6 hours ago, flosken said:

How does you settings look like now? I'm having the same issue and also uses NPM...

hmpf....I swear it worked yesterday...just to be broken again today...so maybe it wasn't NPM related....will investigate further....If you have any findings yourself, please let me know. Yesterday I thought that maybe de-activating Cache Assets in NPM finally did the trick.

Using 9.3-unraid tag currently. I updated to 9.3.45 around 10 hours ago. No issues to report

On 7/30/2025 at 10:25 AM, flosken said:

How does you settings look like now? I'm having the same issue and also uses NPM...

I can't get it to go. As soon as I close my browser (no matter which one I try) completely, the session cookie seems to be gone and I have to login again. If I "only" close the browser tab, without closing the whole browser, the login cookie is still working, so with a new tab of the controller webpage, I am automatically logged in.

But maybe this is all "expected behavior"? I found this (quite old) reddit:

https://www.reddit.com/r/Ubiquiti/comments/8ke4by/does_the_remember_me_option_work_for_anyone_on/

Someone (it seems to be an Unifi employee) posted the following statement:

"Remember me basically means that if you keep the window open, it won't log you out after 10 minutes. I am not 100% sure what happens if you close all tabs (I am more on the firmware side, not backend/UI)".

Edited August 2, 2025Aug 2 by JayDee73

How can I fix this issue here: I have already increased the memory to 4K, and I don't have too many other docker containers running, on a system using overall 32 GIGs of RAM.

I have only two unifi access points

Thanks in advance for help

Edited August 4, 2025Aug 4 by ozkhan1

50 minutes ago, ozkhan1 said:

How can I fix this issue here: I have already increased the memory to 4K, and I don't have too many other docker containers running, on a system using overall 32 GIGs of RAM.

I have only two unifi access points

Thanks in advance for help

What do you mean by ‘increased to 4k’?

1 hour ago, PeteAsking said:

What do you mean by ‘increased to 4k’?

Sorry, changed to 4GB , not K.

37 minutes ago, ozkhan1 said:

Sorry, changed to 4GB , not K.

The docker requires 8gb to function properly as certain times the database maintenance temporarily uses more ram for a short period.

1 minute ago, PeteAsking said:

The docker requires 8gb to function properly as certain times the database maintenance temporarily uses more ram for a short period.

I actually increased it to 10 GB and now it is stuck here: (My system overall has 32GB RAM)

5 minutes ago, ozkhan1 said:

I actually increased it to 10 GB and now it is stuck here: (My system overall has 32GB RAM)

Its hard for me to know what happens if the database kept crashing and ooming due to lack of memory and what this has done to the integrity of the database. Hopefully you can restore a working version and apply 8gb as per the deployment xml and it will work correctly going forward. The 8gb requirement has come up many times in this thread and every time people who have reduced it have created problems for themselves. Once its damaged there is not a lot I can do to repair it.

Edited August 4, 2025Aug 4 by PeteAsking

Hi,

Every 1-2 days the container stops with java error. It runs perfectly fine for a while which is strange. My flag is 11notes/unifi:9.3.43-unraid

I see this is log:

{"time":"2025-07-29T16:37:12.000Z","type":"INF","msg":"starting unifi v9.3.43"}
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  [thread 32927 also had an error]
SIGBUS (0x7) at pc=0x000014ae728192de, pid=1, tid=111
#
# JRE version: OpenJDK Runtime Environment (17.0.15+6) (build 17.0.15+6-Ubuntu-0ubuntu120.04)
# Java VM: OpenJDK 64-Bit Server VM (17.0.15+6-Ubuntu-0ubuntu120.04, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, parallel gc, linux-amd64)
# Problematic frame:
# V  [libjvm.so+0xc192de]  ObjectMonitor::EnterI(JavaThread*)+0x27e
#
# No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /usr/lib/unifi/hs_err_pid1.log
#
# If you would like to submit a bug report, please visit:
#   https://bugs.launchpad.net/ubuntu/+source/openjdk-17
#

[error occurred during error reporting (), id 0xb, SIGSEGV (0xb) at pc=0x000014ae7316d941]

