Cant install Nvidia Drivers

[1liminal1]

Hey guys,
I tried installing some drivers from the Nvidia website that asked me if I wanted to make changes to the dkms kernel in case I wanted to make changes later and I said yes. It failed to install the drivers and I cant install any other drivers using this guide (Updated for bullseye)

Can I fix this?

Error! Bad return status for module build on kernel: 5.15.39-3-pve (x86_64)
Consult /var/lib/dkms/nvidia-current/418.211.00/build/make.log for more information.
dpkg: error processing package nvidia-kernel-dkms (--configure):
installed nvidia-kernel-dkms package post-installation script subprocess returned error exit status 10
dpkg: dependency problems prevent configuration of nvidia-driver:
nvidia-driver depends on nvidia-kernel-dkms (= 418.211.00-1) | nvidia-kernel-418.211.00; however:
Package nvidia-kernel-dkms is not configured yet.
Package nvidia-kernel-418.211.00 is not installed.
Package nvidia-kernel-dkms which provides nvidia-kernel-418.211.00 is not configured yet.

dpkg: error processing package nvidia-driver (--configure):
dependency problems - leaving unconfigured
Processing triggers for libgdk-pixbuf-2.0-0:amd64 (2.42.2+dfsg-1) ...
Processing triggers for libc-bin (2.31-13+deb11u3) ...
Processing triggers for initramfs-tools (0.140) ...
update-initramfs: Generating /boot/initrd.img-5.15.39-3-pve
Running hook script 'zz-proxmox-boot'..
Re-executing '/etc/kernel/postinst.d/zz-proxmox-boot' in new private mount namespace..
No /etc/kernel/proxmox-boot-uuids found, skipping ESP sync.
Processing triggers for update-glx (1.2.1~deb11u1) ...
Processing triggers for glx-alternative-nvidia (1.2.1~deb11u1) ...
update-alternatives: using /usr/lib/nvidia to provide /usr/lib/glx (glx) in auto mode
Processing triggers for glx-alternative-mesa (1.2.1~deb11u1) ...
Processing triggers for libc-bin (2.31-13+deb11u3) ...
Processing triggers for initramfs-tools (0.140) ...
update-initramfs: Generating /boot/initrd.img-5.15.39-3-pve
Running hook script 'zz-proxmox-boot'..
Re-executing '/etc/kernel/postinst.d/zz-proxmox-boot' in new private mount namespace..
No /etc/kernel/proxmox-boot-uuids found, skipping ESP sync.
Errors were encountered while processing:
nvidia-kernel-dkms
nvidia-driver

 

[Neobin]

Error! Bad return status for module build on kernel: 5.15.39-3-pve (x86_64)
Consult /var/lib/dkms/nvidia-current/418.211.00/build/make.log for more information.

No expert, but I highly assume the 418 Nvidia driver (which is currently in the old Debian 10/Buster repository [1] and was only tested with kernels up to 5.10) is too old for the 5.15 kernel (5.15-pve is Debian 11/Bullseye).

So you should use the nvidia-driver from bullseye [2] or bullseye-backports [3] (whichever is newer).

Also do not forget to install the PVE-kernel-headers before: apt install pve-headers (The PVE-kernel has also to be up-to-date and booted on before, because this meta-package only pulls the headers for the latest kernel in.)

Can I fix this?

I don't know, sorry.

[1] https://packages.debian.org/buster/nvidia-driver
[2] https://packages.debian.org/bullseye/nvidia-driver
[3] https://packages.debian.org/bullseye-backports/nvidia-driver

 

[1liminal1]

Thank you for having a crack

Yeah so it seems like its stuck with this 418 driver that wouldn't install. Anything I try from Bullseye, gets stuck on this

Package nvidia-kernel-418.211.00 is not installed.

I've tried to purge nvidia, but... I dont know why it keeps getting stuck with this. I cant even remove it (though it says it is)
Im going to try the nvidia driver clean up tool.

 

[LnxBil]

Just search the forums, we already discussed this a few month ago. You need to use the bullseye-backports nvidia modules and drivers, they work flawlessly (just updated one of my PVE boxes with nvidia to the newest version without a problem.

 

[1liminal1]

I've been though that with a fine tooth comb lol

No matter what I do I get the same error.

Ive added the packes from here and tried

https://forum.proxmox.com/threads/install-nvidia-drivers.99597/page-2

Ived tried these drivers

apt-get install -t bullseye-backports nvidia-driver nvidia-smi

Ive tried

apt install nvidia-drivers (recommended by NVidia detect)

I get this error,

------------------------------
Deleting module version: 470.129.06
completely from the DKMS tree.
------------------------------
Done.
Loading new nvidia-current-470.129.06 DKMS files...
Building for 5.15.39-3-pve
Building initial module for 5.15.39-3-pve
Error! Bad return status for module build on kernel: 5.15.39-3-pve (x86_64)
Consult /var/lib/dkms/nvidia-current/470.129.06/build/make.log for more information.
dpkg: error processing package nvidia-kernel-dkms (--configure):
 installed nvidia-kernel-dkms package post-installation script subprocess returned error exit status 10
dpkg: dependency problems prevent configuration of nvidia-driver:
 nvidia-driver depends on nvidia-kernel-dkms (= 470.129.06-6~deb11u1) | nvidia-kernel-470.129.06; however:
  Package nvidia-kernel-dkms is not configured yet.
  Package nvidia-kernel-470.129.06 is not installed.
  Package nvidia-kernel-dkms which provides nvidia-kernel-470.129.06 is not configured yet.

dpkg: error processing package nvidia-driver (--configure):
 dependency problems - leaving unconfigured
Errors were encountered while processing:
 nvidia-kernel-dkms
 nvidia-driver

SO I check the logs and get this

ERROR: Kernel configuration is invalid.
include/generated/autoconf.h or include/config/auto.conf are missing.
Run 'make oldconfig && make prepare' on kernel src to fix it.


I google that and find THIS!!!

apt reinstall pve-headers-$(uname -r)

Which finally fixed it... at last LOL

 

[1liminal1]

Also do not forget to install the PVE-kernel-headers before: apt install pve-headers (The PVE-kernel has also to be up-to-date and booted on before, because this meta-package only pulls the headers for the latest kernel in.)

So Neobin was closest in the end, But they actually needed to be reinstalled