Connected infrastructure

Rehearse a small recovery

Editorial edition · 16 November 2026

The idea

A recovery document is a hypothesis until someone follows it. A bounded drill checks whether people, access and procedures work together under realistic conditions. The goal is to discover missing steps before an outage makes every minute expensive.

Design choice

Choose a safe test environment and a clearly limited failure. State the expected recovery result, time budget and stop condition before starting. Have someone other than the author follow the procedure, recording unclear instructions and unexpected dependencies.

Worked example

A team rebuilds a staging service from its documented package and restores a sample dataset. The drill discovers that a certificate renewal account was missing from the access list. The procedure is corrected and that step is repeated.

Before you ship

  • Define a safe stop condition.
  • Measure the observed recovery time.
  • Update the procedure and retest the failed step.

Examples are illustrative. Adapt the checks to your own environment.

Further reading

HTTP Semantics, RFC 9110

Search the library