Automatically down after successful mesh creation

AP03(MPP) —— FIT mode

AP05(MAP) —— FAT mode

AP06(MAP)—— FIT mode

AP3 establishes a mesh link with AP5 and AP6 through radio1, and after the establishment, AP03 will be down automatically, and it can be recovered only by restarting the PoE interconnect interface of AP3. However, after the recovery is completed, AP03 will automatically go down again after a while, and the Idle state can only be recovered by rebooting.

1. Check the AP configuration, the configuration is correct and there are no special configurations

#
wlan ap AP03 model xxx
 serial-id xxx
statistics-interval 50
led-mode normal
vlan 1
hybrid-remote-ap enable
portal-service enable
interface wlan-mesh 1
  mesh-port link-type trunk
  mesh-port trunk permit vlan all
radio 1
  channel 44
  radio enable
  channel band-width 40
  mesh-profile 1
  mesh-interface 1
  mesh peer-mac-address xxxx-xxxx-xxxx
  mesh peer-mac-address xxxx-xxxx-xxxx
radio 2
  channel 140
  radio enable
  channel band-width 40
 radio 3
  radio enable
provision
  ip address 192.168.100.3 255.255.255.0
gigabitethernet 1
ten-gigabitethernet 1
  port link-type trunk
  port trunk permit vlan all
wlan ap AP06 model xxx
 serial-id xxx
vlan 1
radio 1      
  channel 44
  radio enable
  channel band-width 40
  service-template 2
  service-template 1
  mesh-profile 1
  mesh peer-mac-address xxxx-xxxx-xxxx
radio 2
radio 3
gigabitethernet 1
gigabitethernet 2
#

2. According to the AP capwap tunnel debug logs, the AP is dropped because a Join message was received in the Run state, and the AC will reset the AP to the online state, generating a single drop record.

[12:47:34]*Jul 20 12:47:33:999 2023 WX1840H CWS/7/FSM: [AP ID: 3] Run -> Idle.
[12:47:34]%Jul 20 12:47:34:000 2023 WX1840H CWS/4/CWS_AP_DOWN: CAPWAP tunnel to AP AP03 went down. Reason: Processed join request in Run state.
[12:47:34]*Jul 20 12:47:34:204 2023 WX1840H WMESH/7/EVENT: [APID:0x3, radio ID:0x1] Received radio down event.
[12:47:34]*Jul 20 12:47:34:217 2023 WX1840H WMESH/7/EVENT: [APID:0x3, radio ID:0x2] Received radio down event.
[12:47:34]*Jul 20 12:47:34:225 2023 WX1840H WMESH/7/EVENT: [APID:0x3, radio ID:0x3] Received radio down event.
[12:47:34]%Jul 20 12:47:34:304 2023 WX1840H APMGR/6/APMGR_AP_OFFLINE: AP AP03 went offline. State changed to Idle.

3. The reason for the downtime is that the AP sends an Echo message to the AC but the AC does not reply, suspecting packet loss on the wired side.

4. Checking the AC diagnostics reveals that AC 1/0/1 and ports receive many TC packets.

%Jul 20 11:22:37:520 2023 WX1840H STP/6/STP_NOTIFIED_TC: Instance 0"s port GigabitEthernet1/0/1 was notified a topology change. %Jul 20 11:22:40:244 2023 WX1840H STP/6/STP_NOTIFIED_TC: Instance 0"s port GigabitEthernet1/0/1 was notified a topology change.

5. After troubleshooting due to AP05 and AP06 connected to the same switch, resulting in a loop at the moment of the mesh up, and thus packet loss occurs.

Connecting AP05 and AP06 to Different Switches Problem Solved.