Add OpenAI retry with exponential backoff for 429 and 5xx responses #374

New issue

Closed

opened 2026-04-10 13:17:29 -04:00 by pook · 1 comment

pook commented

2026-04-10 13:17:29 -04:00

Owner

Create a retryWithBackoff wrapper for the OpenAI chat completion call in document generation. On 429 (rate limit) or 5xx (server error) responses, retry up to 3 times with delays of 1s, 2s, 4s. Respect Retry-After header if present. On final failure, return structured error to client. Do not retry 4xx client errors (except 429). Different from #366 (response timeout) — this handles transient failures, not hangs. Acceptance: transient OpenAI errors recover automatically, logs show retry attempts, final failures return clean errors.

Generated by CEO Planner (priority: 3)

pook added the

agent-task

label

2026-04-10 13:17:29 -04:00

pook commented

2026-04-10 14:47:41 -04:00

Author

Owner

Bulk-closed 2026-04-10 during pipeline triage.

Context: CEO agent had created 100 open agent-task issues against compliancebot, largely duplicates of each other and of the 50 currently-open PRs. Root cause traced to a git-push race in agent-worker executor (dispatch jobs collided on branch agent/dispatch/* because jobId prefix truncated to literal "dispatch"). Fix deployed: runId is now threaded from Paperclip shim through /dispatch → TaskJob → executor, and branches are keyed on a 12-char unique run key.

What to do next:

Triage the 50 open PRs at https://192.168.183.110:3000/pook/compliancebot/pulls — many are ready to merge
CEO should halt new task creation until open PRs drop below 10
Surviving kept issues: #313, #314, #315, #341, #342, #350, #351, #352 (PR review/merge tasks)

This issue was superseded, not abandoned. Reopen if still relevant after PR triage.

Bulk-closed 2026-04-10 during pipeline triage. **Context:** CEO agent had created 100 open agent-task issues against compliancebot, largely duplicates of each other and of the 50 currently-open PRs. Root cause traced to a git-push race in agent-worker executor (dispatch jobs collided on branch `agent/dispatch/*` because jobId prefix truncated to literal "dispatch"). Fix deployed: runId is now threaded from Paperclip shim through /dispatch → TaskJob → executor, and branches are keyed on a 12-char unique run key. **What to do next:** 1. Triage the 50 open PRs at https://192.168.183.110:3000/pook/compliancebot/pulls — many are ready to merge 2. CEO should halt new task creation until open PRs drop below 10 3. Surviving kept issues: #313, #314, #315, #341, #342, #350, #351, #352 (PR review/merge tasks) This issue was superseded, not abandoned. Reopen if still relevant after PR triage.

pook closed this issue

2026-04-10 14:47:42 -04:00

No labels

agent-task

No milestone

No project

No assignees

1 participant

Notifications

Due date

The due date is invalid or out of range. Please use the format "yyyy-mm-dd".

No due date set.

Dependencies

No dependencies set.

Reference

pook/compliancebot#374

No description provided.

Rows
Columns