Unknown runtime specified nvidia.

34 minutes ago, jason0453 said:

Unknown runtime specified nvidia.

Can you please be a bit more specific? What have you done so far? With that less context I really can't help.

 

Have you rebooted or restarted the Docker service like the red notification tells you to after you've installed the driver? Have you read the first post in this thread on how to install the driver correctly?

 

Please always include your Diagnostics if you have such issues.

7 小时前，ich777 说：

你怎么做？

 

在安装驱动程序之后，您是否像红色通知一样通知您该重新启动或重新启动 Docker 服务？是否启动过阅读主题中关于如何安装正确驱动程序的第一篇文章？

 

如果您有此类问题，请始终包含您的诊断信息。

我在额外参数里添加了 --runtime=nvidia 以后docker就不能运行了

提示docker: Error response from daemon: Unknown runtime specified nvidia.

4 minutes ago, jason0453 said:

提示docker: Error response from daemon: Unknown runtime specified nvidia.

Have you restarted the Docker service or your server?

Please post your Diagnostics.

之前好用，后来突然就不好使了

tower-diagnostics-20220311-2337.zip

15 hours ago, jason0453 said:

之前好用，后来突然就不好使了

Have you yet tried to reboot the server?

 

You can also try to upgrade to RC3 and see if it starts working there.

 

I have now tried the exact same driver on my test machine on Unraid 6.10.0-rc2 and it works flawlessly (with a Nvidia T400).

I had a spare 3060 Ti  laying around so thought i'd throw it in my unraid server. NVIDIA-Plugin installed no problem and working brilliantly for my transcoding. 

 

But I am noticing that the fans ramp up to 92% speed around every 30-60 seconds (or so) 

 

Is there a way to prevent that and have the fans sit idle if the GPU is doing nothing?

19 minutes ago, z0ki said:

But I am noticing that the fans ramp up to 92% speed around every 30-60 seconds (or so) 

When transcoding a file or when the transcoding finished? Do you use Plex? Plex usually transcodes a few minutes from the file to buffer it and then throttles itself.

 

Do you have nvidia-persistenced enabled? If not try to issue:

nvidia-persistenced

from a Unraid terminal (you will get no output but this means it's working) and then try to transcode a file again and see if it's the same.

 

If that doesn't help you are maybe out of luck because the FAN speed/curve is set in the BIOS and there is nothing you can do about that.

 

19 minutes ago, z0ki said:

Is there a way to prevent that and have the fans sit idle if the GPU is doing nothing?

nvidia-persistenced should help hopefully.

 

If you want this to persist between reboots simply add the command at the end of your go file.

Thanks for the responses @ich777 under transcoding conditions i expect the GPU fans to ramp up, but this is when the card is just sitting in a IDLE state and not being used by anything.

 

I tried nvidia-persistenced just now and it looks like it did not do anything. So I am guessing you'd be right that its something nvidia put in the vBios

FYI looks like this plugin is not ready for 6.10.0-rc4, got the message that the driver could not be downloaded, immediately rolled back to RC3.

7 minutes ago, bubbl3 said:

FYI looks like this plugin is not ready for 6.10.0-rc4, got the message that the driver could not be downloaded, immediately rolled back to RC3.

It is only an error in the plugin helper. ich777 posted this in another thread.

 

investigated a little further and found the cause of the issue, the naming from the Linux Kernel version changed, updated the Plugin-Update-Helper to take that into account. Hopefully it will work next time.

 

It is safe to reboot because the plugins are downloaded on boot too, just make sure that you have a active Internet connection without any AdBlocking software in front of it.

If you don't have a active Internet connection on boot a plugin removal from the Plugin Error tab and a new installation will fix it.

34 minutes ago, bubbl3 said:

FYI looks like this plugin is not ready for 6.10.0-rc4

Plugin is ready, see here: Click

 

34 minutes ago, bubbl3 said:

immediately rolled back to RC3

Can you try to upgrade again to RC4 and tell me if it is now working?

31 minutes ago, ich777 said:

Plugin is ready, see here: Click

 

Can you try to upgrade again to RC4 and tell me if it is now working?

Downgrade to rc3 then back to rc4 error still appears

7 minutes ago, SimonF said:

Downgrade to rc3 then back to rc4 error still appears

What do you get from:

cat /tmp/update-helper

 

2 minutes ago, ich777 said:

What do you get from:

cat /tmp/update-helper

 

root@computenode:~# cat /tmp/update-helper 
2022.03.19
root@computenode:~# 

Just now, SimonF said:

2022.03.19

Can you try to downgrade once more and reboot so that it pulls the new version 2022.03.20 and then try to upgrade once more?

