Skip to content

tests: reduce background task timing flakiness - #37

Merged
dkundel-openai merged 1 commit into
openai:mainfrom
VOIDXAI:voidxai/issue-33-runtime-test-flakiness
Mar 31, 2026
Merged

dkundel-openai merged 1 commit into
openai:mainfrom
VOIDXAI:voidxai/issue-33-runtime-test-flakiness

Conversation

@VOIDXAI

@VOIDXAI VOIDXAI commented Mar 31, 2026

Copy link
Copy Markdown
Contributor

Summary

  • relax the background-task status wait window in runtime.test.mjs
  • allow a longer window for the brokered cancellation test to observe an in-flight task
  • widen the fake interruptible task window so cancel assertions race against a real running task, not normal completion

Why

Closes #33.

The background-task runtime tests are too timing-sensitive on slower machines. Normal startup variance can leave a job in running long enough to fail assertions even when the implementation is correct.

Validation

  • manually reproduced the background task completion flow and verified the job reaches completed
  • manually reproduced the brokered cancellation flow and verified turnInterrupted: true plus the recorded interrupt in the fake Codex state

@VOIDXAI
VOIDXAI requested a review from a team March 31, 2026 03:38
@dkundel-openai
dkundel-openai merged commit 8e403f9 into openai:main Mar 31, 2026
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.

Background task runtime tests are too timing-sensitive on slower machines

2 participants