- Bronze, 100 points or more
In short, APs will join the controller immediately using the recovery image, but once it downloads 7.6.100 and reboots, it either cannot join the controller or will take up to 40 minutes.
The AP is in local mode. The AP 1040 eventually joined the controller, 40 minutes after we deleted all the files off the flash which is unacceptable.
*Apr 9 15:38:15.842: %CAPWAP-5-SENDJOIN: sending Join Request to 10.10.10.10
*Apr 9 15:38:15.851: %CAPWAP-3-ERRORLOG: Invalid event 10 & state 5 combination.
*Apr 9 15:38:15.851: %CAPWAP-3-ERRORLOG: CAPWAP SM handler: Failed to process message type 10 state 5.
*Apr 9 15:38:15.851: %CAPWAP-3-ERRORLOG: Failed to handle capwap control message from controller
*Apr 9 15:38:15.851: %CAPWAP-3-ERRORLOG: Failed to process encrypted capwap packet from 10.10.10.10
*Apr 9 15:38:16.304: %LINK-6-UPDOWN: Interface Dot11Radio0, changed state to down
*Apr 9 15:38:16.362: %LINK-5-CHANGED: Interface Dot11Radio0, changed state to reset
*Apr 9 15:38:16.427: %CAPWAP-5-JOINEDCONTROLLER: AP has joined controller CONTROLLER-DC
I also had our technician console into a 2602 and we were seeing exact same errors.
SUMMARY OF THIS CASE:
--AP is 1042 and wlc is on 220.127.116.11
--using dhcp option 43 but the AP has static ip
--checked the private config on the AP and it was not sending the discovery request to the desired WLC
--cleared the private config
--primed the AP to desired WLC
--AP sending join request but receiving the following errors:
Apr 4 14:37:47.000: CAPWAP-3-ERRORLOG Go join a capwap controller
*Apr 4 14:36:47.000: CAPWAP-5-DTLSREQSEND DTLS connection request sent peer_ip: 10.10.10.11 peer_port: 5246
*Apr 4 14:36:47.169: DTLS_CLIENT_ERROR: ../capwap/base_capwap/dtls/base_capwap_dtls_handshake.c:90 First fragment for seq 2 is missing
*Apr 4 14:37:17.205: DTLS_CLIENT_ERROR: ../capwap/base_capwap/dtls/base_capwap_dtls_connection_db.c:2176 Max retransmission count reached!
--cleared the flash except rcv
--ap started downloading the image from the wlc but once it reboots the same errors appear again
--research made on the error message and apparently we are hitting a bug: