Hello, is it possible to limit the maximum use of RAM in the Satisfactory Docker? Unfortunately I can't find anything in the settings.

Thanks

13 hours ago, alone.at.home said:

is it possible to limit the maximum use of RAM in the Satisfactory Docker?

May I ask why you want to do that?

This is a really bad idea, if you do that the server will ultimately crash if it wants to assign more RAM than you allow it to.

Hi, I want to host the Satisfactory Docker for some friends but the docker don't start proberly. First I thought it install some files so I waited one hole night that it install the hole files. The foldes size stucked at 6,67 GB. I've delete the docker (+image) and the created "stemcmd" and "satisfactory" folder many times. The Docker starts and do things but stands still every time at this line:

Quote

[2023.08.10-11.55.35:895][  0]PacketHandlerLog: Loaded PacketHandler component: DTLSHandlerComponent ()

Is there a setting to change or a known bug? I've attached the complete log.

Satisfactory Docker Log.txt

27 minutes ago, Dextoo said:

The Docker starts and do things but stands still every time at this line:

How much CPU and RAM does the container consume?

 

27 minutes ago, Dextoo said:

[2023.08.10-11.55.35:895][  0]PacketHandlerLog: Loaded PacketHandler component: DTLSHandlerComponent ()

IIRC this line shows after the container is fully started and waiting for connections.

 

EDIT: If you read one post above yours someone posted another Satisfactory related question so I assume the container is working, but I haven't tried it yet.

8 minutes ago, ich777 said:

How much CPU and RAM does the container consume?

 

10 minutes ago, ich777 said:

IIRC this line shows after the container is fully started and waiting for connections.

That would make me very uncomfortable 😅 I don't own the game so I've tested the connection just with telnet. I'm going to ask a friend to try it out.

1 minute ago, Dextoo said:

This seems correct to me IIRC.

 

1 minute ago, Dextoo said:

I don't own the game so I've tested the connection just with telnet.

I don't think that you will get a answer back from most game servers over telnet...

Most gameservers simply drop the connection if they get a request from something to them unknown.

Hello @ich777!

 

There seems to be an issue with Icarus running on linux, where it runs out of memory blocks, causing the docker container to crash (OOM) and restart. The community seems to have found a fix however, but it's beyond my capabilities to apply it myself; so I thought I'd forward the issue.

 

Icarus "Linux and Docker Support" discord thread discussing the issue.

 

 

Fixed community docker hub page has a ReadMe at the bottom explaining the issue and the fix.

 

 

Is this a fix that can be implemented on your end through the Docker Hub? Or is it something we have to implement on our end? If so, I'll finally have to sit down and learn how to edit docker files on Unraid.  If I understand correctly, I can use the docker console to apply the fix, but it won't "stick" through updates unless I edit the dockerfile itself?

 

Regardless, thanks for taking the time and effort to put out all these containers and make life simple for the rest of us! Much appreciated.

 

TL;DR: Those who have issues with the Icarus Docker crashing with a "Ran out of memory" error, it's a known issue with the dedicated server running out of memory blocks. Community seems to have found a fix, which you can find in the links above.

Edited August 10, 20232 yr by SinoBreizh
Rephrased for clarity

11 minutes ago, SinoBreizh said:

Is this a fix that can be implemented on your end through the Docker Hub? Or is it something we have to implement on our end? If so, I'll finally have to sit down and learn how to edit docker files on Unraid.  If I understand correctly, I can use the docker console to apply the fix, but it won't "stick" through updates unless I edit the dockerfile itself?

This is something that needs to be done on the host and nothing that needs to be done in the container but that or at least how it's described there is not suitable for Unraid.

 

I've never heard of that issue to date or better speaking no one reported that to me about my container, did you experienced this crash already?

Is this discussion on Discord about my container or is it the "semi" official?

 

I'm not a big fan of setting this value to something higher because this can also cause issues on the system itself.

 

TBH I have played only a few hours on the server and it was rock solid.

27 minutes ago, ich777 said:

I've never heard of that issue to date or better speaking no one reported that to me about my container, did you experienced this crash already?

I've had a week and a half of smooth sailing, good performance and no bugs; but just yesterday and today, the friends I host for reported massive slowdowns, physics not working, followed by being kicked out of the server. I've rebooted yesterday, thinking it was a one-off, but it occured again today after about 14 hours of uptime. So I've experienced the issue twice at this point.

