Make dependencies visible.
Document which services rely on each other. Include the less visible parts: name resolution, storage, certificate renewal and scheduled maintenance. Keep this map short enough to stay current.
INFRASTRUCTURE
Simple to explain.
Ready to recover.
Our notes focus on practical foundations: visible dependencies, deliberate boundaries and recovery paths that have been tested.
Keep responsibilities clear.
Give each service a defined purpose and each account the access it needs. Clear boundaries make changes easier to review and problems easier to isolate.
Practice the return path.
Record how to restore a service, which configuration it needs and how to confirm that it is working. Exercise this path in a controlled setting before relying on it.
Change with a record.
Capture the reason for a change, the checks that matter and the way back. A concise record is more useful than a long history without clear decisions.