[Support] Linuxserver.io - Unifi-Controller


Recommended Posts

anyone know if I can switch from the brettm357 docker to the linuxserver.io flavor, and keep all my config settings?

I'd like to be able to "lock" in the version I have without fear of it accidentally upgrading if I fat/ham/booze finger the update button(s).

Link to comment
4 hours ago, PeteAsking said:

So an update.

 

I can tell you that AP's and Switches on this new version(I dont have a USG or any of that stuff) all work fine on this version. No CPU issues or anything and speeds all normal.

I am using TAG: version-6.1.71

Firmware for AP's : 4.3.28.11361

Firmware for switch: 5.43.23.12533

FYI, that's the same AP firmware version that's current on 5.14.23-ls76. I have no clue about switch FW, all I manage are ap's.

 

I wonder if the new v6 AP's are running that same FW revision.

Link to comment
6 minutes ago, jonathanm said:

I wonder if the new v6 AP's are running that same FW revision.

I am still running the 5.14.23 controller and recently added a U6 Lite AP.  It provisioned just fine and installed an update to the 5.43.23.12533 firmware.  Seems to be working fine.

 

My AC APs are on firmware version 4.3.28.11361 and show no updates available.

 

There are available updates to 5.43.23.12533 for the USG and switches, but I have not installed these updates yet.

 

The U6-LR AP requires controller 6.0.34 or higher.

Link to comment
On 3/22/2021 at 1:18 PM, jademonkee said:

Any idea why my Unifi Controller is using so much RAM (4.887GiB)?

I don't know how much it usually uses, but my server almost never goes above 6GB RAM usage but is now currently up at 10GB, so I'm guessing Unifi is the culprit.

See attached.

unifiRAM.png

 

EDIT: A restart of the Docker brought usage back down, but I'd love to know if anyone else has had a similar experience.

 

unifiRAM_restart.png

 

So, after limiting the Docker RAM allocation to 4GB, I tried to log into my Docker this morning and it threw up a web socket error, then a weird 404 (404 was written in red inside the interface, not an actual 404 from the browser).

Looking at the usage, it was up hard against 4GB RAM.

So I had to restart it.

As per recommendations in this topic, I've now disabled DPI as it was only a mildly useful feature for me, anyhow.

I really don't see how constantly growing RAM usage is acceptable. And given the constant complaints on the forum about v6.x, I've grown to regret forking out the money for the USG to replace my EdgerouterX.

Every single feature that it brought into the Unifi interface has been a worse experience than on the EdgerouterX, and it offers absolutely no benefit (I have a 500Mbps connection, so it's not fast enough to handle threat protection (which I only found out AFTER I bought it) and the in-built speed test can't handle an internet connection of that speed, either).

So yeah, not feelin' the love for Ubiquiti right now. Which sucks, coz the WiFi hardware has been great - fast and reliable (except for the problem with Chromecasts introduced in a fw update and promptly fixed).

But the Controller... just terrible. I hope that, by the time v7 rolls around, Unifi Controller will be a fully fleshed out, reliable, usable product.

  • Thanks 1
Link to comment

I'm using some EOL APs (I think they're called UAP v2). There is a version of the controller released by Ubiquiti that is dated after their EOL, suggesting it will not allow configuration changes to EOL devices. It appears all the tags for this docker are deprecated, is there a way to version lock this so I can continue to "upgrade all", or do I need to be vigilant about never upgrading it?

Link to comment
2 minutes ago, jonathanm said:

What do you mean by that? Many (most?) people using this container are using tags to specify the version they want, currently I'm still using linuxserver/unifi-controller:5.14.23-ls76

 

I was looking at the documentation on dockerhub, which says:

Quote

Version Tags

This image provides various versions that are available via tags. latest tag provides the latest stable build from Unifi, but if this is a permanent setup you might consider using the LTS tag.

TagDescription

latestreleases from the latest stable branch.

LTSDEPRECATED - releases from the now EOL 5.6.x "LTS Stable" branch.

5.9DEPRECATED - releases from the now EOL 5.9.x branch.

5.8DEPRECATED - releases from the now EOL 5.8.x branch.

5.7DEPRECATED - releases from the now EOL 5.7.x branch.

