ich777

Please see this post:

Sure thing, but you have to use the keys that live in the folder /luckybackup/.ssh not the ones in /root/.ssh

Schreib das mal in deinen Bug thread und markier @bonienl. Mit 6.10 wurde auf nchan umgestellt.

Welche skripte genau?

I think @alturismo & @Mr.Will will figure that out...

I think there are multiple ways of setting the governor for currently assigned cores to VMs which is either on or off but the proper way should be to read which cores are assigned to the VM from libvirt and change the states only for the assigned cores since those two are trying to accomplish this from what I read here.

Because this is not the Log and this is therefore not the Token. Please remove the container again and also the folder that got created for teamspeak that lives in your appdata directory, pull a fresh copy from the CA App and then, after the container started click on the icon and select Logs:

Yes. Maybe it would be a good thing to use the qemu hooks itself and check in the hooks if a Nvidia GPU is bound to the VM itself and if it is even needed to kill nvidia-persistenced or start it after the VM is ended.

Reden wir von diesen kleinen ausschlögen oder von was anderem? Habt ihr schon mal versucht im Safe Mode zu booten und dort ein paar stunden zu bleiben ob es dort das selbe ist?

Jup. @schoppehermann ich würde vorshlagen das du das den post von @cz13 als Lösung markierst. 😉

@alturismo has a solution for this I think.

@alturismo posted such a script already on the forums, somewhere… 😅 You can also call nvidia-persistenced once and kill it after a few seconds this should pull down your power draw from the GPU too, even if nvidia-persistenced isn‘t running at all. Of course this is also a workaround. It would be of course a welcome addition but the description has to be cristal clear what his plugin is doing (VFIO, that is not for use in Docker containers and so on).

Is the GPU recognized correctly in the Plugin? If the GPU is correctly recognized and listed on the Plugin page then please post on the appropriate support thread from the container because the driver is indeed working and this issue is related to the container since this is also the log output from the container.

Why? As said above, don‘t do this, this handled by the GAME_PORT variable.

No, steamcmd is only used when the container is fresh started to downlaod the game and check for updates, after that it doesn't run anymore, or better speaking it doesn't run anymore when the application (CS:GO) is started. I also have reused the SteamCMD directory from another container and was already in place. Can you check if other services are running on those ports and are reachable from outside? I have a few reports that some ISP block certain ports or port ranges.

I tried it now with 27100, 28015, 27015 and also 27030 (default on my server) and everything is working as it should...

I‘ve sent you a PM.

Was anybody lucky enough to get their hands on a ARC GPU? If yes, please contact me... why, because of this: Yes and no, if the driver also uses /dev/dri, like all currently Intel based iGPUs, it should work OOB or better speaking at least on the Host and how it behaves after that, in terms of usage in Docker containers and so on is to be seen and needs to be tested. Also keep in mind that maybe the Intel Media Driver needs to be updated inside the containers but these are all things that need to be seen if someone can get there hands on a ARC GPU.

@bjmi in case you haven't seen it already: (ZSTD compression is now available with 6.11.1)

Have you set a different port? Please click on the show more section, the port is handled in there. Maybe also post a screenshot how you configured it now.

You don‘t have to reconfigure it, just remove it and install it again.

You can use that for the VM but please not for transcoding, the GT1030 isn‘t capable of transcoding h265 (HEVC)

No issue there, you even can it with a x1 slot from my testing. Exactly, but it also doesn't matter which card you use for what since this card is not that powerful but preferably you would put the card for the VM in the x16 slot because I assume the other PCIe slot (also x16 in terms of width) which is eclectically wired as x4 is running over the chipset. This always depends on your workload, there is always something better... Don't know if you are happy with the T400 why buying something that's more expensive? Of course you can buy a RTX 3050, 3060 or even higher, maybe you want to try the new ARC Intel GPUs... That's all up to you.

Exactly, but you should quote the other user. He meant if Nerdtools is compatible with un-get and vice versa, nothing to do with updates.

Maybe see if the slot has a open back x4 is well enough for transcoding, even x1 is enough... btw. you could also use a riser