Skip to content

Stop leftover process-group members after a harness exits (#391) - #393

Merged
DanielNetzer merged 1 commit into
mainfrom
forgecell/issue-391-95a55efba2702a8465e3101e
Nov 6, 2026
Merged

DanielNetzer merged 1 commit into
mainfrom
forgecell/issue-391-95a55efba2702a8465e3101e

Conversation

@DanielNetzer

Copy link
Copy Markdown
Owner

Implements issue #391.

Molecule: mol-391-6d567eb4f574ec95d82712c9
Formula: forgecell-github-issue
Base: 2a771de3bf09e07b900ef3467248ce5af9b7b379
Delivery: 2d755b5f10f140ba761cfea01656f86f3e3ce8e3

Awaiting exact-commit CI and human review. Harness-reported checks are not independent CI evidence. No merge, deployment or issue closure is authorized.

@DanielNetzer

Copy link
Copy Markdown
Owner Author

Review notes (LOW, not blocking):

  • TestLeftoverHoldingPipeIsReconciledAfterWaitDelay always pays the 1 s WaitDelay plus SIGTERM; that's fine for one test, but don't copy the pattern widely.
  • The helper process ignores errors from os.WriteFile/os.Rename. If they fail, the tests fail later with a less clear readPid error.
  • Successful runs that leave leftovers now take up to 5 s longer to reconcile (and 1 s more if a leftover holds stdout/stderr). That's expected, and reconcileNote records it.

Protected regression: TestLiveManagedDescendantMakesResultUncertain failed by design (Decision 1 in #391 inverts it); accepted with --accept-protected-failures. All 12 checks passed.

@DanielNetzer
DanielNetzer marked this pull request as ready for review November 6, 2026 16:27
@DanielNetzer
DanielNetzer merged commit be7ece6 into main Nov 6, 2026
2 checks passed
@DanielNetzer
DanielNetzer deleted the forgecell/issue-391-95a55efba2702a8465e3101e branch November 6, 2026 16:48
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.

1 participant