@ich777 Swapped the GPU slot from the first to the second GPU slot on the motherboard, started the system & now normally the driver will crash in less than an hour but now seems to be working fine. I even tried transcoding more than 8 streams of HEVC 4K on Plex without any failures. I restarted Plex docker multiple times trying to initiate any failures, but it seems that all is working great.

As I expected, it must be the motherboard dying.

 

Conclusion: nothing wrong with nvidia driver plugin yay.

@ich777 I installed a 1660 into my unraid server to utilize with your docker image. I'm having an issue where the GPU registers in the system (with the nvidia plugin), but when I try to utilize the GPU for transcoding, plex just throws a black screen and never actually transcodes. Hardware transcoding does work with Jellyfin which is what makes me think it's an issue with the docker image. Any help you could provide would be greatly appreciated. I have disabled HDR tone mapping as a troubleshooting step as I saw others were having trouble with it in the plex forum. It did not fix the issue.

PacMan_UpdateContainer_Redacted.pdf

pacman-diagnostics-20220905-0758.zip

Edited September 5, 20224 yr by tusculumgolfer
Add diagnostic file

22 minutes ago, tusculumgolfer said:

plex just throws a black screen and never actually transcodes. Hardware transcoding does work with Jellyfin

I think then it would be best that you post this issue in the according support thread for the Docker container since this is not an issue with the plugin itself.

 

Please note that I always recommend to use the official Plex container, I got a few reports that some of the 3rd party containers won't work properly sometimes with HW transcoding.

I keep having an issue where the docker containers that utilize my gpu die out and then i get a "Bad Parameter" when i try to restart them. 

Docker is giving me this

Error response from daemon: failed to create shim: OCI runtime create failed: container_linux.go:380: starting container process caused: process_linux.go:545: container init caused: Running hook #1:: error running hook: exit status 1, stdout: , stderr: nvidia-container-cli: initialization error: open failed: /proc/sys/kernel/overflowuid: permission denied: unknown

 

I appear to have access to overflowid

cat /proc/sys/kernel/overflowuid
65534

 

I have tried all the different driver versions with no luck.

I am running Version: 6.10.3

9 hours ago, BigWebstas said:

Docker is giving me this

Can you send me the Diagnostics after this happens?

 

Do you have nvidia-persostenced enabled or any power saving script enabled? If yes which one?

For testing purposes please disable it.

Hello,

after following your instructions, unfortunately I can't get Emby to run. I get the following message:

 

Quote

docker: Error response from daemon: failed to create shim: OCI runtime create failed: container_linux.go:380: starting container process caused: exec: "/init": stat /init: no such file or directory: unknown.

 

 

 

A reinstallation of EMBY Docker solved the problem. Even though I don't know what the problem was.

 

2 hours ago, Gorosch said:

A reinstallation of EMBY Docker solved the problem. Even though I don't know what the problem was.

If the issue occurs the next time, please post your Diagnostics.

ive been having an issue for past month or so. I found that Plex wont run when I have a GPU listed, then I looked at the nvidia plugin and it shows that no GPU is there. if I look into system devices, it is there. I tried uninstalling the plugin and reinstalling it. This all worked before, the only difference is the driver version/plugin version.

 

its an Nvidia 1660 Super.

 

backup-diagnostics-20220907-2317.zip

1 hour ago, TheSkaz said:

This all worked before, the only difference is the driver version/plugin version.

It seems like the driver isn't installed fully. Did you wait until the DONE button was displayed on plugin installation?

What is the output from:

nvidia-smi

from a Unraid terminal?

 

Can you please try to click the Update & Download button again and see what happens?

If it says that something is wrong with the driver please wait for it to fully download and restart your server.

 

Otherwise I would recommend that you:

1. Uninstall the plugin
2. Reboot
3. Pull a fresh copy from the CA App and wait until the done button is displayed
4. Reboot

After that the driver should be installed properly and you see your GPU listed on the Plugin page.

12 hours ago, ich777 said:

It seems like the driver isn't installed fully. Did you wait until the DONE button was displayed on plugin installation?

What is the output from:

nvidia-smi

from a Unraid terminal?

 

Can you please try to click the Update & Download button again and see what happens?

If it says that something is wrong with the driver please wait for it to fully download and restart your server.

 

Otherwise I would recommend that you:

1. Uninstall the plugin
2. Reboot
3. Pull a fresh copy from the CA App and wait until the done button is displayed
4. Reboot

After that the driver should be installed properly and you see your GPU listed on the Plugin page.

it said that the checksum was bad. it redownloaded and I rebooted. the GPU is now showing. thank you!

