Skip to content

[UX] Better logging when user program OOM'ed #4199

Closed
@Michaelvll

Description

Currently, if a user program OOM'ed, our logs for the job will not include any information of the failure reason and no returncode will be shown.

We should improve the UX for this with the following steps:

  1. Include necessary logs in the sky logs, e.g, OOM and returncode.
  2. Improve the sky queue with additional failure reasons instead of just a FAILED state of the job.

Version & Commit info:

  • sky -v: PLEASE_FILL_IN
  • sky -c: PLEASE_FILL_IN

Metadata

Assignees

No one assigned

    Labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions