You have a critical database server that constantly backs its files up to the cloud, but its software environment is so finicky that if it encountered a critical failure it would take a long time to get it working again. How would you describe your recovery plan for that service