[Support] ich777 - Application Dockers

[ich777]

18 minutes ago, Rkpaxam said:

does the ssd need to be btfs, zfs or xfs or doesnt it matter? 

That doesn‘t matter.

[Rkpaxam]

ok, i dont know what is causingg it then, like i said i have tried multiple container versions of sql, followed a video found online but i get the same error everytime. It has to be somthing unraid related but i cant figure out what 

Edited April 21 by Rkpaxam

[ich777]

14 minutes ago, Rkpaxam said:

It has to be somthing unraid related but i cant figure out what 

I don‘t think so but can you give me a bit more information how you configured everyhtinng, screenshot from the template and so on?

 

Maybe also try start over.

[Rkpaxam]

Maria DB had been set up with a custom BR0

 

All Variables for user/password/db are set to zabbix 

 

only things i have changed are 

 

the database location (i have this stored on a seprate drive)  i have other databases on there so no idea why this one is playing up

 

removed the links for sql injection as they are not needed 

 

https://imgur.com/a/6p69Le6

[Rkpaxam]

i belive its somthing to do with "Character set utf8 and utf8_bin collation is required for Zabbix server to work properly with MySQL database."

[ich777]

4 minutes ago, Rkpaxam said:

Maria DB had been set up with a custom BR0

And Zabbix is also in a BR0 or is it in the default bridge?

Please keep in mind that you have to enable host access to actually be able to connect to the database if in a custom bridge.

 

2 minutes ago, Rkpaxam said:

i belive its somthing to do with "Character set utf8 and utf8_bin collation is required for Zabbix server to work properly with MySQL database."

Then you have to alter the databse to support this character set.

[Rkpaxam]

all three are on cutom br0 as i didnt want cross over. i dont know how to change that but surly it should work out of the box as i havnt changed anything 

[ich777]

2 hours ago, Rkpaxam said:

all three are on cutom br0 as i didnt want cross over. i dont know how to change that but surly it should work out of the box as i havnt changed anything 

Sorry it took me a bit but I've now tried it now on my server and it starts up fine.

 

I created a database alongside with a user on my MariaDB for Zabbix, fired up Zabbix with the IP from my MariaDB instance (my Server IP), port, username, password, databasename and it was starting fine.

[Rkpaxam]

has to be somthing with my unraid setup then but i cant for the life of me figure out what. ive tryied again with 100% fresh installs and a new DB exact same issue every time. 

[HHUBS]

Is it possible to run ruskdesk server using nginx and without opening the ports? I have rustdesk server and it's working when I open the ports 21115-21119 in my pfsense. But I want, if possible, to close those ports and run it through nginx. 

Edited April 23 by HHUBS

[ich777]

17 minutes ago, HHUBS said:

But I want, if possible, to close those ports and run it through nginx. 

May I ask why? Or can you please explain the benefit?

 

You have to first understand that a reverse proxy is designed for http/https TCP traffic and RustDesk is not http/https traffic, it uses it's own encryption with the key that you have to specify, is encrypted too and also uses UDP ports.

Sure you could use streaming ports but you then also have to open up the necessary ports to your reverse proxy because it can only proxy these ports if they are open.

With all of that said, that will complicate your setup more than needed.

 

So in conclusion in general it is safe to forward the ports from the container in your Firewall because it is not http/https traffic and uses it's own encryption.

 

 

BTW I get these questions also from time to time for my game server containers and it's the same there, in general you can forward the ports but you are complicating the setup more than it is necessary and since you are just streaming the ports this also defeats the purpose of a reverse proxy.

I also can't give support for such setups.

[SudoPacman]

Anyone else having trouble with the MEGAsync container failing to login today?

 

Nothing much in the logs, but thought I'd ask before digging further.

 

Cheers

 

Edit: Turns out the container was having DNS issues.

Edited April 24 by SudoPacman

[Elmojo]

Has anyone been able to get WOL working in Rustdesk outside of a local LAN?  I've asked on the Rustdesk Github, and the answer is "you can't", but that seems...implausible.  I feel like there has to be a way to do such a simple and commonly-needed thing, especially since Teamviewer and Anydesk do it without issue.  Any ideas?

[ich777]

3 hours ago, Elmojo said:

I've asked on the Rustdesk Github, and the answer is "you can't", but that seems...implausible.

Why should that be implausible?

To send a WoL you have to be on the same Subnet and you need a device that is available on the same subnet...

 

3 hours ago, Elmojo said:

I feel like there has to be a way to do such a simple and commonly-needed thing, especially since Teamviewer and Anydesk do it without issue.

I don't think that this is possible if nothing is running Anydesk or Teamviewer on your local network since otherwise this is in fact impossible because how should they perform a WoL in your subnet? Don't get me wrong but otherwise they must have physical access to your local network through the Internet and if they are able to perform this I won't be very happy because this is a big security risk!

 

3 hours ago, Elmojo said:

Any ideas?

Do you use HomeAssistant? If yes, it would be way easier to do it via a WoL package through HomeAssistant or like I do it simply put your PC behind a switchable power outlet (WiFi, Zigbee,...), in your BIOS make sure that the action for Power Loss is set to Always on so that your PC/device is started, with that you won't need a WoL package at all (and maybe save some power since the PC is entirely off).

[Elmojo]

Thanks for the detailed reply!