 

27 minutes ago, ich777 said:

Is this discussion on Discord about my container or is it the "semi" official?

I saw no mention of your container by searching the thread. I don't know of a "semi official" docker, but the one with the fix in the ReadMe is a SteamCMD variant like yours. The discussion itself had a variety of users: desktop linux, hosting providers, some Unraid users.

 

27 minutes ago, ich777 said:

I'm not a big fan of setting this value to something higher because this can also cause issues on the system itself.

Interesting. Following the discussion, I was under the impression that no one had really run a stress test over time yet to determine if it was a flawless fix. But the people at Shockbyte are apparently testing this specific fix for their customers.

 

27 minutes ago, ich777 said:

TBH I have played only a few hours on the server and it was rock solid.

Yeah, I was surprised too; it's strange that it ran fine for over a week, then it all came crashing down in the past few days. Maybe it has something to do with the world size growing as the clients keep playing.

Edited August 10, 20232 yr by SinoBreizh
Clarification

48 minutes ago, SinoBreizh said:

I've had a week and a half of smooth sailing, good performance and no bugs; but just yesterday and today, the friends I host for reported massive slowdowns, physics not working, followed by being kicked out of the server.

But the issue talks about the server crashing or am I wrong, this seems a bit different or am I wrong?

 

49 minutes ago, SinoBreizh said:

I don't know of a "semi official" docker, but the one with the fix in the ReadMe is a SteamCMD variant like yours.

I assume that my container is most certainly working differently then the "semi-official" container since I do many things differently than most other container maintainers.

 

50 minutes ago, SinoBreizh said:

But the people at Shockbyte are apparently testing this specific fix for their customers.

For now this is a workaround not a fix in my opinion, this needs to be fixed from the developers in the engine itself.
I got so many requests about this game to create a container but in the early stages is was plagued by bugs and was really unstable, even for a proper connection you had to do a workaround.

I'm not really willing to accept another workaround for a thing that needs to be fixed. It should not be necessary to increase that limit for a game server.

 

Again, you have to do that on the host if you are really want to change that limit, execute this form a Unraid terminal (but please keep in mind that it will be reset on a restart from Unraid, because Unraid works differently):

echo "262144" > /proc/sys/vm/max_map_count

to check if it was applied successfully issue this command and it should give you 262144 back:

cat /proc/sys/vm/max_map_count

 

27 minutes ago, ich777 said:

But the issue talks about the server crashing or am I wrong, this seems a bit different or am I wrong?

It is the same issue; the slowdown and physics breaking is just an early symptom of the server running out of memory blocks. It crashes when it runs completely out, and kicks everyone out before restarting to recover. I get the exact same errors Ran out of memory allocating 0 bytes with alignment in the exact same circumstances as mentionned in the discussion and ReadMe, with more than 30Gb of spare physical memory available. 

 

36 minutes ago, ich777 said:

For now this is a workaround not a fix in my opinion, this needs to be fixed from the developers in the engine itself.
I got so many requests about this game to create a container but in the early stages is was plagued by bugs and was really unstable, even for a proper connection you had to do a workaround.

I'm not really willing to accept another workaround for a thing that needs to be fixed. It should not be necessary to increase that limit for a game server.

No pressure, I understand; I can imagine it'd be a nightmare as a maintainer to keep track of, and troubleshoot, every baind aid fix -  which may or may not work, when the real solution lies upstream with the game itself. From what I understand, even the Windows dedicated server is still in "beta".

 

40 minutes ago, ich777 said:

Again, you have to do that on the host if you are really want to change that limit, execute this form a Unraid terminal (but please keep in mind that it will be reset on a restart from Unraid, because Unraid works differently):

Commands worked as you explained, thanks! I'll do a quick test run that way, to see if this "fix" even works in the long run. Will it be reset if the docker container is restarted (not Unraid as you mentioned, but the docker container itself)? I back up the save files with the appdata backup plugin every night, and it restarts the game containers in the process.

 

Either way, in the long run I think I'll just spin up a Windows VM if the issue keeps coming up, until the Icarus devs release a native Linux server client (there seems to be some demand for it), or adress the issue specifically. I'll take your word that this fix is far from ideal anyways, I'd rather not play around with things I don't fully understand!

 

If you have the time and opportunity to do so though, I'd be curious to see if you can replicate the issue on your end. Intuitively, I'd imagine running out of memory blocks becomes more of a thing when the world grows larger (as players explore) and more populated (with structures, equipment and creatures). So it could take some time before the issue comes up.

