Snapshot, backup, replication and archive: choose by recovery task

Ask a recovery question first

Saying “snapshots are enabled” does not explain how usable data returns after an incident. A snapshot records a point-in-time state; its usefulness depends on where it resides and whether that state is suitable for the application. A backup policy coordinates copy creation, retention and recovery. Replication transfers data to another endpoint, while an archive serves long-term retention and retrieval.

From incident to evidence

Independence and retention

Replication can also propagate an operational mistake. Copies on different devices may still share a risk if the same administrative account can delete them. Document access rights, management dependencies, retention and deletion rules separately. If immutable storage or isolation is required, confirm the actual product version and policy. A second chassis alone does not establish an independent recovery process.

A short policy example

Illustrative scenario: working files need user-level rollback, an operational database needs application-validated recovery, and retained records need searchable retrieval. Put these objectives on separate policy lines. Give each line a data owner, retention rule, acceptable period of data loss and agreed start and end points for recovery time.

At acceptance, go beyond a successful copy-creation log. Open a sample in the target environment, record validation by the authorised application owner, and feed any gaps back into the recovery plan.