Aglet

Triage webhook acknowledgement before processing

A receiver can return success before it has safely recorded or queued an event. Choose one delivery and map arrival, response, process lifetime, validation or storage, and downstream effect. Separate durable handoff, in-memory buffering, early response, and sender retry policy before calling the event lost.

Establish what is happening

  1. Freeze one acknowledgement

    Record endpoint, environment, event identity, arrival time, response status and time, receiver revision, process state, durable receipt, and downstream result. Note whether an interruption occurred before or after any durable handoff.

  2. Mark response ordering

    Compare the response boundary with validation, durable receipt, queue or worker handoff, processing, and effect. Distinguish a slow event that remains recoverable from a success response with no local recovery handle. Keep sender retry observations separate.

  3. Bound possible loss

    Group cases by receiver revision, response timing, interruption point, event identity, retry result, and known effect. Separate one missing receipt from a systematic early response. Preserve unresolved events for reconciliation rather than replaying them blindly.

What to carry forward

Triage is complete when one delivery has a reproducible ordering, first unsafe boundary, affected scope, and explicit unknowns. Hold blanket response changes. Route receipt durability, worker handoff, interruption, and sender retry questions separately with owners.

Technical background: Stripe 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