Evan Lezar

I think your understanding is correct. Using the toolkit daemonset should configure the toolkit as required. Note that you would also need to set the `ACCEPT_NVIDIA_VISIBLE_DEVICES_AS_VOLUME_MOUNTS="true"` for the toolkit to...

@Alwinator could you please confirm your GPU operator version and values that you have applied to your helm deployment?

> OpenShift 4.10.35 with Kubernetes 1.23 OS on nodes: Red Hat Enterprise Linux CoreOS 410.84.202209231843-0 CRI-O 1.23.3-17.rhaos4.10.git016b1ca.el8 GPU Operator version: 22.9.0 Helm version: v3.6.3 Thanks. And the values used to...

@Alwinator as mentioned in https://github.com/NVIDIA/gpu-operator/issues/421#issuecomment-1281983276 could you ALSO please set the `ACCEPT_NVIDIA_VISIBLE_DEVICES_AS_VOLUME_MOUNTS="true"` in the `toolkit.env` and see if this addresses your issue? What is happening now is that the device...

You seem to be running the device plugin under WSL2. This is not currently a supported use case of the device plugin. The specific reason is that device nodes on...

@patrykkaj I think that in theory this could be done by outside contributors and is simplified by the recent changes to support Tegra-based systems. What I can see happening here...

@Vinrobot thanks for the work here. Some thoughts on this: We recently moved away from `nvidia-gpu-monitoring-tools` and use bindings from [`go-nvml`](https://github.com/NVIDIA/go-nvml) through [`go-nvlib`](https://gitlab.com/nvidia/cloud-native/go-nvlib) instead. I think the steps outlined in...

@Vinrobot yes, it is an issue that `gpuallocator` still uses `gpu-monitoring-tools`. It is on our roadmap to port it to the `go-nvml` bindings, but this is not yet complete. The...

This is most likely due to how [pynvml](https://pypi.org/project/pynvml/) (or this one https://pypi.org/project/nvidia-ml-py/#history) is wrapping the underlying NVML library. It seems to require certain `_v2` symbols that are not present in...

The device plugin can be used with docker, containerd, or cri-o as long as these container engines are configured accordingly. Support for jetson nano devices is in progress and should...