The ray job status is always RUNNING

Ray Libraries (Data, Train, Tune, Serve) Ray Train
1

How severe does this issue affect your experience of using Ray?

  • High: It blocks me to complete my task.

I built a training pipeline using ray data and ray train. After the training task is completed, the status of ray is always RUNNING.

In the dashboard, it can be observed that some tasks of ray data are still in running status, but I actually want this task to be in the successful status.

image
image1113×1120 47.7 KB

I hive 2 questions:

  1. How do I know when the training task is over?
  2. How do I set the status of SUCCEEDED?

ps: I submitted the job this way: ray.job_submission.JobSubmissionClient.submit_job — Ray 2.9.3

1 Like
2

Are you still running into this? Could you ask this on the Ray Slack?