Skip to content

The Unmanaged Service Accounts Keys Audit job is flaky #39188

Closed

Activity

  1. Amar3tto commented on Jul 1, 2026

    @Amar3tto
    Collaborator
  2. added 2 commits that reference this issue on Jul 15, 2026
    e65186e
    0d1f25d
  3. added this to the 2.76.0 Release milestone on Jul 15, 2026
  4. added a commit that references this issue on Jul 17, 2026
    dca9622
  5. github-actions commented on Jul 20, 2026

    @github-actions
    ContributorAuthor

    Reopening since the workflow is still flaky

  6. added a commit that references this issue on Jul 23, 2026
    5e49938
  7. github-actions commented on Jul 28, 2026

    @github-actions
    ContributorAuthor

    Reopening since the workflow is still flaky

  8. Amar3tto commented on Jul 28, 2026

    @Amar3tto
    Collaborator

    @rwiggles could you take a look

  9. rwiggles commented on Jul 28, 2026

    @rwiggles
    Contributor

    Not sure how to proceed, just noted that this was already failing when I was working on my PR.

  10. self-assigned this
    on Jul 30, 2026
  11. added a commit that references this issue on Jul 30, 2026
    1738d11
  12. HansMarcus01 commented on Aug 3, 2026

    @HansMarcus01
    Contributor

    Hi everyone,

    I was reviewing recent GitHub Actions logs and noticed that the "Unmanaged Service Accounts Keys Audit" workflow keeps failing consistently.

    I wanted to mention that my pull request #39177 directly addresses this issue. During my initial investigation, I discovered that the beam_Infrastructure_PolicyEnforcer workflow already runs the same account_keys.py script and—crucially—has the SMTP environment variables correctly configured (the absence of these variables is what causes the standalone action to fail with a socket.gaierror).

    To eliminate code duplication and avoid deploying redundant runners that perform essentially the same task, PR #39177 completely removes the isolated unmanaged service accounts workflow and consolidates the audit task into the existing PolicyEnforcer.

    I apologize for the delay in following up; I hadn't realized this redundant action was still causing pipeline failures. Please let me know if any further changes are needed or if we can proceed to review and merge PR #39177 to resolve this!

    @pabloem

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

Metadata

Metadata

Assignees

Type

No type

Projects

No projects

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions