Establish what is happening
Capture the retained record
Record event and delivery identifiers, endpoint, environment, failure class, first and last attempt, and current disposition. Do not copy secrets or unredacted payloads. Note whether the record can be matched to a receipt or side effect.
Compare review fields
Check identity, timing, response, attempt history, redaction, resource, and owner against a known failed fixture. A visible error without an event ID still leaves replay ambiguous. Keep provider delivery history separate from local dead-letter state.
Bound recovery candidates
Group records by failure class, age, endpoint, event identity, receipt state, and available replay path. Separate recoverable validation or transport failures from completed or unsafe effects. Mark missing fields as evidence gaps.
What to carry forward
Triage ends with a findable failed event, failure history, redaction boundary, current effect state, and recovery gap. Keep the item quarantined while identity or side-effect status is unknown. Route retention and replay decisions separately. Keep replay separate from ordinary processing.
Technical background: Twilio documentation.
Keep the decision with the work.
Use a Work Item in Aglet to record the problem, the evidence you have, and the next decision. Add an owner and priority, then keep updates in the discussion so the next person can follow the reasoning.
Create an account See the product workflow