omv6 and nvidia gpu

nsas02 you should stick to the versions that are provided with the repositories. OMV6 / Debian 11 comes with nvidia 4xx drivers, OMV7 / Debian 12 comes with nvidia 5xx drivers. There is no difference in my tutorial for 4xx and 5xx drivers.

i am trying to install the nvidia drivers on OMV 7 according your guide but running into this error:

dpkg: dependency problems prevent configuration of nvidia-driver:
 nvidia-driver depends on nvidia-kernel-dkms (= 525.147.05-4~deb12u1) | nvidia-kernel-525.147.05 | nvidia-open-kernel-52
5.147.05 | nvidia-open-kernel-525.147.05; however:
  Package nvidia-kernel-dkms is not configured yet.
  Package nvidia-kernel-525.147.05 is not installed.
  Package nvidia-kernel-dkms which provides nvidia-kernel-525.147.05 is not configured yet.
  Package nvidia-open-kernel-525.147.05 is not installed.
  Package nvidia-open-kernel-525.147.05 is not installed.

dpkg: error processing package nvidia-driver (--configure):
 dependency problems - leaving unconfigured
Processing triggers for libc-bin (2.36-9+deb12u4) ...
Processing triggers for initramfs-tools (0.142) ...
update-initramfs: Generating /boot/initrd.img-6.1.0-18-amd64
Processing triggers for update-glx (1.2.2) ...
Processing triggers for glx-alternative-nvidia (1.2.2) ...
update-alternatives: using /usr/lib/nvidia to provide /usr/lib/glx (glx) in auto mode
Processing triggers for glx-alternative-mesa (1.2.2) ...
Processing triggers for libc-bin (2.36-9+deb12u4) ...
Processing triggers for initramfs-tools (0.142) ...
update-initramfs: Generating /boot/initrd.img-6.1.0-18-amd64
Errors were encountered while processing:
 nvidia-kernel-dkms
 nvidia-driver

tried installing on a plain debian VM and it went without errors

Wich kernel are you using on omv7 ? Do you use the proxmox-kernel ? And did you try to remove old nvidia packages ? For me it seem that there are some older dependencies

i am using the stock OMV7 kernel

Linux omv6 6.1.0-18-amd64 #1 SMP PREEMPT_DYNAMIC Debian 6.1.76-1 (2024-02-01) x86_64 GNU/Linux)

did perform an apt purge *nvidia* serveral times, but did not help.

omv6 is the hostname of my install, but i am on 7 actually

I am using the pve kernels. They do work fine. Try to switch to the pve kernel and enable the backports repositories.

switching to pve kernel allows to install the nvidia drivers without error, but i run into issues with sudo nvidia-xconfig - it does not find the GPU:

$ sudo nvidia-xconfig

Using X configuration file: "/etc/X11/xorg.conf".

VALIDATION ERROR: Data incomplete in file /etc/X11/xorg.conf.
                  Device section "Device0" must have a Driver line.


WARNING: error opening libnvidia-cfg.so.1: libnvidia-cfg.so.1: cannot open shared object file: No such file or
         directory.


ERROR: Unable to find any GPUs in the system.

Backed up file '/etc/X11/xorg.conf' as '/etc/X11/xorg.conf.nvidia-xconfig-original'
Backed up file '/etc/X11/xorg.conf' as '/etc/X11/xorg.conf.backup'
New X configuration file written to '/etc/X11/xorg.conf'

although in lspci i can clearly see the device

02:00.0 3D controller: NVIDIA Corporation GP104GL [Tesla P4] (rev a1)

Hm, I didn’t run into that issue until now.

Does it work if you leave out the xconfig part?

Or may be it helps if you uninstall nvidia-xconfig with purge parameter and reinstall it ?

apt-get remove nvidia-xconfig —purge

Zitat von molnart

switching to pve kernel allows to install the nvidia drivers without error, but i run into issues with sudo nvidia-xconfig - it does not find the GPU:

Code

$ sudo nvidia-xconfig

Using X configuration file: "/etc/X11/xorg.conf".

VALIDATION ERROR: Data incomplete in file /etc/X11/xorg.conf.
                  Device section "Device0" must have a Driver line.


WARNING: error opening libnvidia-cfg.so.1: libnvidia-cfg.so.1: cannot open shared object file: No such file or
         directory.