{"time":"2025-08-02T08:17:44.000Z","type":"INF","msg":"starting unifi v9.3.43"}
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGBUS (0x7) at pc=0x000015360d47d412, pid=1, tid=30
#
# JRE version: OpenJDK Runtime Environment (17.0.15+6) (build 17.0.15+6-Ubuntu-0ubuntu120.04)
# Java VM: OpenJDK 64-Bit Server VM (17.0.15+6-Ubuntu-0ubuntu120.04, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, parallel gc, linux-amd64)
# Problematic frame:
# V  [libjvm.so+0xc7d412]  PerfLongVariant::sample()+0x22
#
# No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /usr/lib/unifi/hs_err_pid1.log
#
# If you would like to submit a bug report, please visit:
#   https://bugs.launchpad.net/ubuntu/+source/openjdk-17
#

[error occurred during error reporting (), id 0xb, SIGSEGV (0xb) at pc=0x000015360dda7941]

I have attached the file from the container that the error log refers to.

hs_err_pid1.log

3 minutes ago, daninet said:

Hi,

Every 1-2 days the container stops with java error. It runs perfectly fine for a while which is strange. My flag is 11notes/unifi:9.3.43-unraid

I see this is log:

{"time":"2025-07-29T16:37:12.000Z","type":"INF","msg":"starting unifi v9.3.43"}
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  [thread 32927 also had an error]
SIGBUS (0x7) at pc=0x000014ae728192de, pid=1, tid=111
#
# JRE version: OpenJDK Runtime Environment (17.0.15+6) (build 17.0.15+6-Ubuntu-0ubuntu120.04)
# Java VM: OpenJDK 64-Bit Server VM (17.0.15+6-Ubuntu-0ubuntu120.04, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, parallel gc, linux-amd64)
# Problematic frame:
# V  [libjvm.so+0xc192de]  ObjectMonitor::EnterI(JavaThread*)+0x27e
#
# No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /usr/lib/unifi/hs_err_pid1.log
#
# If you would like to submit a bug report, please visit:
#   https://bugs.launchpad.net/ubuntu/+source/openjdk-17
#

[error occurred during error reporting (), id 0xb, SIGSEGV (0xb) at pc=0x000014ae7316d941]

{"time":"2025-08-02T08:17:44.000Z","type":"INF","msg":"starting unifi v9.3.43"}
#
# A fatal error has been detected by the Java Runtime Environment:
#
#  SIGBUS (0x7) at pc=0x000015360d47d412, pid=1, tid=30
#
# JRE version: OpenJDK Runtime Environment (17.0.15+6) (build 17.0.15+6-Ubuntu-0ubuntu120.04)
# Java VM: OpenJDK 64-Bit Server VM (17.0.15+6-Ubuntu-0ubuntu120.04, mixed mode, sharing, tiered, compressed oops, compressed class ptrs, parallel gc, linux-amd64)
# Problematic frame:
# V  [libjvm.so+0xc7d412]  PerfLongVariant::sample()+0x22
#
# No core dump will be written. Core dumps have been disabled. To enable core dumping, try "ulimit -c unlimited" before starting Java again
#
# An error report file with more information is saved as:
# /usr/lib/unifi/hs_err_pid1.log
#
# If you would like to submit a bug report, please visit:
#   https://bugs.launchpad.net/ubuntu/+source/openjdk-17
#

[error occurred during error reporting (), id 0xb, SIGSEGV (0xb) at pc=0x000015360dda7941]

I have attached the file from the container that the error log refers to.

hs_err_pid1.log

Same issue as the other guy. Looks like you assigned 4gb to the docker image for some reason and overrode the default amount

For non-Unraid servers, is it still recommended to use Unraid tags with the 99:100 setting?

22 minutes ago, Iceman24 said:

For non-Unraid servers, is it still recommended to use Unraid tags with the 99:100 setting?

No this image is tested under unraid only. The tags without -unraid appended are for other kinds of boxes.

Thanks for creating this! Just noticed that I had an ancient docker image that was still running Unifi Controller v. 8.0.24 and was not being updated any more and changed to this.

Steps (for me) for a successful upgrade from 8.0.24 -> 9.3.43 (most up-to-date unraid-version of the image at the time of posting):

1. Backup old Unifi Network Application / Controller (Settings -> System -> Backups -> Download)

2. Stop old container

3. Install this container (I used default settings except I had to change port 1900 -> any free port, I used 1902 and this was due to Jellyfin using same port and it was not configurable). Once installed and start open the webUI.