On 9/8/2022 at 2:45 AM, ich777 said:

It seems like the driver isn't installed fully. Did you wait until the DONE button was displayed on plugin installation?

What is the output from:

nvidia-smi

from a Unraid terminal?

 

Can you please try to click the Update & Download button again and see what happens?

If it says that something is wrong with the driver please wait for it to fully download and restart your server.

 

Otherwise I would recommend that you:

1. Uninstall the plugin
2. Reboot
3. Pull a fresh copy from the CA App and wait until the done button is displayed
4. Reboot

After that the driver should be installed properly and you see your GPU listed on the Plugin page.

I completed the uninstall and reinstall like you laid out here and I can see one of the two GPUs in the Nvidia Driver plugin again (the other GPU is in pass-through to a VM so not sure if it should show up here). I have the server diag from before the change as well if your interested. Let me know if you want the one after and I can upload it too.

tworivers-diagnostics-20220910-1756.zip

Edited September 11, 20224 yr by Herbiewalker

5 hours ago, Herbiewalker said:

the other GPU is in pass-through to a VM so not sure if it should show up here

it shouldnt, while a disk is vfio bind and/or active in a VM it shouldnt show up anymore, so you should be fine then

6 hours ago, Herbiewalker said:

the other GPU is in pass-through to a VM so not sure if it should show up here

Like @alturismo already mentioned because you've bound the card to VFIO it isn't visible anymore for the host OS:

03:00.0 VGA compatible controller [0300]: NVIDIA Corporation GM206GL [Quadro M2000] [10de:1430] (rev a1)
	Subsystem: Dell GM206GL [Quadro M2000] [1028:1190]
	Kernel driver in use: vfio-pci
03:00.1 Audio device [0403]: NVIDIA Corporation GM206 High Definition Audio Controller [10de:0fba] (rev a1)
	Subsystem: Dell GM206 High Definition Audio Controller [1028:1190]
	Kernel driver in use: vfio-pci

 

Hi,

New to Unraid I have a Dell R720 and a Quadro P2000 I use with Jellyfin, I upgraded my OS to 6.10.3 and after that I could not get the NVidia Drivers to work it did not seem to see my card any more I then downgraded back to 6.9.2 with no luck if you start the NVidia Drivers download it will start and stay on the download screen and will never give me the done button. I have now upgraded to 6.11.0-rc5 and all my problems are gone I have my Quadro P2000 back and the Nvidia Drives installed 

21 minutes ago, HomeStudios said:

New to Unraid I have a Dell R720 and a Quadro P2000 I use with Jellyfin, I upgraded my OS to 6.10.3 and after that I could not get the NVidia Drivers to work it did not seem to see my card any more I then downgraded back to 6.9.2 with no luck if you start the NVidia Drivers download it will start and stay on the download screen and will never give me the done button. I have now upgraded to 6.11.0-rc5 and all my problems are gone I have my Quadro P2000 back and the Nvidia Drives installed 

Have you any Diagnostics from where you got the issues? I really can't help with that little information.

 

The issue with the Done button is usually caused because it can't properly communicate with GitHub (this is where the driver packages are located) and/or GitHub has issues with it's downloads again.

Over here sometimes (but this happens really occasionally) I get download speeds of about 100KB/s and as you can imagine for a driver package of about 200-300 MB this can take a long time.

 

I also would recommend the next time you got such an issue that you:

1. Remove the Plugin
2. Reboot
3. Grab a fresh copy from the CA App
4. Reboot

If you never get a Done button then I would recommend that you wait a few hours and remove the plugin in the meantime and then try it again.

28 minutes ago, ich777 said:

Have you any Diagnostics from where you got the issues? I really can't help with that little information.

 

The issue with the Done button is usually caused because it can't properly communicate with GitHub (this is where the driver packages are located) and/or GitHub has issues with it's downloads again.

Over here sometimes (but this happens really occasionally) I get download speeds of about 100KB/s and as you can imagine for a driver package of about 200-300 MB this can take a long time.

 

I also would recommend the next time you got such an issue that you:

1. Remove the Plugin
2. Reboot
3. Grab a fresh copy from the CA App
4. Reboot

If you never get a Done button then I would recommend that you wait a few hours and remove the plugin in the meantime and then try it again.

