I’m running into some issues getting an Unifi USG adopted into the controller. The controller is ran locally. I have factory reset the USG and was able to navigate to its default IP and set the LAN1 IP so it would be on our subnet. The site is already established.
I have added it to the network and am still able to log into the USG and ping it. But when I log into the controller I cannot get it to show up in the list of devices. While in the USG it has a field for the controller inform URL. I read on anther forum that you have to enter it multiple times for it to finally show in the controller. I have entered it so many times it isn’t funny.
I dug down through the list of clients and found it there.
I guess my next step would be to factory reset it again and start over. Maybe I set something up wrong initially.
I will also add that we do already have 1 USG adopted in the “default” site within the controller. But I have added another site that I am wanting to add the new one to. Would this cause any problems to anyone’s knowledge?
something’s weird, it should just work. What does your USG use for DNS? is your controller on https / is your cert good?
(e.g. I’m using https://somehost.domain.com/ and I initially forgot to setup nginx reload in my letsencrypt cron… it’s good that devices check the cert)
You would think that to adopt the USG into the network you would plug into the LAN port. Well for some reason that isnt the case and you have to plug the WAN port of the USG into the network. As soon as I did that I was able to adopt the USG.
Next problem now though is I can’t and don’t want to leave the USG plugged in the WAN port. If I try to swap it over to the LAN port it will disconnect from the controller. Then even if I plug it back into the WAN it will not reconnect.
Either way you If you can ssh into the USG, you can run the set-inform command/script to change the controller address. (e.g. http://192.168.0.11:8080/inform) - that should just work.
Unifi devices have this loop that just makes http (or https) requests to the controller every 30s to report state and check for any configuration updates.
Optionally they might try to reach the controller using STUN (udp nat port punching service), which would then allow the controller to make changes to the device more frequently, but it’s not strictly required.
As long as they can reach the controller via http (or https) it’s all good - your controller will have a green online dot.
If you’re running on a lan, and your controller changes IPs, your device will freak out and go into this “can’t reach the controller mode”. Your controller is expected to notice that and fix the inform URL.