8 hours ago, mgutt said:

Is there a reason why this container defaults to "bridge" network? For me it's only reliable with "host" network (else it does not adopt devices after restart).

Bridge works fine if all the ports are set properly and your devices know the correct hostname to connect to for adoption. @jademonkee hit all the major points.

 

I'm using bridge and new out of the box AP's find the controller just fine, and as long as the override is set in the controller options everything continues to work. I have 5 AP's on the local LAN and 30+ managed over the internet. My set inform points to a FQDN which forwards in to the container without issue.

Also using bridge here without issues.

@linuxserver.io

Can you add this to the template description?

https://forums.unraid.net/topic/78060-support-linuxserverio-unifi-controller/page/29/?tab=comments#comment-929058

 

I think it would be useful as the problem reoccurs.

On 11/20/2020 at 3:41 AM, jonathanm said:

I'm not, directly anyway. I keep database backups, which I should be able to apply to a fresh instance. I have all my relevant unifi ports exposed on a subdomain, so it's easy to spin up another unifi controller on whatever is handy and point the port forwards to the appropriate ip and internal ports.

hi @jonathanm can you teach me how you set this up? my unifi docker has the same ip address as unraid server, if i spin up another unifi controller it would need a different IP. i believe my APs look for the controller via the IP address, which i assume is the IP address of unraid that we used for the "override inform host" setting.

i would like to be able to easily spin up a backup unifi controller if ever my unraid server has to go down for a few days.

9 minutes ago, limawaken said:

i believe my APs look for the controller via the IP address, which i assume is the IP address of unraid that we used for the "override inform host" setting.

 

On 6/17/2021 at 1:40 PM, jonathanm said:

My set inform points to a FQDN

 

11 minutes ago, limawaken said:

can you teach me how you set this up?

All the needed ports are forwarded through my firewall, and my FQDN resolves to my WAN address. If I need to point to a different controller app I just change the fowarding rules to the new LAN ip. Your firewall MUST properly support hairpinning / loopback.

18 minutes ago, jonathanm said:

On 6/18/2021 at 1:40 AM, jonathanm said:

My set inform points to a FQDN

 

is this the same as setting the override inform host to the FQDN in the controller GUI?

 

19 minutes ago, jonathanm said:

All the needed ports are forwarded through my firewall, and my FQDN resolves to my WAN address. If I need to point to a different controller app I just change the fowarding rules to the new LAN ip. Your firewall MUST properly support hairpinning / loopback.

doing it this way means that the unifi controller gui will also be accessible from the internet?

can it be set up for local access only?

2 minutes ago, limawaken said:

doing it this way means that the unifi controller gui will also be accessible from the internet?

can it be set up for local access only?

Yes.

 

I haven't tried not forwarding the GUI port, but I guess in theory it might work if all the other needed ports are forwarded correctly. I need remote access to the GUI, as 90% of my managed AP's are outside my LAN. It's https with a strong password, so I'm not too worried about it.

 

Here is the article I used when configuring my setup.

https://help.ui.com/hc/en-us/articles/218506997-UniFi-Ports-Used

Hi Team

 

Trying to set up Email notifications in Unifi Controller :latest. When trying to use either gmail or hotmail with the following credentials i am unable to send out. 

 

smtp.gmail.com > all combinations of ports 25, 465 and 587 with SSL checked / unchecked 

smtp.office365.com > all combinations of ports 25, 465 and 587 with SSL checked / unchecked

 

I am wondering if this an issue with SMTP issue with GMAIL/HOTMAIL as they routinely use 2FA now, or a docker communications limitation? My technical skills are limited.  

 

Some help would be appreciated. 

On 6/26/2021 at 10:39 AM, jonathanm said:

Yes.

 

I haven't tried not forwarding the GUI port, but I guess in theory it might work if all the other needed ports are forwarded correctly. I need remote access to the GUI, as 90% of my managed AP's are outside my LAN. It's https with a strong password, so I'm not too worried about it.

 

Here is the article I used when configuring my setup.

https://help.ui.com/hc/en-us/articles/218506997-UniFi-Ports-Used

There are alternative things you can to to harden this more such as locking the firewall rules down to the public IP the site connects from or if its a dynamic IP having the remote site establish a wireguard connection back to the primary site so that the inform URL can be an internal IP. If both those options are unavailable and the remote site has a dynamic ip another thing you can do is setup a dyndns service on the internet router there and then have a device on your primary network via a script write the ip of the remote site to a text file that can be imported as an alias to the firewall every 15 minutes or so so that if the IP changes the firewall rule is still locked down to an IP. This would ensure after 15 minutes that the AP's are allowed through to the firewall in case of an ip change.

All of a sudden this container won't load.  When trying to bring up the WebGUI all I get is "Cannot Connect".

 

There are no errors in the logs and I haven't even logged into the GUI in months because everything worked.

 

Tried reinstalling it, changing branches, changing to bridge/host/br0, deleted all appdata and copied it back over.

 

What's going on?  How do I fix this?

