Lab
Automated Backup and Restore Drill
Back up a sample service, check the archive, and practise restoring it somewhere clean.
- Linux
- Bash
- Backups
- Recovery
- Operations
Scenario
The point of this lab is to test the restore, not just the scheduled backup.
Challenge
If the only copy is on the same machine, the archive is never checked, or the restore steps are unclear, a backup may not help when something breaks.
Approach
I’d write a script that checks its prerequisites, creates a dated backup, records the result, and validates the archive. Old copies would only be removed according to a written retention period, and the test backup would be kept away from the sample service data.
Verification
I’d restore into a clean test location, compare the files and service data, and write down any missing steps. I’d only report a restore time after actually running the drill.