communication webhook_reliability ai_generated true

GitHub webhook delivers same event multiple times because endpoint is slow or returns non-2xx

ID: communication/github-webhook-delivery-timeout-retry

Also available as: JSON · Markdown
85%Fix Rate
88%Confidence
3Evidence
2023-01-01First Seen

Version Compatibility

VersionStatusIntroducedDeprecatedNotes
any active — — —

Root Cause

GitHub webhooks timeout after 10 seconds and retry failed deliveries. There's no configurable retry policy. Slow endpoints receive duplicate deliveries. The X-GitHub-Delivery header is unique per delivery attempt, not per event.

generic

Workarounds

  1. 95% success Return 200 immediately and process webhook asynchronously 
    Enqueue the raw webhook payload for background processing. Return 200 within 1 second.
  2. 90% success Use event ID from payload (not delivery header) for idempotency 
    Dedup key: f'{event_type}:{payload["action"]}:{payload["id"]}'  # unique per logical event
  3. 82% success Check webhook delivery logs in GitHub repo settings for debugging 
    Settings -> Webhooks -> Recent Deliveries shows each attempt with request/response details

Dead Ends

Common approaches that don't work:

  1. Process webhook synchronously and return 200 after completion 88% fail

    If processing takes >10s, GitHub considers the delivery failed and retries. You get duplicate processing of the same event.

  2. Use X-GitHub-Delivery header as idempotency key 82% fail

    X-GitHub-Delivery is unique per delivery ATTEMPT, not per event. Retries have different delivery IDs. Use the event payload's action+id for dedup.