Decide where the work belongs
Map trust and backlog
For each endpoint, record failed events, environment, receiver revision, secret version label, pending state, and any effect. Separate delayed work from evidence that an unverifiable event was processed. Keep receiver revision beside backlog evidence.
Compare safe recovery
Assess updating receiver configuration, using a documented overlap, pausing delivery, or reviewing a bounded replay. Record authority, secret-handling, and deduplication needs. Do not print or copy signing secrets to speed diagnosis.
Set rotation order
Queue any processed unverifiable event and shared receiver lag first, then blocked essential delivery and isolated configuration errors. Set owner, review time, version evidence, and proof. Keep old-secret expiry claims tied to observed policy. Document the fallback for review.
What to carry forward
Return a rotation queue with trust consequence, endpoint reach, version evidence, safe containment, owner, and replay proof. Escalate unverifiable processing. Keep rank provisional when overlap or receiver deployment state is unknown. Keep strict verification ahead of backlog reduction.
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