Skip to content

simple_scheduler_state_manager: count worker disconnects toward max_job_retries#2352

Open
amankrx wants to merge 1 commit into
TraceMachina:mainfrom
amankrx:fix/worker-disconnect-retries
Open

simple_scheduler_state_manager: count worker disconnects toward max_job_retries#2352
amankrx wants to merge 1 commit into
TraceMachina:mainfrom
amankrx:fix/worker-disconnect-retries

Conversation

@amankrx
Copy link
Copy Markdown
Collaborator

@amankrx amankrx commented May 20, 2026

Description

Avoid worker disconnects that keeps on running the build forever.

Fixes # (issue)

Type of change

Please delete options that aren't relevant.

  • Bug fix (non-breaking change which fixes an issue)

How Has This Been Tested?

Added tests and tested it in a working build environment.

Please also list any relevant details for your test configuration

Checklist

  • Updated documentation if needed
  • Tests added/amended
  • bazel test //... passes locally
  • PR is contained in a single commit, using git amend see some docs

This change is Reviewable

Copy link
Copy Markdown
Collaborator

@MarcusSorealheis MarcusSorealheis left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

..ExecutionMetadata::default()
},
error: Some(make_err!(
Code::Aborted,
Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I was re-reading this one. This is the wrong error code. Code::Aborted is for concurrency/transaction conflicts. For an action that failed because the worker repeatedly died running it, the right code is INTERNAL (the safe generic).

If we could verify what the failure mode was, like OOM or something else, we could get more specific like RESOURCE_EXHAUSTED.

Copy link
Copy Markdown
Collaborator

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

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

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

2 participants