01-26-2015 06:18 PM
Does anyone have a step by step for enabling email on the Amazon instances? My understanding is that Amazon only allow TLS email out, and the controller can only send SSL out, so the fix is to install a SMTP client on the server. A step by step for installing the SMTP client would be very helpful.
Thanks in advance.
01-27-2015 12:02 AM - edited 01-27-2015 12:09 AM
Just a short note for those having trouble with adopting their APs (going from Adopting to Disconnected). I was SSHing into my APs one at a time and performing a set-inform via CLI but it would never take (when I adopted in controller it would go from Adopting to Disconnected after 30 seconds or so and never leave disconnected state). It wasn't until I launched the Discovery tool on my laptop (had it installed on desktop but desktop still couldn't see APs, not sure why) and clicked Manage and then set the inform that way, then I was able to successfully adopt each AP without it going to Disconnected state.
Just to check, you know that you need to do the set-inform twice - once before adopting and once afterwards?
01-27-2015 07:58 AM
Nope, did not know that! I kept reading the three steps it printed after I ran the first set-inform and it didn't make much sense, but now it does. Thanks!
01-27-2015 09:00 AM
Related question - have a running instance with the wizard at step 1. When I attempt to restore from backup of my current UniFi configuation, the process fails out.
If I bang on the restore function for awhile occasionally it goes to 'restoring and restarting' but then returns to the wizard as thoughnothing has happened.
Using a .unf file created on a Mac but doubt that matters given the .unf file format, no?
I thought this might be a controller version issue - but I'm running 3.2.7 and I believe the instance is as well.
Thanks in advance....
01-27-2015 11:48 AM
Also, to clarify - I'm attempting to restore from the .unf file via the web through to the instance. Perhaps this .unf file needs to be local? I may be getting a port blockage while attempting to send the file via the web.
Apologies again for being a bt of a noob aorund Ubuntu - can anyone share the SSH language needed to upload the file to the correct directory on the instance and then how to access the file via the wizard?
Thanks in advance.
02-03-2015 03:13 PM
So someone sent me an updated guide for setting up the AWS Controller. I go to the elastic ip I set up which brings me to a setup wizard for the controller. I go through the wizard in which it has me create a username and password. After the wizard it takes me to the login screen. But when I put in my username and password I just created, all I get is "invalid username and password." Any one else experience this or know why its doing that? Thanks.
02-04-2015 02:18 AM - edited 02-04-2015 02:25 AM
I have run into this before - my hunch is that you've got a different set of credentials cached on your machine from a different login. My experience has been that the UniFi software looks to that cached info regatdless of whether you're looking at a local 'instance' or one in the cloud.
A quick way to test would be to enter any other credntials you may have used on that specific computer before. I'll bet the prior credentials work correctly.
Not sure if this behaviour is intentional or a bug.
Also - can you say a little more about the more current step-by-step? I'd be curious to see what you've got.
02-04-2015 02:24 AM
A bit more clarity on this issue - the trick is that I had unknowinglt installed the 2.x version of UniFi and was attempting to restore using a 3.x backup file. I'm guessing they're not backward compatible, which is what led to the errors I was seeing.
When I deleted the prior installation and re-installed using the 'unifi-rapid' designation, the v3.2.9 wizard appeared and the restore worked just fine.
I can also verify that installing the v4.6 beta works properly and can be restored from the same v3.2 backup.
There's nothing on the wizard landing page to indicate which version is in play. I'll post a screengrab of the v3.x wizard - slightly different-looking from the 2.x version - so anyone else who has this same experience can tell what version they're working with.
02-04-2015 08:15 AM
Here is the upgated guide I used.
Ironically, I used the exact same password that my software unifi controller uses; so that shouldn't be the problem. Someone suggested that if I have special characters in my password; that could be the issue. And I do have a '?" in my password. So might redo my intance and try the wizard agian.
02-04-2015 01:27 PM
Thanks for that how-to. Adds some good context.
FWIW, I have a special character in my password and haven't nad an issue with tranistioning from one controller to another.
02-08-2015 02:42 AM
Really struggling here and any help greatly appricated.
Just installed a UAP pro and set up AWS cloud service. everything has gone so fluidly until now. I followed the helpful advice on this forum and various walkthroughs on youtube.
Where i am, the controller is running ok and i can click about on it in the AWS cloud. I have set all port rules and checked them again.
I can see the UAP using the discovery tool, I go to set it to 'inform'. it says pending and then adopting but it never pops up on the controller.
i am setting the Pointing the device to https://<myelasticIP>:8080/inform is that correct??
In the following tutorial ( https://www.youtube.com/watch?v=Qqc-OgmCByw ) it explains using a ping tool to check that device is talking to the server. Being a massive newby to this can someone explain how i do that??
Like i say any help or new ideas very very welcome
02-08-2015 10:55 PM
Thanks for the idea but sadly that didnt work . . . .
Also in my last post i gave the wrong the link, Please follow as below . . .
How does he use the Ping tool??
02-09-2015 04:24 PM
This tutorial by bradynapier has cut to the chase for getting the AWS instances running and updated. Your precise step by step provided everything that was needed and put an end to the frustrating searches for the tips of others that did not work. Thanks
02-09-2015 08:29 PM
This guide has been updated:
If followed, it works flawlessly. My last hurdle, is that I was using the wrong AMI. Be sure to use the one the guide has pictured.
02-18-2015 08:22 AM
I have a UniFi AP-LR running lastest beta version and currently the controller is installed on a local computer. This access point is already adopted and running OK.
I want to change the controller to a EC2 instance which i already have up and running too.
When i run the Discovery Utility tool, the acess point wont show here.
Im running version 4.6.0
Will appreciate your help!
02-18-2015 08:29 AM
Only unadopted APs show up in the discovery utility, or adopted APs where the controller is offline. Also, the discovery tool is L2 only, if you are running in an EC2 instance, you'll need to follow the L3 adoption process.
02-18-2015 08:40 AM - edited 02-18-2015 08:43 AM
Hello and thanks for your reply.
Im not familiar with the L3 adoption process. Could you please provide me some extra info so i can read about it and give it a try?
I also tried this. dont know if im doing it wrong or not. i did it but still cant see the access point on my EC2 controller.
configuring Ubiquiti UniFi APs Remotely. Find AP's Local IP. SSH to AP. Default Usr/Pw= ubnt/ubnt Run command to Adopt APs: set-inform http://UniFiServerIP:8080/inform
02-18-2015 08:53 AM - edited 02-18-2015 08:55 AM
One important note - won't address the issue you're running into here, but still valuable with EC2 instances - if you haven't already, get a (free) 'Elastic IP' from AWS and use that in your inform URLs. Essentially acts as a static IP for your EC2 instance and will avoid issues later on.