5 minutes ago, ich777 said:

Can you try to downgrade once more and reboot so that it pulls the new version 2022.03.20 and then try to upgrade once more?

Yes looking better now.

 

1 minute ago, SimonF said:

Yes looking better now.

Thank you as always for your help!

 

Really strange that it reports that CoreFreq isn't found...

Tried as well:

• upgraded to RC4
• error, but after reboot all is good
• downgraded to RC3
• reboot, all is good
• upgrade to RC4, no more errors
• reboot, all is good

 

Thanks for fixing this so fast!

I tried upgrading to RC4 and noticed that all dockers that use the nvidia driver fail to start with a bad parameter error. Checked the nvidia driver plugin and it shows that no drivers are installed. When I try to manually download the driver I get this-

+==============================================================================
| WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING
|
| Don't close this window with the red 'X' in the top right corner until the 'DONE' button is displayed!
|
| WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING - WARNING
+==============================================================================

----------------Downloading Nvidia Driver Package v-----------------
---------This could take some time, please don't close this window!------------

---------------Can't download Nvidia Driver Package v----------------

 

 

brunnhilde-diagnostics-20220320-0910.zip

After rebooting the nvidia driver plugin is no longer installed.

1 minute ago, wgstarks said:

After rebooting the nvidia driver plugin is no longer installed.

Do you have a active Internet connection on boot, so to speak do you run any AdBlocking software or Firewall on Unraid?

 

Please go to your Plugins page and see if you got a Plugins Error tab, if so please remove the Nvidia Driver plugin and install it from the CA App again.

1 minute ago, ich777 said:

Do you have a active Internet connection on boot, so to speak do you run any AdBlocking software or Firewall on Unraid?

Yes. I don't use any ad blockers.

 

2 minutes ago, ich777 said:

Please go to your Plugins page and see if you got a Plugins Error tab, if so please remove the Nvidia Driver plugin and install it from the CA App again.

Nvidia driver plugin showed up in Apps>Previous Apps and I have re-installed it. It showed that it downloaded and installed the driver as part of the installation process. There wasn't any prompt for a reboot but I did anyway. Found this in my email inbox so I'm guessing a reboot was overkill (although this wasn't the first time the app was installed)-

Event: Nvidia Driver
Subject: Notification
Description: Please make sure to disable and enable Docker if you installed the Nvidia driver for the first time! Settings -> Docker -> Enable Docker 'No' -> Apply -> Enable Docker 'Yes' -> Apply
Importance: alert

 

 

Regardless, it seems the issue is now resolved. All dockers have started successfully.

On 11/20/2020 at 5:20 PM, Saurabh Sharma said:

Will this work on a GTX560Ti? I need it for Jellyfin. nVidia website only recommends up to version 390.138. On my app search, nothing came up with "nvidia" so my card may be too old for this world. Does that mean it is not compatible? 

And just to confirm, I can't have this on Plex without paying up for plex pass, right?

seems not. there is no 390.xx opptions.  sad 

my nvs4200m cannot compatible, either

 

Latest Legacy GPU version (390.xx series): 390.147

Mar 23 22:05:31 Tower root: -----------------Installing Nvidia Driver Package v510.60.02-------------------
Mar 23 22:06:29 Tower kernel: Linux agpgart interface v0.103
Mar 23 22:06:29 Tower kernel: nvidia: loading out-of-tree module taints kernel.
Mar 23 22:06:29 Tower kernel: nvidia: module license 'NVIDIA' taints kernel.
Mar 23 22:06:29 Tower kernel: Disabling lock debugging due to kernel taint
Mar 23 22:06:29 Tower kernel: nvidia-nvlink: Nvlink Core is being initialized, major device number 246
Mar 23 22:06:29 Tower kernel: NVRM: The NVIDIA NVS 4200M GPU installed in this system is
Mar 23 22:06:29 Tower kernel: NVRM:  supported through the NVIDIA 390.xx Legacy drivers. Please
Mar 23 22:06:29 Tower kernel: NVRM:  visit http://www.nvidia.com/object/unix.html for more
Mar 23 22:06:29 Tower kernel: NVRM:  information.  The 510.60.02 NVIDIA driver will ignore
Mar 23 22:06:29 Tower kernel: NVRM:  this GPU.  Continuing probe...
Mar 23 22:06:29 Tower kernel: NVRM: No NVIDIA GPU found.

it seems that my GPU is too old.

Is there anyway to make my gpu work ? 

I've tried to download the '.run' file of 390.xx. but i cannot cope with the installation on the unraid Slackware system, gcc glib and other depends software.