG: Retransmission count for packet exceeded max(CAPWAP_WTP_EVENT_REQUEST., 12)", then disassociates from the controller, and reassociates almost immediately. The issue is the users get disassociated from the AP and call the helpdesk.
A counter measure at one site was to add the CAPWAP traffic (udp ports 5246 & 5247) to the controller in our QOS Platinum policy (setting the DSCP bit to 'ef'), but that doesn't seem to help.
We're using Flexconnect with central authentication, local switching.
A couple of questions:
1) The Platinum queue on the QOS is showing over 500 kbps when the only thing put in that queue is the CAPWAP traffic - there aren't any phones. Why so much bandwidth for authentication and control traffic?
2) What is happening with the APs that they can't talk to the controller that causes the issue in the first place? Bandwidth doesn't seem to be an issue.
Below are some config and outputs:
AP-1242#show capwap reap status
AP Mode: REAP, Connected
Radar detected on:
AP-1242#show capwap reap association
REAP Data Switching: Local
2960#show int fa0/22
Hardware is Fast Ethernet
Full-duplex, 100Mb/s, media type is 10/100BaseTX
Last input 00:00:22, output 00:00:00, output hang never
Last clearing of "show interface" counters never
Input queue: 0/75/0/0 (size/max/drops/flushes); Total output drops: 0
I think you are hitting a bug (CSCse92856) specific to 1242 AP. Solution given is "Enable Proxy ARP on the default-gateway device of your AP". You can try that & see.
Even I cannot view detail of this bug as of insufficient access permission.Therefore I do not know more details about this bug fix & which software version affected,etc. Better you contact Cisco TAC & get more information.
I ran across this post while revisiting the problem so posting for all to benefit.
The excessive UDP CAPWAPP traffic was a direct result of the global multicast setting on the controller. When enabled, we were seeing ~500kbps of CAPWAPP to this site. Once I disabled multicast on the controller, the rate dropped to about 15kbps total for the site.
The APs at the remote sites are in flexconnect mode and the controller is in the data center. As a result, any multicast that the controller receives on any of its interfaces, is automatically converted to unicast and sent across the WAN to all flexconnect APs in an individual stream.
For those that need to enable/disable this setting, it was a "hitless" change for us. Did not require a reboot, did not cause an outage.
Controller --> Multicast on the left hand side --> "Enable Global Multicast" checkbox is unchecked.
Transferring Crash file from standby: Login to the Active WLC in HA.
From CLI: (Cisco Controller) >transfer upload datatype crash (Cisco
Controller) >transfer upload filename (Cisco
Controller) >transfer upload mode tftp (Cisco Controller) >transfer
This is the start of a display filter cross reference between Wireshark
and OmniPeek. The 1st installment is a table of advanced filters. More
filters will be added as time allows. It is a living doc, so check back
for changes every so often Please feel f...