Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[Feature] Support get k8s pod log for interactive RayJob #2701

Open
1 of 2 tasks
Moonquakes opened this issue Dec 31, 2024 · 3 comments
Open
1 of 2 tasks

[Feature] Support get k8s pod log for interactive RayJob #2701

Moonquakes opened this issue Dec 31, 2024 · 3 comments
Labels
enhancement New feature or request triage

Comments

@Moonquakes
Copy link

Search before asking

  • I had searched in the issues and found no similar feature requirement.

Description

Currently, RayJob supports interactive submission, but there is a problem that without the submitter pod, the log information of RayJob cannot be obtained on the k8s cluster. Is there any solution?

Use case

No response

Related issues

No response

Are you willing to submit a PR?

  • Yes I am willing to submit a PR!
@Moonquakes Moonquakes added enhancement New feature or request triage labels Dec 31, 2024
@rueian
Copy link
Contributor

rueian commented Jan 1, 2025

Hi @Moonquakes,

You can find job logs at /tmp/ray/session_date/logs/job-driver-{job_id}.log in your ray cluster pods.

@Moonquakes
Copy link
Author

Hi @rueian, Thanks for your reply, but when RayJob is killed, the head node no longer exists, and in normal mode I can always get the corresponding log information through the submitter pod. Is there any way to achieve a similar effect in interactive mode?

@rueian
Copy link
Contributor

rueian commented Jan 2, 2025

Hi @Moonquakes,

The job driver logs are streamed to the stdout of the submitter (the ray job submit cli tool) and are stored in the logs folder I mentioned earlier. These are the only two places the logs present.

If you don’t have the former logs stored somewhere, then persisting the latter somehow is the only choice available right now. I would suggest you follow the tutorial here for persisting them https://docs.ray.io/en/master/cluster/kubernetes/user-guides/persist-kuberay-custom-resource-logs.html

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
enhancement New feature or request triage
Projects
None yet
Development

No branches or pull requests

2 participants