ich777

What do you mean exactly with GUI the VM?

What have you done so far? What are you typing in, with no context I can't help...

Yes, but not that new... You can at least try if you can get it to work.

No, but don't do a manual modprobe. Please note that Alder Lake isn't fully supported yet and I get many messages about that. That has nothing to do with the plugin but rather with the Kernel. Please reboot once more and the send over the Diagnostocs again. Please also keep in mind that many users reported a complete crash from the entire server when using an Alder Lake iGPU for HW transcoding.

I will look into this ASAP, please give me a few days... hope that's not to late for you...

Please share your Diagnostics again. Something seems really wrong on your system if such a thing happens...

Completely or for only a short time? Only because it crashed you had to reinstall Windows?

What have you done that it worked before? What do you mean exactly with that? That is just normal at time when the driver is loaded during boot.

Thank you for the update @dlandon! Really much appreciated!

@b3rs3rk can you help here? Something seems wrong...

That was already discussed and could cause really much confusion for new users, also it would be a bit difficult on how to implement that in the CA App since you need more different templates, the paths can be different, the ports,... I think also another reason was how many Cluster templates that you put on the CA App, 1, 2, 3,..., 11,... Creating a cluster is nothing for beginners and if a beginner uses the wrong template from the CA App or thinks that something like this is easy to setup I also have to give support for. That's why it was decided to recommend the post and not put it on the CA App. Yes I know, but keep in mind Modding is not a Mod. Modding means for me any modification to the server what is done by the user, that also includes that more servers/containers act as one server.

I don't know what you mean with that? My template is for single server hosting and nothing more... Modding is always up to the users itself but I try to help where I can but I've never set up a ARK cluster... Exactly that is what his GitHub is for and I think the post that is recommended tells you that IIRC.

Can you tell me what this is? I found not much about this... Where can I get LEGO Universe, nothing that I found on Steam... I also saw that they have a massive ammount of issues open about many events not triggering, are you really sure that it is worth the time to look into this?

I would strongly recommend to not convert everything into AV1 just yet and wait if there comes something up new that maybe is better and/or establishes itself more than AV1. Also as said once, barely any device supports AV1 in hardware and that means you have to transcode every single stream to the devices again on your server and that would be a real pain I think...

From the error it seems that something with the transcoding is failing. May be try to stop the container, delete the whole cache folder for Jellyfin and start the container again. Don't know if something is wrong with the permissions or anything else. Can you try to force a transcode from a file in your local network (click on the gear icon when playing a file and change the quality to something lower than the source)?

Exactly because it's loaded through the plugin if you've enabled Autostart. modprobe corefreqk will do the job just fine but keep in mind that is what the plugin does if you've enabled Autostart.

Both of them can be blacklisted. Thank you @Squid!

Will look into this ASAP, maybe I've messed up the configuration... This is usually the intel_pstate driver where only powersave and performance are availalbe if you want to use a ondemand governor you have to manually edit your syslinux.conf and add this to get the acpi-cpufreq driver by adding this: intel_pstate=disable and reboot. Or you can use the intel_pstate driver in passive mode by adding this to your syslinux.conf: intel_pstate=passive and reboot so that it acts more like the acpi-cpufreq driver and get also the ondemand governor. But only add one argument both won't work, hope that makes sense to you. Here is some documentation on how this is working: intel_pstate & ARCH CPU frequency scaling WIKI

Das kann nicht funktionieren da die Karte jedes mal an VFIO gebunden war und damit für das Host System mehr oder weniger nicht zu sehen ist weil sie eben für die Nutzung in VMs reserviert ist und somit "sieht" das Plugin die Karte nicht. Genau deswegen wird sie wieder in dem Plugin angezeigt. Lass die Karte bitte an VFIO gebunden wenn du sie durchreichn willst und schmeiß das Plugin runter. Diese Einstellung würde ich wählen aber mit aktiviertem CSM und auch wirklich in den Legacy mode booten, mit dem erhältst du Erfahrungsgemäß die besten Ergebnisse, auch wenn die karte an VFIO gebunden ist. Dann funktioniert sie in den VMs aber auch nicht mehr. Genau für diesen Zweck ist VFIO da, deswegen bekommst du auch keine Ausgabe vom Mellanox Plugin und das die Karte eben nicht gefunden wurde. Ich empfehle dir mal klein anzufangen, eins nach dem anderen zu machen und die Grafikkarte mal links liegen zu lassen. Für die Mellanox CSM aktivieren und in Legacy booten -> PCIe ACS Override -> Downstream -> reboot. Die Mellanox Karte an die VM durchreichen, wie machst du das eigentlich? Hast mal einen screenshot? Dann probieren ob die startet.

Wo hast du denn den Monitor angeschlossen? Wenn du ihn auf der iGPU angeschlossen hast dann stell bitte sicher das du das Primäre Output der Grafikkarte auf die iGPU im BIOS eingestellt hast und installier sicherheitshalber das Intel-GPU-TOP Plugin. Wenn du noch zusätzlich das Nvidia Driver Plugin installiert hast öffne bitte ein Terminal von Unraid selbst und gib das hier ein: sed -i '/disable_xconfig=/c\disable_xconfig=true' /boot/config/plugins/nvidia-driver/settings.cfg und Starte deinen Server neu (das bitte nicht machen wenn du den monitor an der Nvidia Karte angeschlossen hast um die GUI von Unraid zu sehen).

Already answered on Discord, you can try to compile it but I really can't tell if it is working. Also keep in mind that I forgot to remove the container from DockerHub and I will be removing it in the next few days from DockerHub.

What tuners are you using or how did you integrate it? This seems like a configuration issue since on my server it works just fine with TVHeadend and 4 Tuners.

Please use the support thread for GVT-g by going to the Plugin tab in Unraid and click Support Thread at the GVT-g plugin.

What does the Vendor drop down display?

Hast du eine oder 2 GPUs verbaut?