Skip to content

Running Container Lost GPU Permission #50

Description

@yukirora

How to reproduce the issue:

  1. Create a job on the platform and run nvidia-smi in a infinite loop, it works well
  2. Run the sudo systemctl daemon-reload on the VM in the job
  3. The job will show 'Failed to initialize NVML: Unknown Error' failure

Related Issues:
NVIDIA/gpu-operator#485
NVIDIA/nvidia-container-toolkit#48

Activity

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Metadata

Metadata

Labels

No labels
No labels

Type

No type

Projects

No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions