[Support] - Unifi-Controller: Unifi. Unraid. Reborn.

[andyd]

19 hours ago, PeteAsking said:

It seems like unraid is killing the docker. Check you didnt run out of ram globally. Test by stopping all VM’s and All dockers and starting only this one. If it starts then you have some issue above the container you must resolve. 

 

Globally as in the amount of RAM I have installed on the server? I have 48GB - it's never above half that. Didn't have any VMs running at the time either. I'll check though what's going on with usage.

 

18 hours ago, bmartino1 said:

If you have the extra parameter, which is in the current template for downloading from the community app store. The extra parameter is set to 8 GB which is a pinning parameter you would want to lower that.

Yes I would recommend lowing the other 2 to 2 GB. Then try again.

Ok I'll see about getting the latest template

[PeteAsking]

1 hour ago, andyd said:

 

Globally as in the amount of RAM I have installed on the server? I have 48GB - it's never above half that. Didn't have any VMs running at the time either. I'll check though what's going on with usage.

 

Ok I'll see about getting the latest template

Ok let us know

[andyd]

As a follow up to the memory issue I was having the other day...

 

I lowered the limit to 2048 and set extra parameters to 4GB

 

But ever since adding the container my server has been experiencing issues with Docker. Obviously it could be coincidental but hard to ignore.

 

1. I saw in a comment that the container should be on bridge network - I had it on a custom network ... does it need to be bridge?

 

2. I see port 1900 is being used by Plex again - should I remap the Unifi controller to another port like 1901?

[PeteAsking]

7 hours ago, andyd said:

As a follow up to the memory issue I was having the other day...

 

I lowered the limit to 2048 and set extra parameters to 4GB

 

But ever since adding the container my server has been experiencing issues with Docker. Obviously it could be coincidental but hard to ignore.

 

1. I saw in a comment that the container should be on bridge network - I had it on a custom network ... does it need to be bridge?

 

2. I see port 1900 is being used by Plex again - should I remap the Unifi controller to another port like 1901?

Test by stopping all dockers except the Unifi one. If the issues go away then you have some sort of conflict. This troubleshooting step is mandatory if you want to get to the bottom of it. 

[andyd]

Yeah I’ll try that but since the issues don’t manifest till a few days later I’m thinking it makes sense to ensure setup is correct. 
 

should I be changing the port? Network doesn’t matter?

[PeteAsking]

2 hours ago, andyd said:

Yeah I’ll try that but since the issues don’t manifest till a few days later I’m thinking it makes sense to ensure setup is correct. 
 

should I be changing the port? Network doesn’t matter?

A custom network should be fine so long as you are not running into the docker macvlan issues unraid have been trying to fix. I’m personally not affected so can’t tell you much about that. 

[bmartino1]

16 hours ago, andyd said:

As a follow up to the memory issue I was having the other day...

 

I lowered the limit to 2048 and set extra parameters to 4GB

 

But ever since adding the container my server has been experiencing issues with Docker. Obviously it could be coincidental but hard to ignore.

 

1. I saw in a comment that the container should be on bridge network - I had it on a custom network ... does it need to be bridge?

 

2. I see port 1900 is being used by Plex again - should I remap the Unifi controller to another port like 1901?

 

Regarding networking stuff, all I can say is review previous post in forum...:

and

I recommend Macvlan for unifi to fix a great deal of things... I recommend using the custom network to static asign its ip and to remove port conflict.
 

Every 5 days tells me your server may be experiencing pre-hardware failure. Reboot the machine with a monitor and keyboard and run memtest.

We have to assume a great many things. So, unless you are not running a standard unraid system per the unraid docs... Such as using or edit the go file to do other task. The issues you are experiencing are a per client issues related to your setup/hardware.

 

port 1900 is teh ssdp used by upnp per iana guidelines: https://www.speedguide.net/port.php?port=1900

yeah networking and predesigned ports... https://www.iana.org/assignments/service-names-port-numbers/service-names-port-numbers.txt

it is hard to redirect shared ports. Both your plex and unifi should be set to docker custom network and giving a statin lan ip...(depending on your configurations... - custom eth0 / custom br0 / custom bon0) Doing this fixes same port sharing. Depdingin g on the docker it can be hard to fix and use/change ports. port 1900 in plex is the mindlna server, while 1900 on unifi is a hardware device inform connection port. If you are using bridge (docker default network) both docker are in conflict receiving and using bad data from both. 

[Heciruam]

Hello,

sorry if I ask so boldly. I just have no idea what I am doing wrong.

So I set this up with the 11notes/unifi:8.1.127-unraid tag and everything else standard except the network type which is br0 with a defined IP adress.

When I check the log I see this
 

{"time":"2024-05-09T20:06:12.000Z","type":"INF","msg":"starting unifi controller"}

And when I try to access the webUI I get the following message on the page
 

Bad Request
This combination of host and port requires TLS.

Does anybody know what I'm doing wrong?

Thank you in advance.

[PeteAsking]

5 minutes ago, Heciruam said:

Hello,

sorry if I ask so boldly. I just have no idea what I am doing wrong.

So I set this up with the 11notes/unifi:8.1.127-unraid tag and everything else standard except the network type which is br0 with a defined IP adress.

When I check the log I see this
 

{"time":"2024-05-09T20:06:12.000Z","type":"INF","msg":"starting unifi controller"}

And when I try to access the webUI I get the following message on the page
 

Bad Request
This combination of host and port requires TLS.

Does anybody know what I'm doing wrong?

