Evan Lezar

Thanks for the confirmation @zvier. @gwgrisk Note that with newer versions of systemd and using systemd cgroup management, it is also required to specify the `PASS_DEVICE_SPECS` option. It is thus...

The primary issue here is that the device plugin is started in privileged mode to have access to the device nodes for enumeration. This means that the `NVIDIA_VISIBLE_DEVICES` environment variable...

@premalathak12 is the `nvidia` runtime configured as the default runtime in crio? If not, a runtime class must be created and associated with the `nvidia` runtime, and the runtime class...

@A-Akhil could you please provide information on the helm values that you are providing to the plugin? Do any of the plugin containers show any logs?

@A-Akhil have you installed the NVIDIA Container Toolkit on the host and configured containerd with this runtime? Furthermore if the `nvidia` runtime is not the default runtime for Containerd you...

@zhangQiWorr are you deploying the device plugin using helm? The use of MPS in the device plugin requires both GFD and an additional component to manage the lifecycle of the...

@jakubkrzykowski this sounds like a bug. Looking at the helm helper, we definitely don't check for the `null` value. Would you be able to submit a PR with a fix?...

This should be addressed by #666.

@tariq1890 I'm going to merge this. Could you create a follow-up that adds a config option to allow the deprecated labels to be disabled and optionally omit them. Something along...

@yunfeng-scale since you mention the GPU Operator being used, could you please confirm the GPU Operator version that is being used to deploy the v0.14.0 version of the device plugin?