Q:
How to verify the configuration sync status on all the IAPs in a cluster?
A: There is an index called "Config ID" in "show aps" command which could be used to double check the configuration sync status for all the APs. If the config sync counters are found to be the same for all the APs across the cluster, it means all the changes applied so far on the VC, has been propagated to the slave IAPs successfully.
When all the IAPs in the Cluster are in sync, Config ID will be the same as highlighted below.
Instant-ca:54:86#show aps
3 Access Points
---------------
Name IP Address Mode Config Id Config Csum Ext SSID Active Age
---- ---------- ---- --------- ----------- --------------- ---
f0:5c:19:ca:54:86 10.10.0.3* access 5 2867 enable 4d:18h:18m:54s
f0:5c:19:c7:1c:7c 10.10.0.5 access 5 48758 disable 4d:18h:17m:47s
f0:5c:19:c7:2e:8c 10.10.0.6 access 5 48758 enable 4d:18h:16m:13s
end of show aps
As we have created applied some new configuration, post executing commit apply, Config ID would have incremented by 1.
Instant-ca:54:86# configure terminal
We now support CLI commit model, please type "commit apply" for configuration to take effect.
Instant-ca:54:86 (config) # wlan ssid-profile test
Instant-ca:54:86 (SSID Profile "test") # vlan 1
Instant-ca:54:86 (SSID Profile "test") # exit
Instant-ca:54:86 (config) # exit
Instant-ca:54:86# commit apply
Config ID incremented by 1 for all the APs in the cluster, post receiving the configuration from VC.
Instant-ca:54:86#show aps
3 Access Points
---------------
Name IP Address Mode Config Id Config Csum Ext SSID Active Age
---- ---------- ---- --------- ----------- --------------- ---
f0:5c:19:ca:54:86 10.10.0.3* access 6 2867 enable 4d:18h:44m:54s
f0:5c:19:c7:1c:7c 10.10.0.5 access 6 48758 disable 4d:18h:43m:47s
f0:5c:19:c7:2e:8c 10.10.0.6 access 6 48758 enable 4d:18h:42m:13s
end of show aps
If you consider the below example, there is a difference in Config sync ID for two slave IAPs. This means, those IAPs has not received the configuration update made on VC twice so far. This could be due to many reasons, couple of them as follows, PAPI packets(UDP 8211) that carries configuration information might have fragmented and lost, packets from VC to slave IAPs that's part of different switch stack getting dropped for a brief moment due to uncertain network issue, etc.
Instant-ca:54:86#show aps
3 Access Points
---------------
Name IP Address Mode Config Id Config Csum Ext SSID Active Age
---- ---------- ---- --------- ----------- --------------- ---
f0:5c:19:ca:54:86 10.10.0.3* access 2 2867 enable 4d:17h:23m:54s
f0:5c:19:c7:1c:7c 10.10.0.5 access 0 48758 disable 4d:17h:22m:47s
f0:5c:19:c7:2e:8c 10.10.0.6 access 0 48758 enable 4d:17h:21m:13s
end of show aps
As long as the communication between the IAPs are fine, no matter how there are separated on the network, configuration sync should be fine. If such issues are seen even after ensuring the VC/Slave communication to be fine, please open a case with Aruba TAC for further troubleshooting.
Note: Output trimmed as it's too long.