Sandbox Handbook·2 min read

Scenario 2: prod-payments-ledger-db — Primary Failover, CAB Gate

A high-blast-radius failover that shows the platform's rule: a step that big always requires a change ticket.

The primary payments ledger database degraded, blocking writes for three dependent services: payment-settlement-svc, ledger-reconciliation-svc, and fraud-scoring-svc.

Steps to test

  1. Open INC0000002.
  2. Steps 4–7: db_replica_lag_seconds climbs past the 30-second threshold.
  3. Steps 8–9: blast radius of 3 downstream services.
  4. Step 12: approved (blast-radius reason).
  5. Step 13: the remediation step is flagged high blast radius and requires a change ticket, linked to CAB-approved change CHG0000027.
  6. Step 14: executed live, verification RESOLVED.
Outcome: confirms the platform's rule that a high-blast-radius step always requires a change ticket.
Back to Sandbox Handbook overview
← PreviousScenario 1: checkout-api — Connection Pool ExhaustionNext →Scenario 3: ehr-patient-records-db — Awaiting Review