9 hours ago, SinoBreizh said:

No pressure, I understand; I can imagine it'd be a nightmare as a maintainer to keep track of, and troubleshoot, every baind aid fix

This is nothing that I can integrate into my container since this needs to be applied on the host.

 

9 hours ago, SinoBreizh said:

From what I understand, even the Windows dedicated server is still in "beta".

My container uses the Windows client because there is no Linux server available.

 

9 hours ago, SinoBreizh said:

Will it be reset if the docker container is restarted (not Unraid as you mentioned, but the docker container itself)?

Only when the Host is restarted, so to speak Unraid.

 

9 hours ago, SinoBreizh said:

Either way, in the long run I think I'll just spin up a Windows VM if the issue keeps coming up, until the Icarus devs release a native Linux server client (there seems to be some demand for it)

But that doesn't mean that will and I don't think that they will do that since they struggle even with the Windows version that much, look for example at The Forest or Sons of the Forest they have and probably will never have a native Linux server.

 

9 hours ago, SinoBreizh said:

If you have the time and opportunity to do so though, I'd be curious to see if you can replicate the issue on your end.

I don't think I have time for that, in the little spare time that I have left over currently.

3 hours ago, ich777 said:

My container uses the Windows client because there is no Linux server available.

I'm aware; I meant that in the sense that while Linux hosting through wine has no official support from the devs, even the Windows instance of the dedicated server is still technically in beta. And as you said they seemed to have struggled to get this far to begin with.

 

3 hours ago, ich777 said:

Only when the Host is restarted, so to speak Unraid.

Got it, thanks!

 

3 hours ago, ich777 said:

But that doesn't mean that will and I don't think that they will do that since they struggle even with the Windows version that much, look for example at The Forest or Sons of the Forest they have and probably will never have a native Linux server.

You're probably right, it's wishful thinking. I've been thinking about migrating all my game servers to a Windows VM with a proper server manager GUI like WindowsGSM or similar. But I can't deny the resource efficiency and convenience of docker containers with editable files shared through SMB.

 

3 hours ago, ich777 said:

I don't think I have time for that, in the little spare time that I have left over currently.

No worries, thank you for taking the time to answer my questions regardless! I now have a better understanding of the problem and what can and can't be done about it. Makes the whole thing a whole lot less aggravating once I manage to wrap my head around it.

Edited August 11, 20232 yr by SinoBreizh

Anyone know how in the FiveM container I can manually update TxAdmin?

 

It's probably a bit hard to keep up with all of the updates but I'd love to get it updated to the latest version.

 

Actually, how would I go about updating the Game version as well as the txAdmin.

Edited August 11, 20232 yr by Milhouz

2 hours ago, Milhouz said:

Actually, how would I go about updating the Game version as well as the txAdmin.

Simply restart the container and make sure that manual updates are disabled in the template.

Howdy! I am trying to setup a Quake III Arena server, and this is what I am running into. My install only has the one PAK file.

 

Any tips?

 

Thanks!!

 

48 minutes ago, sp00nix said:

Howdy! I am trying to setup a Quake III Arena server, and this is what I am running into. My install only has the one PAK file.

Did you restart the container after placing the file in the correct directory?

Does the path in the template also point to the physical location on your machine?

Yeah I restarted it. Initially I tried the pk3 files from the CD, the from the install (it's on a win98 box) and restarted each time.

 

The path looked like it didn't go deep enough so I pointed it right to the folder. Still no luck.

Looking again I don't think that was the right move. I'll have to poke around more when I get back to a desktop.

1 hour ago, sp00nix said:

The path looked like it didn't go deep enough so I pointed it right to the folder. Still no luck.

But the path in your screenshot from above and says something like "virtual_machines" are you sure that the path is correct?

 

Also this was a really bad idea what you've did there, you should only changthe path to something like /mnt/virtual_machines/appdata/quake3

 

What you've done here is basically you've said the container that it should all files that belong in the folder quake3 now in a subfolder and you now even have a deeper path.

22 minutes ago, ich777 said:

But the path in your screenshot from above and says something like "virtual_machines" are you sure that the path is correct?

 

That is just the name assigned to the SSD I use for VMs and Dockers.

 

I'll uninstall and reinstall and try again in case I boogered it up.

Alright, got that sorted out. The default install path was in /tmp and I think that was throwing myself and the system off.

