Backup and Recovery · PostgreSQL
Backup, Recovery, and the Calm Restore
A focused two-week intensive on backup verification, point-in-time recovery, and the operational habits that make a 03:00 restore feel boring rather than brave.
About this cohort
Backups are the part of the job nobody notices until they fail. This intensive is built around the principle that an untested backup is an opinion, not a backup. You will rehearse point-in-time recovery on PostgreSQL with WAL archiving, validate a full and differential strategy on SQL Server, and run a clean restore from a logical export on MySQL. Every lab finishes with a written verification report — short, honest, and useful to read later. We close the cohort with a paired restore drill where one participant operates and another observes, then they swap.
Inclusions
- Three engine-specific recovery rehearsals in two weeks
- WAL archiving setup and PITR validation lab on PostgreSQL
- Paired restore drill with peer observation and notes
- Verification report template used by working operations teams
- Reading on backup retention, regulatory windows, and audit trails
By the end you can
- 01 Perform point-in-time recovery on PostgreSQL with confidence
- 02 Validate a backup strategy against a defined recovery point objective
- 03 Write a verification report that another operator will trust at handover
Programme lead
Park Jihoon
Database Lab Engineer who maintains the cohort sandboxes. He breaks our clusters on purpose so participants can practice repairing them.
Common questions
From the cohort
-
Two weeks well spent. The verification report template is now part of our team handbook.
Hye-rinHye-rin
Reliability engineer 4.7/5
-
Calm tone throughout, no bravado. Restore drills do feel boring afterwards, which is the point.
Operator at a healthcare providerOperator at a healthcare provider