Thank you in advance.

Put https:// in front of the address

 

eg:

 

https://IPADDRESS:8443

[Heciruam]

34 minutes ago, PeteAsking said:

Put https:// in front of the address

 

eg:

 

https://IPADDRESS:8443

Wow. That was embarrassing. Thanks!

[Dynamiteboy]

When restoring from backup on unifi:8.1.127-unraid no users seem to migrate which does not allow me to login.

 

mongo --port 27117 ace --eval "db.admin.find().forEach(printjson);" shows no admin users. I remember my password fine for the local login, but nothing seems to work. Ideas?

[PeteAsking]

2 hours ago, Dynamiteboy said:

When restoring from backup on unifi:8.1.127-unraid no users seem to migrate which does not allow me to login.

 

mongo --port 27117 ace --eval "db.admin.find().forEach(printjson);" shows no admin users. I remember my password fine for the local login, but nothing seems to work. Ideas?

Did you not have the ‘Sync Local Admins with SSO’ option ticked?

[Dynamiteboy]

9 minutes ago, PeteAsking said:

Did you not have the ‘Sync Local Admins with SSO’ option ticked?

Where is that option? and should it be ticked or unticked? I had a local account setup under the owner account. I even created a separate "unraid" admin account for testing and neither are allowing me to login.

 

 

Edited 20 hours ago by Dynamiteboy

[PeteAsking]

29 minutes ago, Dynamiteboy said:

Where is that option? and should it be ticked or unticked? I had a local account setup under the owner account. I even created a separate "unraid" admin account for testing and neither are allowing me to login.

 

 

How did you create the backup?

[Dynamiteboy]

4 hours ago, PeteAsking said:

How did you create the backup?

Download from within the Unifi Network App

[PeteAsking]

9 minutes ago, Dynamiteboy said:

Download from within the Unifi Network App

It seems this is a common issue with the Unifi controller. Here is an example here: https://community.ui.com/questions/Logins-not-working-after-restoring-backup-to-new-machine/023647bc-2953-421f-ac05-4a19072d702c

under legacy interface can you go to settings and then ‘admins’ and review the list of admins there and possible create a new one. Under the modern interface the sso option that links your unifi account on their website to your local deployment is under system - advanced and allows you to remote manage it at https://unifi.ui.com

you can try enabling this also and seeing what happens after a restore. 
 

let me know after doing some troubleshooting like this. 

 

 

[Dynamiteboy]

18 minutes ago, PeteAsking said:

It seems this is a common issue with the Unifi controller. Here is an example here: https://community.ui.com/questions/Logins-not-working-after-restoring-backup-to-new-machine/023647bc-2953-421f-ac05-4a19072d702c

under legacy interface can you go to settings and then ‘admins’ and review the list of admins there and possible create a new one. Under the modern interface the sso option that links your unifi account on their website to your local deployment is under system - advanced and allows you to remote manage it at https://unifi.ui.com

you can try enabling this also and seeing what happens after a restore. 
 

let me know after doing some troubleshooting like this. 

 

 

I don't see a settings -> admins in the legacy interface. You sure it still exists in Network 8.1.127? I am running this on a Dream-Machine that has a built-in controller as well.

Also not seeing that sso option setting in system -> advanced in the new interface.

I already tried creating a separate admin account as you can see as well. Unless my Mongo query is wrong, doesn't look like there a user there at all. I am restoring from 8.1.127 -> 8.1.127 so this is pretty much a non-starter.

 

This is the command I was running mongo --port 27117 ace --eval "db.admin.find().forEach(printjson);"

https://community.ui.com/questions/How-to-recover-access-to-a-unifi-controller-when-the-admin-password-has-been-lost-/bea32e54-c8cd-4d08-ba98-b836acebfeb4

 

All that outputs is like my email and name which shouldn't be the username?

 

Edited 14 hours ago by Dynamiteboy

[PeteAsking]

6 hours ago, Dynamiteboy said:

I don't see a settings -> admins in the legacy interface. You sure it still exists in Network 8.1.127? I am running this on a Dream-Machine that has a built-in controller as well.

Also not seeing that sso option setting in system -> advanced in the new interface.

I already tried creating a separate admin account as you can see as well. Unless my Mongo query is wrong, doesn't look like there a user there at all. I am restoring from 8.1.127 -> 8.1.127 so this is pretty much a non-starter.

 

This is the command I was running mongo --port 27117 ace --eval "db.admin.find().forEach(printjson);"

https://community.ui.com/questions/How-to-recover-access-to-a-unifi-controller-when-the-admin-password-has-been-lost-/bea32e54-c8cd-4d08-ba98-b836acebfeb4

 

All that outputs is like my email and name which shouldn't be the username?

 

This seems to be a know issue on the unifi forums. Can you try the site export and site import method?

 

https://help.cloudunifi.com/unifi/migrating-sites-with-site-export-wizard/

[Dynamiteboy]

5 hours ago, PeteAsking said:

This seems to be a know issue on the unifi forums. Can you try the site export and site import method?

 

https://help.cloudunifi.com/unifi/migrating-sites-with-site-export-wizard/

Sure, what do I need to delete from the docker folders in order to restart the setup process from the beginning?

[PeteAsking]

16 minutes ago, Dynamiteboy said:

Sure, what do I need to delete from the docker folders in order to restart the setup process from the beginning?

Just remove the docker and ssh into the server and under app data remove any reminants of the docker if there is anything. Then install again and it will be blank.