 

Anywho looks like i have the files in the right spot, more files have populated around it, but now the console and the logs only show up for a split second, and I'm not sure it is even running.

 

 

 

EDIT:

I can see the logs after stopping the docker, here is what it has:

 

text  error  warn  system  array  login  

0 files in pk3 files
Sys_Error: Couldn't load default.cfg
Q3 1.32c linux-i386 May  8 2006
----- FS_Startup -----
Current search path:
/quake3/.q3a/osp
/quake3/osp
/quake3/q3ded/osp
/quake3/.q3a/baseq3
/quake3/baseq3
/quake3/q3ded/baseq3

----------------------
0 files in pk3 files

Running in restricted demo mode.

----- FS_Startup -----
Current search path:
/quake3/.q3a/demota
/quake3/demota
/quake3/q3ded/demota

----------------------
0 files in pk3 files
Sys_Error: Couldn't load default.cfg
Q3 1.32c linux-i386 May  8 2006
----- FS_Startup -----
Current search path:
/quake3/.q3a/osp
/quake3/osp
/quake3/q3ded/osp
/quake3/.q3a/baseq3
/quake3/baseq3
/quake3/q3ded/baseq3

----------------------
0 files in pk3 files

Running in restricted demo mode.

----- FS_Startup -----
Current search path:
/quake3/.q3a/demota
/quake3/demota
/quake3/q3ded/demota

----------------------
0 files in pk3 files
Sys_Error: Couldn't load default.cfg
---Ensuring UID: 99 matches user---
usermod: no changes
---Ensuring GID: 100 matches user---
usermod: no changes
---Setting umask to 000---
---Checking for optional scripts---
---No optional script found, continuing---
---Checking if Quake3 is installed---
---Quake 3 found, continuing...---
---Taking ownership of data...---
---Starting...---
---Checking if .pk3 files are present---
---pak0.pk3 found, continuing---
---Prepare Server---
---Checking if 'server.cfg' is present---
---'server.cfg' found, continuing---
---Checking if 'maprotation.cfg' is present---
---'maprotation.cfg' found, continuing---
---Server ready---
---Starting server---
---Ensuring UID: 99 matches user---
usermod: no changes
---Ensuring GID: 100 matches user---
usermod: no changes
---Setting umask to 000---
---Checking for optional scripts---
---No optional script found, continuing---
---Checking if Quake3 is installed---
---Quake 3 found, continuing...---
---Taking ownership of data...---
---Starting...---
---Checking if .pk3 files are present---
---pak0.pk3 found, continuing---
---Prepare Server---
---Checking if 'server.cfg' is present---
---'server.cfg' found, continuing---
---Checking if 'maprotation.cfg' is present---
---'maprotation.cfg' found, continuing---
---Server ready---
---Starting server---

Edited August 11, 20232 yr by sp00nix

15 hours ago, sp00nix said:

I can see the logs after stopping the docker, here is what it has:

Something is wrong with you .pak file, there are you get these messages over and over:

15 hours ago, sp00nix said:

0 files in pk3 files

which basically means that something is wrong.

 

To what is your appdata share set in terms of Use Cache or what primary storage is set?

 

I can not reproduce this, I've attached my log since I've now tried it, you can also see that bots are already playing on the server...

q3.log

Yeah, I'm not sure. I have tried 3 PK3 files now. I found some on a github.

 

At the end of my log I noticed this:

 

---pak0.pk3 found, continuing---
---Prepare Server---
---Checking if 'server.cfg' is present---
---'server.cfg' found, continuing---
---Checking if 'maprotation.cfg' is present---
---'maprotation.cfg' found, continuing---
---Server ready---
---Starting server---

 

So it eventually found it? The thing is if i have any PK3s in there it seems to not do anything, the webgui shows that the docker is using 0% CPU and 0bytes of RAM, if i remove the PK3 files it at least runs to some degree. Is there a specific version of the PK3 files i would need, or is it pretty universal?

 

 

Appdata is on an SSD in a pool for the sole purpose of running VMs and dockers, no alternative storage is set.

 

 

 

These are the PK3 files i found on that github, decided to dump them all in there. Same result.

 

I don't know what info you might find useful. I'm still pretty green to unRAID and my understanding of the Linux file structure is pretty weak.

 

Thank you ich777 for great docker. Can i ask how to add more valheim servers for one steamcmd? Can it be done? Thanks for help.