4. WebUI should start fine if you are using Network type: Bridge during docker installation. Once the startup wizard shows up you can select to restore from backup and use the one you just downloaded. It says that it is doing restart but it doesn't do this automatically (unknown bug). You can open Unraid webgui in another tab and wait until the container is shut down (red light). Then you can start it again and jump back to the unifi tab.

5. Once you have got it started your devices may say that they are waiting for adoption. If so, head to Settings -> System -> Advanced -> Inform Host and override it to your server's ip (without declaring any ports, just the ip)

6. Now your devices should show up as connected but it may take up a few minutes.

7. After that I made another backup just in case and updated the image to the newest stable one that was available in the first page of this topic (in my case docker repository tag "11notes/unifi:9.3.43-unraid"). You can just edit your current template and change repository tag and applying changes pulls the newer version and all other settings should be kept identical.

In the future I think I will eventually install Unifi OS but until it's official and stable. Until then I continue to use this one.

@PeteAsking @bmartino1
Ultimately fixed the issue I was having by changing the container to host with everything else being the unraid default.
I can finally remote access!
Got the green globe and everything works!

8 hours ago, MrTroll said:

@PeteAsking @bmartino1
Ultimately fixed the issue I was having by changing the container to host with everything else being the unraid default.
I can finally remote access!
Got the green globe and everything works!

cool then the stun issue you were having is a internal docker cross talk issues between other host networks. Thanks for sharing and giving a solution.

with the release of Unifi OS. I'm not sure what the next evolutions will be like yet.

Currently using 9-unraid tag and updated to the latest controller version when it became available last night. No issues have popped up so far.

On a side note object based networking is working well in this update with me being able to route all traffic from my Apple TV's via Nord VPN quite easily

4 hours ago, fat said:

Currently using 9-unraid tag and updated to the latest controller version when it became available last night. No issues have popped up so far.

On a side note object based networking is working well in this update with me being able to route all traffic from my Apple TV's via Nord VPN quite easily

Tnx for the update

I believe 9.4.19 has not caused any problems for anyone, if this is the case I will update this to be the latest version for people to use later. Let me know if this is not the case.

• P

9.4.19 supported.

3 hours ago, YourNightmar3 said:

Im running this container with the following configuration:

Tag: unifi:9.3-unraid

Network Type: Bridge (Default)

Use Tailscale: Off (Default)

Privileged: Off (Default)

I left all the ports unchanged.

Basically running everything completely default with no changes in configuration.

But i cannot adopt my devices. It just gets stuck on "Adopting" forever.

And sometimes the devices switch to "Offline", and then back to "Adopting"

I did some searching on google and apparently the most common cause for this problem is that the docker container is telling the devices the wrong ip address to communicate to. I also see on the "Homepage" of the Unifi app (the container) that it lists it's internal docker ip address, not my Unraid server IP:

Is this part of the problem or is this fine? What can i do to solve this "Adopting" stuck issue?

Answered many many times in this thread. Update your inform ip in the settings.

I had a similar issue with a newer USW Flex 2.5g switch that just would not adopt while all my other devices had no issues. I read a ton of posts about this sort of thing and checked what my server was set to do and it did seem to be handing out the host name of my server. I overrode the inform host value to be the UnRaid server IP and it worked perfectly afterwards so I'd try that.

On 8/4/2025 at 10:47 AM, PeteAsking said:

Same issue as the other guy. Looks like you assigned 4gb to the docker image for some reason and overrode the default amount

In fairness, if you go through the community apps and install, 4gb is the default max on that. I just did the install and upped it to 8Gb based on this thread, but the app install needs to be refreshed.

1 hour ago, Sinister Crayon said:

In fairness, if you go through the community apps and install, 4gb is the default max on that. I just did the install and upped it to 8Gb based on this thread, but the app install needs to be refreshed.

That is ofd because I checked the xml and it states

“<ExtraParams>--memory=8G</ExtraParams>”

So unclear why your is 4gb

Is there an iOS app that will connect with this controller? I usually use my iPhone to connect to the webUI (I’m lazy that way 😁) and the webpage isn’t optimized for iOS. I found one app named “Unifi” but when I logged in it just showed me networks from several years ago.