liveBook · Manning

Canary deployment

As explained in Microservices Patterns, Second Edition

A canary release runs a new software version alongside the current version and initially routes only a small share of traffic to it.

A canary deployment reduces the impact of production bugs by separating deployment—running an updated component in production—from release—routing traffic to it. The updated component runs alongside the current version and initially receives only a small share of traffic. section 5.5.2 section 5.5.2

A progressive rollout proceeds by:

  • Increasing the share of traffic routed to the updated component until it handles all traffic. section 5.5.2
  • Using telemetry from the updated component to detect problems; roll back if it indicates failure. section 5.5.2

Canaries help limit user disruption, but some strategies start the application twice, slowing deployment. Evaluate the trade-off against the risk of exposing users to a faulty update. section 5.5.2 section 6.1.2

How Microservices Patterns, Second Edition covers it

How deep it goes
Defines traffic ramp-up and telemetry-triggered rollback, while deferring deeper incremental-release details.
Examples
Uses an updated component running beside its current version, initially receiving a small share of traffic.
What you can do after
Operate a progressive rollout by increasing traffic exposure and rolling back when telemetry indicates a problem.
How it connects
Connects progressive deployment with deployment and release decoupling, telemetry, observability, and rollback.

Also explained in

Compare how 11 books explain canary deployment →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Microservices Patterns, Second Edition. The section links point to the parts of the book it draws on.