I had a similar issue today where the mongod process inside this container ran away with memory upon start up, before the UI was accessible. I had a 6G memory limit on the container as it has been quite a memory hog in the past with slow leaks -- bumping that up to 15G allowed whatever initialization process to complete and now it's running fine and I was able to lower the memory limit again.

I just installed this docker on my unRAID server and left everything as default. When I start it I get Execution Error. If I change the network type to Host, then it does start, but I can't access the WebUI. This is all new. I just bought ubiquiti stuff, not a migration.

7 hours ago, Lectoid said:

I just installed this docker on my unRAID server and left everything as default. When I start it I get Execution Error. If I change the network type to Host, then it does start, but I can't access the WebUI. This is all new. I just bought ubiquiti stuff, not a migration.

I figured it out. Another docker was using the same port. I changed the Unifi port, but that didn't work. Had to change the port on the other docker.

5 hours ago, Lectoid said:

I figured it out. Another docker was using the same port. I changed the Unifi port, but that didn't work. Had to change the port on the other docker.

Just out of curiosity: are you running it as Bridge or Host? Bridge is preferred and will work fine as long as you set the correct ports and then set the hostname for adoption in the settings.

Just go back a page in this topic to see more info on that.

On 5/25/2021 at 8:47 PM, PeteAsking said:

I updated to 6.2.25 also on the weekend. It seems actually a good version. I have had no issues and you can close the advert these days also.

 

My switch firmware is 5.43.36.12724 and the AP's are now 5.43.36.12724 so unifi have internally made the version numbers the same even though they are 2 different firmware files and sizes. Im sure this wont cause any confusion and be overlooked by anyone at some point in the future.

 

Despite this it is all working so guess this version is good to go.

 

Pete

6 weeks later, how goes it? Any problems from the upgrade to v6? Anything you like about it more than v5?

I'm too afraid to even upgrade the firmware on my APs to the new v5 from v4 without also making the jump to controller v6.

No problems, all seems to be working. As you cant provision the new devices without upgrading it seems like its just a matter of time until you will be forced to update anyway. Especially if you want that wifi 6 speeds the new equipment have. This seems like a pretty good version except for the fact you have to bounce back and forth between the classic and new interface to expose all options which is annoying. There was a video on youtube from that pfsense guy about the new version and these annoying quirks if you want to look that up but otherwise all good.

 

 

10 hours ago, jademonkee said:

6 weeks later, how goes it? Any problems from the upgrade to v6? Anything you like about it more than v5?

I'm too afraid to even upgrade the firmware on my APs to the new v5 from v4 without also making the jump to controller v6.

 

46 minutes ago, PeteAsking said:

No problems, all seems to be working. As you cant provision the new devices without upgrading it seems like its just a matter of time until you will be forced to update anyway. Especially if you want that wifi 6 speeds the new equipment have. This seems like a pretty good version except for the fact you have to bounce back and forth between the classic and new interface to expose all options which is annoying. There was a video on youtube from that pfsense guy about the new version and these annoying quirks if you want to look that up but otherwise all good.

 

 

 

Just asking this because I’m lazy and don’t want to hunt for your previous post, are you running latest or is it another version?

On 7/8/2021 at 4:04 PM, wgstarks said:

Just asking this because I’m lazy and don’t want to hunt for your previous post, are you running latest or is it another version?

repository is set to linuxserver/unifi-controller:version-6.2.25 on my docker.

Would anyone know why this Unifi controller (V5.14.23) docker container would be trying to access something at the stretchoid.com domain?  I noticed a log entry in my firewall that this container is trying to access this domain.

1 hour ago, mifronte said:

Would anyone know why this Unifi controller (V5.14.23) docker container would be trying to access something at the stretchoid.com domain?  I noticed a log entry in my firewall that this container is trying to access this domain.

Have you visited http://stretchoid.com/ ?

They lay it out pretty clearly.

 

I'm definitely not advocating blindly filling out their form, but maybe after some due diligence you could choose to interact with them or just ignore it.

I have updated to version-6.2.26 today just fyi.

35 minutes ago, PeteAsking said:

I have updated to version-6.2.26 today just fyi.

Why?

13 hours ago, PeteAsking said:

I have updated to version-6.2.26 today just fyi.

Let us know what you think of it in a couple weeks. I still haven't taken the plunge, although it seems like most of the big problems were sorted in this latest release.

15 hours ago, jonathanm said:

Why?

 

Its received multiple fixed (this version) and people seem happy on the forums. A small subset of users are having issues with vlans and a memory leak but nobody can work out what causes the people affected to have the issue. I see the docker image is regularly updated (last update 1 day ago) for this version so this particular one seems to be the one they are working the most on and have released no new version for over 2 months now.

 

2 hours ago, jademonkee said:

Let us know what you think of it in a couple weeks. I still haven't taken the plunge, although it seems like most of the big problems were sorted in this latest release.

 

Will do. So far Im having no issues whatsoever.

I have had no issues and its working as expected.