A provider may stop or restart a virtual machine during maintenance.
The service must continue operating.
Which design is appropriate?
Design for provider maintenance and restarts with redundancy, persistence, monitoring, and recovery procedures.
Detailed explanation
Incorrect. Assuming provider maintenance never occurs ignores an expected operational event.
Correct. Cloud services still require availability planning for maintenance and failures.
Incorrect. Temporary disks may lose data during restart or destruction.
Incorrect. Stopping monitoring and backups reduces recoverability.
Try it yourself
An example you can run in a temporary verification environment.
printf '%s\n' 'design for provider maintenance and instance restart'Expected result
design for provider maintenance and instance restartKey points
- Maintenance
- Restart
- Redundancy
- Recovery
Notes
- Environment: クラウド可用性設計の概念確認
- Command output formatting can vary slightly by distribution or tool version.
- Run the example in a temporary directory or process when possible.
Foundation review
Availability
Design so that a single instance failure does not stop the whole service.
Operations
Review maintenance notices, automated repair, and recovery objectives as part of operations planning.