1

I have a pair of Cisco 5508 controllers with close to 50 APs shared between the two. After some maintainence - changing from (the unsupported) LACP to LAG for the port-channel I have repeated issues with WAPs rejoining controllers.

The first section shows the console output from an AP that performs the re-join. The second section shows the disparity between the AP up time and the Association up time.

*Jan 15 15:25:59.004: %DTLS-5-SEND_ALERT: Send FATAL : Close notify Alert to 10.0.0.102:5246
*Jan 15 15:25:59.096: %WIDS-6-DISABLED: IDS Signature is removed and disabled.
*Jan 15 15:26:10.000: %CAPWAP-5-DTLSREQSEND: DTLS connection request sent peer_ip: 10.0.0.102 peer_port: 5246
*Jan 15 15:26:12.135: %CAPWAP-5-DTLSREQSUCC: DTLS connection created sucessfully peer_ip: 10.0.0.102 peer_port: 5246
*Jan 15 15:26:12.139: %CAPWAP-5-SENDJOIN: sending Join Request to 10.0.0.102
*Jan 15 15:26:31.589: %DTLS-5-SEND_ALERT: Send FATAL : Close notify Alert to 10.0.0.102:5246
*Jan 15 15:26:42.000: %CAPWAP-5-DTLSREQSEND: DTLS connection request sent peer_ip: 10.0.0.102 peer_port: 5246
*Jan 15 15:26:44.268: %CAPWAP-5-DTLSREQSUCC: DTLS connection created sucessfully peer_ip: 10.0.0.102 peer_port: 5246
*Jan 15 15:26:44.268: %CAPWAP-5-SENDJOIN: sending Join Request to 10.0.0.102
*Jan 15 15:26:47.064: %CAPWAP-5-JOINEDCONTROLLER: AP has joined controller CONTROLLER2
*Jan 15 15:26:59.005: %WIDS-6-ENABLED: IDS Signature is loaded and enabled

OUTPUT FROM A "Show AP Uptime":

AP Name      AP Up Time                Association Up Time
------------ -----------------------   -----------------------
WAP01        375 days, 11 h 23 m 12 s   261 days, 01 h 19 m 52 s
WAP02        445 days, 09 h 21 m 28 s   222 days, 02 h 20 m 31 s
WAP03        215 days, 05 h 02 m 03 s   215 days, 05 h 00 m 26 s
[output cut]
TEST-AP      0 days,   01 h 21 m 51 s   0 days, 00 h 35 m 48 s
WAP04        136 days, 21 h 40 m 31 s   0 days, 00 h 14 m 50 s
WAP05        432 days, 10 h 20 m 00 s   0 days, 00 h 11 m 14 s
WAP06        332 days, 09 h 35 m 09 s   0 days, 00 h 03 m 38 s

Software version 8.2.130.0

Improve this question
6
  • Are you sure your LAGs are functioning correctly? Is the problem on both controllers?
    – Ron Trunk
    Commented Jan 15, 2018 at 17:26
  • Yeah, I'm pretty sure and the issue is on both controllers. Switch side they're configured "mode on" and as a trunk.
    – Odyssey
    Commented Jan 15, 2018 at 17:42
  • Are you sure there's not any MAC flap or something in the log of your switches?
    – user36472
    Commented Jan 15, 2018 at 18:10
  • Also, have you rebooted the WLC after enabling LAG?
    – user36472
    Commented Jan 15, 2018 at 18:11
  • No MAC flap, and the WLC was rebooted after enabling LAG (but that over a year ago). It wasn't rebooted after changing from LACP to LAG on the Catalysts.
    – Odyssey
    Commented Jan 16, 2018 at 9:30

1 Answer 1

Reset to default
0

The issue was with the LAG/Etherchannel after changing from LACP to LAG (mode on). Resetting (shut -> no shut) the port-channel interface resolved the problem.

A reboot of the WLC would have probably also resolved the issue.

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.