Skip to content

fix: raise agent apply and fallback timeouts for large fix batches - #170

Open
cwbcheng wants to merge 6 commits into
jeremymcs:mainfrom
cwbcheng:codex/raise-agent-timeouts
Open

fix: raise agent apply and fallback timeouts for large fix batches#170
cwbcheng wants to merge 6 commits into
jeremymcs:mainfrom
cwbcheng:codex/raise-agent-timeouts

Conversation

@cwbcheng

@cwbcheng cwbcheng commented Aug 7, 2026

Copy link
Copy Markdown

Problem

PR #438 had 9 approved review-comment fixes batched into a single agent run. The codex apply timed out (exit 124) after the default 30-minute limit, and the code-owner fallback (also 30 minutes) timed out the same way. All 9 fixes were left uncommitted and un-pushed; GitHub kept showing the review threads unresolved.

The root cause is capacity mismatch: a single codex run processing a large batch (many files, dotnet build/tests) exceeds 30 minutes.

Fix

  • Raise the default applyFixesWithAgent timeout from 30 min (1800000) to 90 min (5400000) in server/agentRunner.ts.
  • Raise CODE_OWNER_FALLBACK_TIMEOUT_MS from 30 min to 90 min in server/babysitter.ts.

This gives large fix batches (and their fallback retries) enough headroom to finish.

Verification

  • After applying the fix locally, PR #438's 7 failed fixes were retried in batches (3 + 4) and all completed: 19 feedback items resolved (fixes committed and pushed, review threads replied and resolved), 29 rejected (bot summary / non-actionable), 0 failed.
  • npm run check passes. server/babysitter.test.ts passes (110/110). The only failures in server/agentRunner.test.ts are pre-existing sandbox EACCES-vs-ENOENT spawn issues unrelated to this change (they fail identically on main).

Rejected review-thread feedback previously fell through the GitHub
follow-up path silently: needsGitHubFollowUp only considered accepted
items, so a rejected review comment was never replied to and its
conversation was never resolved, contradicting the agent prompt that
says to reply with the rejection reason and resolve the conversation.

Now rejected review threads (excluding PatchDeck's own status/audit-trail
replies) are collected as follow-up tasks, get a 'Rejected - no code
change made' reply with the reason, and the conversation is resolved.
Audit-trail verification skips the code-change audit for these items and
only requires thread resolution, and the final status update keeps the
rejection decision while advancing the thread-resolved flag.
Localize the comments PatchDeck posts to GitHub:
- Agent prompt now requires Simplified Chinese for every comment,
  thread reply, and summary.
- Follow-up replies use Chinese headlines (已拒绝/已在提交/已在最新更新)
  with 原因 for rejected threads, and 正在回复 for quoted references.
- Status messages (accepted/in-progress/needs-attention/verifying/resolved)
  are localized.
- APP_STATUS_COMMENT_PATTERN now recognizes both English and Chinese
  status markers so recovery of existing status replies keeps working.

Tests updated to assert the Chinese wording.
The internal-reply exclusion in needsGitHubFollowUp only matched
'PatchDeck status comment' and 'Automation audit trail follow-up', but
classifyNonActionableAppFeedback also produces 'PatchDeck audit trail
comment' and 'PatchDeck agent command comment'. A rejected item carrying
one of those reasons was treated as real reviewer feedback, which made
PatchDeck reply to its own follow-up comment, producing a new audit
trail, which was then replied to again - an infinite loop.

Widen the exclusion to all PatchDeck-authored marker reasons and add a
regression test that a rejected audit-trail comment never triggers a
follow-up reply.
The code-owner fallback agent is asked to commit and push its own
changes, but agents sometimes finish without doing so (e.g. a truncated
model response). Previously PatchDeck marked the fallback run completed
and removed the worktree, discarding the agent's edits and leaving the
PR branch and review threads untouched.

After a successful fallback agent run, PatchDeck now:
- commits any uncommitted agent edits in the worktree,
- pushes the local head to the PR head branch,
- verifies the remote head matches the local head.

Adds a regression test that a fallback agent leaving uncommitted edits
results in git add/commit/push to the PR branch.
The DeepSeek-backed codex agent frequently times out on long review
comments: evaluation had a 3-minute cap and apply had a 15-minute cap,
which was not enough for multi-file fixes (e.g. 21 generated fixtures).
When apply timed out mid-run, the uncommitted worktree edits were
discarded.

Raise evaluation to 7 minutes and the default apply timeout to 30
minutes (matching the code-owner fallback timeout) so complex review
threads can be evaluated and fixed to completion.
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