Unable to Update/install NVIDIA drivers in template Unable to connect Server: tinyproxy/1.10.0

User Support
1

I am trying to follow the guide to use Virtual GL to pass through a GPU and use it on a single monitor.

I created a new template based on Debian-12-xfce and am trying to install the NVIDIA drivers into it.

I am getting the following errors:

$ sudo apt install nvidia-driver nvidia-kernel-open-dkms cuda nvidia-smi nvidia-settings
Reading package lists… Done
Building dependency tree… Done
Reading state information… Done
nvidia-driver is already the newest version (545.23.08-1).
cuda is already the newest version (12.3.1-1).
nvidia-smi is already the newest version (545.23.08-1).
nvidia-settings is already the newest version (545.23.08-1).
The following packages will be REMOVED:
nvidia-kernel-dkms
The following NEW packages will be installed:
nvidia-kernel-open-dkms
0 upgraded, 1 newly installed, 1 to remove and 91 not upgraded.
Need to get 4,630 kB of archives.
After this operation, 52.5 MB disk space will be freed.
Do you want to continue? [Y/n] y
Ign:1 Index of /compute/cuda/repos/debian12/x86_64 nvidia-kernel-open-dkms 545.23.08-1
Ign:1 Index of /compute/cuda/repos/debian12/x86_64 nvidia-kernel-open-dkms 545.23.08-1
Ign:1 Index of /compute/cuda/repos/debian12/x86_64 nvidia-kernel-open-dkms 545.23.08-1
Err:1 Index of /compute/cuda/repos/debian12/x86_64 nvidia-kernel-open-dkms 545.23.08-1
Invalid response from proxy: HTTP/1.0 500 Unable to connect Server: tinyproxy/1.10.0 Content-Type: text/html Connection: close [IP: 127.0.0.1 8082]
E: Failed to fetch https://developer.download.nvidia.com/compute/cuda/repos/debian12/x86_64/./nvidia-kernel-open-dkms_545.23.08-1_amd64.deb Invalid response from proxy: HTTP/1.0 500 Unable to connect Server: tinyproxy/1.10.0 Content-Type: text/html Connection: close [IP: 127.0.0.1 8082]
E: Unable to fetch some archives, maybe run apt-get update or try with --fix-missing?

I get this error when I set sys-net for the update, sys-firewall for the update, sys-whonix for the update, and sys-vpn for the update.

I have also tried sudo apt update in the template and get nothing with the same sort of errors.

I connected the template to VPN to download the NVIDIA key through curl. I know I should have done this in another temporary VM and copied it, but I don’t have time and may end up deleting this template anyway. It is mostly for testing.

What am I doing wrong?

2

I also am making this template from a cloned template that has the drivers installed but I wanted to add the open versions of the drivers.

3

Another problem, when I run lspci | grep -e VGA

it shows 2 compatible controllers, the one I attached in Settings for the HVM and another.

4

Uh oh. Your card is not supported by any driver version up to 545.23.08.
A newer driver may add support for your card.
Newer driver releases may be available in backports, unstable or experimental.

I also am getting this error message after running nvidia-detect

In another community guide, it was mentioned that Arch has more bleeding edge releases and Debian has older releases and so the person prefers Arch. I didn’t get it. Now I get it.

I want to run something that uses Ubuntu and so I don’t want to try to run that using Arch. I can’t run it in Fedora.

Does this Virtual GL Qube need to be based on the same template as the Qube using it? Can I run a program in Ubuntu using an Ubuntu template and attach a GPU using Virtual GL somehow in an Arch template?

sorry i’m so dum

5

it also shows no GPU on nvidia-smi or just shuts down