Hi sorry no Diagnostics next time I will add some if I have any problems. 
Removed rebooted and all that from what I can see it never got the .md5 file this was over a day and still never got to the done button, I have also done a manual download off all the files needed and added it into the correct location with me doing that I got to a point that I could start the Nvidia Driver but then I get the no card detected errors. I did see in some posts that someone upgraded to 6.11.0-rc4 if I recall so I did the upgrade to 6.11.0-rc5 and redownloaded from the CA app and took less than 5 min all done and working now. 

48 minutes ago, HomeStudios said:

I did see in some posts that someone upgraded to 6.11.0-rc4 if I recall so I did the upgrade to 6.11.0-rc5 and redownloaded from the CA app and took less than 5 min all done and working now.

Please keep in mind that I downgraded my test server to 6.10.3 and verified that the download is working indeed, I think there is some weird issue with GitHub sometimes and certain files.

 

Anyways, glad that all is working now.

4 hours ago, ich777 said:

Please keep in mind that I downgraded my test server to 6.10.3 and verified that the download is working indeed, I think there is some weird issue with GitHub sometimes and certain files.

 

Anyways, glad that all is working now.

Yes thanks all working now I will keep going on 6.11 for now, I also think it's weird as it was working 100% on 6.9 it was after the upgrade to 6.10 I have started to see some weird thing happening.

But thanks all working good now had some Dockers I had to reinstall but looks like it's all working again from what I can see.

 

Edited September 14, 20224 yr by HomeStudios

After I install the driver, the Docker service cannot be started. Please help me diagnose the problem. Thank you.

 

EDIT: SOLVED

 

Edited September 16, 20224 yr by ich777
Removed Diagnostics

20 minutes ago, unraidxiaobai said:

Please help me diagnose the problem.

You inject a private Docker registry to /etc/docker/daemon.json from your go file.

Please remove that from your go file and reboot your server.

After you‘ve rebooted you will see that your daemon.json has the changes for the nvidia runtime in it.

If you want to keep that private Docker registry, you have to also customize your injection into the file that it includes this changes for the nvidia runtime.

 

From what I saw it also looks very suspicious that you have:

{undefined

at the start, are you sure that this is correct?

I am running Unraid 6.10.3 under the esxi 6.7u3 with passthrough devices.  I have found that current nvidia drivers do not work in that configuration.  Last driver version that was working correctly is 470.82.00 . 

 

How do I build / install this driver on my system ?  How do I prevent nvidia-driver package from "upgrading" that to an incompatible version ?

 

Here is the Nvidia thread explaining the issue. https://forums.developer.nvidia.com/t/nvidia-smi-no-devices-were-found-vmware-esxi-ubuntu-server-20-04-03-with-rtx3070/202904/36

49 minutes ago, Darek said:

I am running Unraid 6.10.3

I have to be completely honest, I'm a little frustrated because you are double posting (Private Message and here in the support thread).

I won't answer any quicker...

 

I've already answered via PM but I will copy paste it here and leave the first paragraph out because you've don't mentioned the Unraid version in your PM:

Quote

 

55 minutes ago, Darek said:

alternatively provide a way to install it and not automatically upgrade/remove

 

What do you mean with this? If you upgrade the Unraid version or on reboots?

If you don't want to upgrade it on reboots then simply change the driver version to the version that you want to stay on instead of selecting latest.

I can't/wont change the behavior for Unraid upgrades since I don't know how long Nvidia will support the legacy 470.xx drivers, if they want to they can support tomorrow (or even today). That's the reason why there is no legacy button in the driver plugin in terms of the version selection.

 

Quote

 

55 minutes ago, Darek said:

The 470.82 drivers were the last proprietary drivers to work in that configuration

 

I can't compile these drivers for newer Unraid versions because they won't compile against newer Kernel versions which Unraid is on, if you want that old drivers you have to stick with that old Unraid versions.

 

Quote

 

55 minutes ago, Darek said:

515.43.04/kernel-opensource

 

Please remember this is the proprietary driver and not the open source one... I also won't compile the open source Kernel module because this would basically break the driver for every card pre Ampere so to speak this would also break compatibility with your system because you are having a Turing card.

 

BTW, are you sure that driver version 470.141.03 is also not working because these are the "newer" legacy drivers which actually support newer Kernel versions: Click

I'm having trouble with my P600, it fails when starting the container on both Plex and Jellyfin

I'm using the R5 2400G, is it possible that the iGPU is messing with the NV somehow?

I also have a diagnostics zip, but since I'm checking if creating it from the CLI anonymises data (gui seems borked somehow)

7 hours ago, DavidNguyen said:

I also have a diagnostics zip

Please post them, without them I really can‘t say much…

 

7 hours ago, DavidNguyen said:

(gui seems borked somehow)

What do you mean with that?