Back up what makes a small server unique
A compact recovery plan built around configuration, state, and a tested restore.
On a small server, most installed software can be downloaded again. The irreplaceable parts are usually configuration, application state, credentials, and the decisions needed to connect them. A useful backup plan starts with that distinction.
Inventory the important state
Keep service configuration, deployment notes, firewall rules, and persistent application data in the inventory. Record the software version and where it came from. For databases, use a backup method that produces a consistent snapshot rather than copying live files indiscriminately.
Keep a separate copy
A second directory on the same disk is convenient for undoing a mistake, but it is not protection against losing the server. Keep an encrypted copy on a separate system with restricted access and a retention policy.
Avoid granting the production machine permission to delete every historical backup. Separate credentials and append-only or versioned storage can reduce the impact of a compromised source.
Write the order of recovery
A short recovery note should identify the required operating system, packages, data locations, service accounts, and startup order. Include the verification steps, but store secret values in the protected backup rather than in a public guide.
- Prepare a clean host and install the recorded software versions.
- Restore configuration and data with the correct ownership.
- Check configuration before starting services.
- Verify the application through its normal client path.
Actually restore it
Test a restore on an isolated machine. Measure the time, inspect the result, and update the recovery note with anything missing. A backup that has never been restored is an assumption, not a demonstrated recovery path.