by swinging traffic back 2. Database changes a. Create two DB – put one in read only mode, backup & restore to new environment, swing traffic b. Decouple DB changes from application changes i. Make

availability 8. Startups & shutdowns 9. Faults & errors 10. Circuit breaker trips 11. Delays 12. Backup success/failure d. USE TELEMETRY TO GUIDE PROBLEM-SOLVING i. Don’t practice “Mean Time until Declared

faster) iii. Production deployments (sixty times higher change success rate) iv. Mean time to restore service (168 times faster) 4. An Introduction to The DevOps Handbook xxi 5. PART I—THE THREE WAYS