Maybe I misunderstand how TV and Anydesk are working, but somehow they are both able to wake my remote machines (several of them) from sleep over the internet.

Regardless, I appreciate your explanation.  I do use HA, but only on my network here at home, not at any of the other locations where there are machines I'll need to wake.  Would it still be of any use?  I suspect no.

For now, the simplest solution has been just to leave the machines awake (never sleep) and let the monitors turn off.

We're not really concerned with power consumption, since it's not many machines in total, and they are spread out over various locations.

[ich777]

12 hours ago, Elmojo said:

Maybe I misunderstand how TV and Anydesk are working, but somehow they are both able to wake my remote machines (several of them) from sleep over the internet.

I don't know how they are doing it but it can't be a real WoL package because that simply won't work.

 

12 hours ago, Elmojo said:

Would it still be of any use?  I suspect no.

No, but if you choose the power outlet method and bind the power outlets to your HomeAssistant instance then it would be of course possible.

 

12 hours ago, Elmojo said:

For now, the simplest solution has been just to leave the machines awake (never sleep) and let the monitors turn off.

We're not really concerned with power consumption, since it's not many machines in total, and they are spread out over various locations.

Than this is your best bet.

[Elmojo]

10 hours ago, ich777 said:

Than this is your best bet.

Thanks, I'll just stick with the simple solution for now.

Edited April 25 by Elmojo

[eLpresidente]

Hey i just installed your TS3 App...

and have some questions.. i just created a Cname ts3.host.com and in NGNIX Proxy relinked it like it shoud  (maybe like it Should ) it doest work. 

 

Which port ist the right to forward to in NGINX and what do i have to put in ts3 ? 

 

Havent got it to work..

Networktype is Bridge

 

And in Teamspeak3 i just put the ts3.host.com for Serverip ? 

also tried ts3.host.com:9987

 

 

could this be the Problem?

 

 

 

Edited April 27 by eLpresidente

[iatemenchi]

On 4/17/2024 at 5:33 AM, ich777 said:

Please read this post, this is actually a reply from the Mega support team about the new release:

 

 

If you got further issues please try to contact the Mega support team, just for information purposes the container uses the latest version: megasync_5.2.0-2.1_amd64.deb

I messed with some stuff and figured out that if I don't have the shares use the cache drive, everything is fine. Makes sense since the error kept talking about a sync error with the drive.

 

I also heard back from Mega support. I'm not sure they bothered to look at the logs I sent, but they mentioned the app is out of date. I verified by looking at the current versions.

[AwesomeAustn]

On 4/15/2024 at 8:52 AM, SudoPacman said:

Heard back from support.

I suspect it might be due to differing versions of MEGA between the originator and the importer. I guess I'll see how it goes over the next few days... if it persists I'll try grabbing the debug logs.

Response here:
 

 

I am having the same issue and logging out and reinstalling it did nothing. Is there anyway to go back a version? I only see the latest tag. That, or hopefully the new version on the post above this one will fix it.

[ich777]

8 hours ago, eLpresidente said:

i just created a Cname ts3.host.com

The main question is why?

You usually create to a TS3 server via your hostname and port like host.com PORT

 

8 hours ago, eLpresidente said:

Which port ist the right to forward to in NGINX and what do i have to put in ts3 ? 

You have to forward all ports since one is the query port, then there is one for the voice channel and one for data transfer.

 

8 hours ago, eLpresidente said:

And in Teamspeak3 i just put the ts3.host.com for Serverip ? 

also tried ts3.host.com:9987

As said above you have to forward all ports because this is not a web server that proxies http/https traffic...

Sure you could use streaming ports but the benefit from this is non existent and you are overcomplicating your setup.

 

It is fine to forward the ports in your Firewall like it is for a game server (game servers are also not using http/https).

 

You can read a bit more here where I write and rant about it a bit more (it's another application but it applies here too) :

 

 

[ich777]

5 hours ago, iatemenchi said:

I verified by looking at the current versions.

They just released a new version...

I already updated the container.

(Maybe the found an issue which is with the new version fixed)

 

5 hours ago, iatemenchi said:

I messed with some stuff and figured out that if I don't have the shares use the cache drive, everything is fine. Makes sense since the error kept talking about a sync error with the drive.

I assume that's when you mount a share to the container where the share is meant to not stay on the cache drive or is this the case for all shares on the cache (which would not make much sense).

[ich777]

4 hours ago, AwesomeAustn said:

That, or hopefully the new version on the post above this one will fix it.

Please update your container, already updated it.

[eLpresidente]

1 hour ago, ich777 said:

It is fine to forward the ports in your Firewall like it is for a game server (game servers are also not using http/https).

 

oh i see, that was my mistake, i don't know why i wantet it to proxy via NGINX <-- 

 

Just Openend the Ports in my Fritz.Box and created a CNAME without Proxy in cloudflare. now it works. 

 

Thanks for your help. 

[ich777]

4 minutes ago, eLpresidente said:

Just Openend the Ports in my Fritz.Box and created a CNAME without Proxy in cloudflare. now it works. 

Usually such containers are safe to expose to the Internet and forward the ports directly and as said above, there is not much benefit to route it through a reverse proxy (which is in general possible) since you are overcomplicating and maybe also introducing other weird behaviour or other issues.