The 3-2-1 backup rule: simple enough to remember, strong enough to matter

Three, two, one

The classic 3-2-1 model keeps three copies of important data, stores them across two types of media, and places one copy offsite. It deliberately avoids betting every recovery path on the same disk, computer, room, or provider.

The working copy counts as one. A local backup on separate storage provides fast recovery. An encrypted offsite copy protects against theft, fire, and failures that affect the whole location. Cloud storage can fill the offsite role, but synchronization alone is not always a backup: destructive changes may synchronize too.

A successful command is not proof of recovery

A backup job can finish while producing an incomplete or corrupt archive. Verification should include a checksum or integrity test and occasional restoration into a safe, separate location. For databases and game worlds, application consistency matters too; copying files in the middle of writes may preserve the wrong state.

Automated retention also needs guardrails. Expiring old archives saves space, but the cleanup must never delete the newest known-good backup merely because a newer attempt failed.

A small recovery routine

List the data you could not recreate, assign each item a recovery frequency, and document one restoration procedure. Schedule the jobs, alert on failures, and perform a sample restore. The test often reveals missing encryption keys, forgotten credentials, or files that were never included.

• Back up configuration and identity data, not only the largest content folder.
• Protect backup credentials separately from the system being backed up.
• Record when the last restoration test passed.