ERROR: Unable to find any GPUs in the system.

Backed up file '/etc/X11/xorg.conf' as '/etc/X11/xorg.conf.nvidia-xconfig-original'
Backed up file '/etc/X11/xorg.conf' as '/etc/X11/xorg.conf.backup'
New X configuration file written to '/etc/X11/xorg.conf'

Alles anzeigen

although in lspci i can clearly see the device

02:00.0 3D controller: NVIDIA Corporation GP104GL [Tesla P4] (rev a1)

i better stop playing with that, because those drivers completely broke docker somehow and i has to restore my whole OMV installation from backup.

You definitely do not need the X configuration file but just in case got no issue:

Your X configuration file has been successfully updated. Installation of the NVIDIA Accelerated Graphics Driver for Linux-x86_64

(version: 550.54.14) is now complete.

The 550 driver is not working - at least on my system. Plex is not hw-transcoding. The latest stable nvidia driver on debian 12/bookworm is the 525 version.

Working for me

For me too, but plex is not hw-transcoding with that driver. As far as I remeber you are not using plex, right Nefertiti ?

chris_kmn no I am using Emby but I can do some test in few days with Plex.

I just did the OMV6 -> OMV7 update on the weekend, which failed miserably twice in a row for me. Going through the logs, it was complaining about <guess what> the Nvidia drivers... I had to completely purge the Nvidia drivers from my system before the upgrade would successfully complete. However, this left me with no transcoding h/w, so I had to go into Jellyfin and Plex dockers and completely de-reference the Nvidia integration.

I did not spend a lot of time debugging, but from what I can see, the 470 drivers (and perhaps my video card itself?) are no longer supported in the Debian version that comes with OMV7 (just a guess). Part of the upgrade script output made reference to installing "Nouveau" Nvidia drivers, but at this point, screw it, I'm simply going to yank the card from my system and do direct streaming for everything!

Time to buy a new, Intel based server with video capability in the CPU itself...

I haven‘t had any issues with upgrading to omv7 - but I followed a tutorial that was posted somewhere here in the forum for upgrading from 6 -> 7.

And yes, omv7 with debian 12 uses nvidia 5xx drivers.

Zitat von chris_kmn

omv7 with debian 12 uses nvidia 5xx drivers.

I might try to re-run the Nvidia driver install guide found elsewhere in this forum to see if it grabs the 5xx drivers and they actually get used by my Plex install, BEFORE I ditch my graphics card...

Zitat von Cap10Canuck

I might try to re-run the Nvidia driver install guide...

Tried to install the Nvidia drivers again as per the install guide and the install stopped at the following:

"This system has a graphics card which is no longer handled by the NVIDIA driver (package nvidia-driver).

You may wish to keep the package installed - for instance to drive some other card - but the card with the following chipset won't be usable.

06:00.0 VGA compatible controller [0300]: NVIDIA Corporation GK104 [GeForce GTX 760] [10de:1187] (rev a1)

The above card requires either the non-free legacy NVIDIA driver (package nvidia-tesla-470-driver) or the free Nouveau driver (package xserver-xorg-video-nouveau). Use the update-glx command to switch between different installed drivers. │

Before the Nouveau driver can be used you must remove NVIDIA configuration from xorg.conf (and xorg.conf.d/)."

At this point I may just throw in the towel, given it errored-out at only step 3 of a relatively complex set of installation instructions.

Zitat von chris_kmn

For me too, but plex is not hw-transcoding with that driver. As far as I remeber you are not using plex, right Nefertiti ?

I must say I never use plex, but the container is installed so it was easy to enable it I got pass anyway. So in Plex Transcoder-Advanced under Hardware transcoding device My Nvidia GPU is listed along the Intel one, and it is running with this driver5.50.54.14 Same for Tdarr no issue.

finally i got this running. for me the steps where the following:

1) install proxmox kernel

2) disable apparmor

3) install the drivers according to the instructions, but ignore the xconfig part

now everything seems ok, but i could not try it as in the meantime i moved all my containers that could utilize the GPU to a different VM, only to realize i can't pass the GPU there because its tied with my HBA in a single IOMMU group

I ran into this issue I mentioned post #216 RE: omv6 and nvidia gpu