I've noticed that some of our 3.6 GHz 450 APs end up using the on board GPS rather than the signal through the power port. These are all running 13.4, and they're all set to autorun + free run. None of them have cables in the timing ports.
Sync Pulse Source : On-board GPS
Sync Pulse Status : Receiving Sync
Sync Pulse Status - On-board GPS : Receiving Sync
Sync Pulse Status - Timing Port/UGPS : No Sync
Sync Pulse Status - Power Port : Receiving Sync
UGPS Power Status : Power Off
Satellites Visible : 21
Satellites Tracked : 14
Checking the event log gives the following:
11/23/2015 : 14:15:49 MST : :Time Set
11/23/2015 : 14:15:59 MST : Acquired sync pulse from Power Port.
11/24/2015 : 07:23:52 MST : Lost sync pulse from Power Port.
11/24/2015 : 07:23:52 MST : Acquired sync pulse from On-board GPS.
11/24/2015 : 07:23:54 MST : :GPS Date/Time Set;
01/26/2015 : 06:44:09 MST : :Time Set
11/24/2015 : 09:14:54 MST : :Time Set
11/28/2015 : 16:59:49 MST : Lost sync pulse from On-board GPS.
11/28/2015 : 16:59:49 MST : Acquired sync pulse from Power Port.
So it looks like if the power port sync is lost briefly, the AP keeps using the on-board GPS indefinitely rather than switching back once the power port sync returns. The only time they revert back to the power port is when the on board GPS drops.
My first question is: does this actually matter? Would having two different sync sources cause issues? It might explain a really strange intermittent problem a customer is having. Maybe having the GPS antenna for the CMM at a different height than the APs leads to the sync being slightly out of phase?
If it is a problem, what do I do about it? Can I disable the on board GPS through a telnet command?