 

Your method of versioning isn't documented on dockerhub https://hub.docker.com/r/linuxserver/unifi-controller

Link to comment

I just updated the docker to the latest version.  Running 6.1.71.  If I log in locally, there are no adopted devices.  If I go in through the cloud it shows the site online with the right number of devices, clients, etc, but when I click launch, it connects and there is nothing there.  If I relaunch the controller, I'll get alerts that all of my devices have reconnected, but nothing there in the GUI.

 

anything obvious I might be missing?

 

Thanks,

 

-dev

Link to comment

reverted to backup from a few days ago and same issue.  Reverted the docker image back to one from a month ago and got:

 

UniFi Controller startup failed

We do not support upgrading from 6.1.71.

 

Hoping I don't have to go to my appdata folder backup as I've made some changes since that was last Sunday

Link to comment
On 3/27/2021 at 1:56 PM, PeteAsking said:

You set the override inform url in the controller somewhere I forget the exact place. 

 

Hey, thanks for that. I set this up and it seemed to work until today. Today inform IP got overwritten with dockers internal IP again, so this doesn't help.

Link to comment

Solved: Totally not the right password, it looks like the controller was using some other login, I dunno. I found the correct one, and it is NOT the cloud/gui one I updated, so thats that...

 

Original Post: 

Anyone point me in the correct direction?

 

I can't log into my instance any longer. I only have this app/docker and one AP. 

 

I changed my ui.com password recently (since their breech was a lot worse than they let on). I always had issues logging in, honestly, and had to use the password from the cloud, but now neither the old or new one works on the local docker instance, so no idea what I did.

 

Any ideas if I can change/reset it? Or should I just make an entire new docker instance/login and let it refind my AP?

Edited by 00b5
Solved
Link to comment
  • 2 weeks later...

I'm seeing this in the logs:
 

[services.d] starting services
[services.d] done.
2021-04-27 10:36:23,747 main ERROR Error processing element InMemoryAppender ([Appenders: null]): CLASS_NOT_FOUND

2021-04-27 10:36:24,096 main ERROR Unable to locate appender "InMemoryAppender" for logger config "root"

 

Link to comment
58 minutes ago, Redindian said:

I'm seeing this in the logs:
 



[services.d] starting services
[services.d] done.
2021-04-27 10:36:23,747 main ERROR Error processing element InMemoryAppender ([Appenders: null]): CLASS_NOT_FOUND

2021-04-27 10:36:24,096 main ERROR Unable to locate appender "InMemoryAppender" for logger config "root"

 

Yes. Same. Tried both the linuxserver and the BrettM versions. Both have the same error. I can’t even get the BrettM version to run anymore. It just keeps crashing. 

Edited by slipknottin
Link to comment

This error seems to occur starting version 6.1.71, but a number of other people reporting this using Raspberry Pi or Synology say it is working fine despite this error. I have also noticed increased CPU usage when it fails. My docker is limited to 2GB of memory. Could this be related? I will try and remove this limit and see how long it runs then.

Link to comment
52 minutes ago, Redindian said:

I'm seeing this in the logs:
 


[services.d] starting services
[services.d] done.
2021-04-27 10:36:23,747 main ERROR Error processing element InMemoryAppender ([Appenders: null]): CLASS_NOT_FOUND

2021-04-27 10:36:24,096 main ERROR Unable to locate appender "InMemoryAppender" for logger config "root"

 

I routinely updated the docker today and saw the same error. Not sure if it's because of the update or happened before. Current version for me is 6.1.71

Link to comment

I've installed the Unifi-Controller docker using the default option, which in this case is version 6.1.71.  Unfortunately Unifi has declared my APs EOL and so any further updates to the Controller past 6.1.71 will break my ability to configure them.  I needed something newer than 5.9 to get my USW-Aggregation working otherwise i would have left it at that.  My questions are as follows:

 

1. Will leaving the docker repository poined at: linuxserver/unifi-controller run any risk of it updating itself past the current version?  If so is there a way to disable this?

 

2. Is there a way to save a complete copy of the current docker build (not just the configuration) to restore from if needed in future and the "latest" branch has moved forward?

 

Thanks in advance for any advice!

 

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Restore formatting

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.