Closed cocakohler closed 1 year ago
Please try installing the libnvidia-gl-510-server
package in addition to the current packages.
Many of the OpenGL files are broken out into their own packages, and we don't install these by default in DeepOps because our clusters are mostly not used for graphical workloads.
@ajdecon So doing sudo apt install libnvidia-gl-510
should be the way to go?
Do you expect any side-effects by doing that?
That's my suggestion based on the error presented. I'm not experienced with Isaac Sim so I don't know for sure this will work. 😄
I don't expect any side effects from the extra package(s) being installed.
This issue is stale because it has been open for 60 days with no activity. Please update the issue or it will be closed in 7 days.
sudo apt install libnvidia-gl-510-server made the fix.
I deployed DeepOps K8s on a single node with 4 RTX A6000 GPUs built in. I can run GPU-accelerated containers (e.g. nvdia-smi) without an issues. But i'm running into problems when running the NVIDIA Isaac Sim container https://catalog.ngc.nvidia.com/orgs/nvidia/containers/isaac-sim When i start the container i get the follwing error:
Fatal Error: Can't find libGLX_nvidia.so.0
As far as I understand this libGLX_nvidia.so.0 is part of the Vulkan (https://developer.nvidia.com/vulkan) which should be part of the driver.
When I do
gpu@gpu01:~$ dpkg-query -l | grep nvidia
i get the following output:
I'm afraid of doing a manual update of the driver package. Any ideas?