fix(server): bound provider command lane stalls - #124
Merged
Conversation
A second recurrence showed that session restart, not only session stop, can hold ProviderCommandReactor's single DrainableWorker indefinitely. Run each provider command in a detached fiber, join it for at most 30 seconds, then release the global lane and request cleanup in the background. Keep upstream divergence to one marked import and one marked call site; the implementation and regression test are fork-owned additive files. Co-Authored-By: Codex <noreply@openai.com>
github-actions
Bot
added
size:M
vouch:trusted
This was referenced Aug 20, 2026
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
1 participant
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
A provider session restart reproduced the same global prompt-queue outage through a second path: any lifecycle command that never returns can hold the single ordered provider-command worker and prevent later turn starts from waking the durable coordinator.
Keep the upstream ordered worker unchanged, but run each command through a 30-second bounded join. A timed-out command releases the lane immediately and receives a detached interruption request, including when its provider finalizer is uninterruptible. The upstream-owned reactor gains only one marked import and one marked call site; the implementation and regression test are additive fork-owned files.
Validation:
Implemented with GPT-5.6 Sol in Codex/T3 Code.
Need help on this PR? Tag
@codesmith-botwith what you need. Autofix is disabled.