Repository navigation
[codex] Reconcile stale running jobs when pid exits - #225
Closed
LawrenceChiu95 wants to merge 1 commit into
Closed
LawrenceChiu95 wants to merge 1 commit into
LawrenceChiu95 wants to merge 1 commit into
Conversation
LawrenceChiu95
force-pushed
the
codex/reconcile-stale-running-jobs
branch
from
April 19, 2026 09:34
dda9158 to
0d6000d
Compare
LawrenceChiu95
marked this pull request as ready for review
April 19, 2026 09:34
xoonjaeho
added a commit
to xoonjaeho/codex-plugin-cc
that referenced
this pull request
Aug 1, 2026
…openai#379) A tracked job's completion is written by the launcher process that owns it. If that launcher dies first -- crash, killed session, machine sleep -- the record is frozen at running/queued with a stale pid forever: `status` keeps listing it, `result` refuses it, and the rows accumulate. Adopted upstream PR openai#379 verbatim. It reconciles at `listJobs`, the single read chokepoint every consumer goes through, so status/result/cancel are all fixed at once. Preferred over the competing openai#225 because its liveness probe is fail-closed: only an explicit ESRCH marks a process dead, while openai#225 defaults to "dead" on any unexpected errno. Wrongly reporting a live job as failed is far worse than leaving a zombie -- codex turns routinely sit silent for many minutes. Added the test that openai#379 was missing: a probe failing with EPERM (or any non-ESRCH error) must leave the job running. Mutation-checked -- flipping that default to "dead" passed the whole upstream suite before this test existed. Windows suite: 96 tests / 84 pass / 12 fail, against a 91 / 79 / 12 baseline. No new failures.
dkundel-openai
added
the
duplicate
ChatGPT Codex Connector
Collaborator
|
Duplicate of #425. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
runningjobs when the tracked pid no longer existsstate.json) and per-job file (jobs/<id>.json)failedRoot Cause
/codex:statustrusted persisted state forrunningjobs without validating whether the worker process was still alive. If a worker died unexpectedly, the job could remainrunningindefinitely and appear stuck.Changes
plugins/codex/scripts/lib/state.mjsprocess.kill(pid, 0))listJobs()for stalerunningjobs with dead pidsfailedwitherrorMessage,completedAt, andpid: nulltests/state.test.mjslistJobs auto-reconciles stale running jobs when pid is no longer aliveValidation
node --test tests/state.test.mjsnode --test tests/runtime.test.mjs --test-name-pattern "status shows phases|status preserves adversarial review kind labels|status --wait times out cleanly|cancel without a job id ignores active jobs from other Claude sessions|cancel with a job id can still target an active job from another Claude session|stop hook logs running tasks to stderr without blocking when the review gate is disabled"