ich777

Please continue in this thread and see my comment there:

Have you yet tried the official Plex container? I‘ve also gone throught the other topic but without any further information I can‘t help, at least the Diagnostics would be helpful. From the OP from the other thread, is this the web client from Plex, it is a known issue if you set a transcoding profile there that it won‘t work. What is the source material HEVC, AVC, HDR,…?

Are you really sure that this patch solves your issue? When does the machine crash? This patch, at least it seems to me, is related to another issue where the guest driver causes a crash even when nothing is running in the VM. The most crashes happen because you exceed the assigned vGPU memory and this patch won't solve that. BTW you could also use a older driver version in the VM to avoid this issue that you've linked.

@zoggy as you've said: This is something we should discuss on the support thread from the Nvidia Driver plugin because the plugin is not part from the OS nor is the Plugin-Update-Helper. This is caused because the driver version from the legacy driver changes from time to time and the Plugin-Update-Helper simply can't find the 470.xx revision driver which you have selected and it falls back to the latest version. For the second recommendation about simply not downloading it, it would be the same outcome and simply not possible since even if the Plugin-Updated-Helper doesn't download the driver, the new Unraid version or better speaking the new Kernel version needs a updated driver which is compatible with the Kernel and as said above, even if the Plugin-Update-Helper doesn't download the driver for the new Kernel version the Plugin itself would do that on boot and since it can't find the same static driver version that you've set it to it falls back to the latest version. That's because of the new Kernel version and the Nvidia legacy driver changes from time to time that I've explained above. The new driver version for Unraid 6.12.0 is: 470.182.03 instead of 470.141.03 for Unraid 6.11.5 Hope that explains everything but I also have to say I won't change that behaviour since I don't know how long Nvidia is going to support the legacy driver and how long it will work with the container runtime <- which is required for the use from Nvidia cards in Docker containers. As a little side note, if the container runtime is dropping support for the legacy driver I will also stop compiling these legacy drivers. May I ask for what do you use the GT710? The GT710 is a pretty bad card for transcoding anyways because it even doesn't support h265 (HEVC) and in comparison it is really power hungry compared to something more recent like a Nvidia T400 which you can get for pretty cheap (often times for below 100,-), it only draws about 1-4 Watt in idle and it is based on the Turing architecture. Hope that helps and answers all your questions.

Can you describe your issues? I think otherwise nobody here can help you correctly. GVT-g always heavily depends on the implementation in the Firmware from the Motherboard manufacturer.

What game are we talking about? This shouldn‘t prevent the dedicated server from starting, I usually don‘t include SDL because it‘s not needed for the dedicated server to work. The full log output would be also helpful.

If it is still boot looping then the container isn‘t properly updated. Sorry, I‘m currently on the go and can‘t tell drom the top of my head which one you meed to change. You could also do this with a startup game parameter. EDIT: Look here at the command line options: https://satisfactory.fandom.com/wiki/Dedicated_servers

Please update the container itself. I pushed a update to the container yesterday to make it compatible with update 8 I tested both the stable and experimental branch and both are working fine on my server. Did you also change them in the config?

Please read this post:

@mgutt bitte update mal das Paket aus deinem Repository: Click oder benutzt un-get um powertop zu installieren. Ich habe schon lange ein upgedates package in meinem Repository.

Please don't do that, I will revert the change from the base image. Thank you for the report.

Wait, so is there a modification to the container needed?

Thanks for the report, please force a update from the container itself (Docker page -> enable Advanced View -> click Force Update at the container -> disable Advanced View).

I think this is not easily possible, I would rather ask the developers from the mods when they will be updated.

This is now fixed, please update the container itself.

Thank you for the report I will fix that ASAP and report back. Should be done in an hour or two.

You need to enable validation. This is working perfectly fine since I‘ve tried it yesterday and another user here was also able to install it fine on hos system and I even updated the container yesterday to be compatible with the new engine: Pleaee read the description from the variable GAME ID, you just have to append: -beta experimental to the GAME ID and it will work just fine. As said above, you have to enable validation for this so that it pulls the update, if validation is not working, like it is the case for some users, try it on a fresh installed container.

Yes, just like I've wrote it above, this was no mistake or laziness...

Of course it's fixed because I've tried it, both the old version and experimental... Yes, the binary changed.

Thanks for the heads up. Fixed, please force a update from the container itself, both stable and experimental will now launch correctly.

Please use exclude like that: domains/* steam/* backup/* Did you made sure that you've enabled console mode like mentioned here:

Please post scrennshots what you did exactly for more help.

Just put: -batchmode -nographics in front of it here: This is basically the same and you don't have to do anything with the command line.

I will try that tomorrow and report back. Hope that's good enough.

What should I fix? This is a thing the game devs should fix in the first place. You have to understand that this would be the same as if the devs would push an update without validation enabled the file would be always overwritten. So to speak they should add headless= to the boot.cfg by default.