Using Canary Release Pipelines to Achieve Continuous Testing

Sauce AI for Test Authoring: Move from intent to execution in min.|xBack to ResourcesBlogPosted June 20, 2018

Using Canary Release Pipelines to Achieve Continuous Testing

Want to lead your testing routine to include production? Canary-yellow freeing pipelines may be the solution you involve.

This article explain what canary-yellow release pipelines mean and how they can aid you test in production in order to achieve uninterrupted testing.

What Is a Canary-yellow Release Pipeline?

Image credit: Greg Sypolt

A canary release grapevine is a deployment tactic to prove in product to trim the risk of introducing a new application issue in product. The purpose of canary release pipelines and test is to get an idea of how an application will perform in production by gradually route a pocket-size percentage of user to the new version before making it uncommitted to the total consumer foundation.

A deployment strategy that includes canary liberation allow team to find problems sooner rather than later by shifting testing to the right—in early words, into production. This doesn ’ t mean that you shouldn ’ t also test sooner in the ontogeny pipeline, of trend. But by adding testing in production, you derive another lineament guard around your application. The new tactic of testing in product is an essential part of your testing strategy to deliver lineament apps to customer unceasingly.

Image credit: Greg Sypolt

Canary Release Pipeline and Testing

With a canary-yellow liberation grapevine in place, you can perform canary-yellow examination.

Canary testing focusing on sending a little part of production traffic to the new version environment in production (to some turn of user) and execute a pre-defined list of canary tests.

They typically include testing for the following:

SUSA automates exploratory testing with persona-driven behavior, catching bugs that scripted automation misses.

Infrastructure
API
Performance
Lightweight User Acceptance
Visual
Application availability

This proficiency reduces the endangerment of introduce new issues with uninterrupted testing methods and additional monitoring—by gradually routing a small subset of users to the new version of the application.

Image recognition: Greg Sypolt

Post-Canary Production Testing

The test train doesn & # x27; t stop after canary releases. Once a canary-yellow freeing has pass and you have completed canary tests, you should continue to perform testing, logging, and monitoring in production to expose and identify issue within the application, network, or base to help fix them before end users are affected. Here are a couple of testing method you may deal for canary post-release deployment in a live production environment:

Teeingis similar to TAP equivalence where we send traffic to multiple host. In this case, we send traffic to growing servers for debugging.
Chaos Testingis where you require to introduce a failure to guarantee your application, net, or substructure can recover from intentional disaster. Yes—we desire to break the covering or infrastructure deliberately to control it mechanically recovers (self-healing).

Conclusion

Traditionally, testing has be something that pass to the “ left ” of covering freeing. With canary freeing pipelines, however, you can test efficaciously in production while minimizing the risk assort with run tryout on production applications. While you should absolutely proceed testing pre-release, canary test are a useful strategy for helping to achieve the goal of uninterrupted testing, which provides the most racy, comprehensive test coverage.

Greg Sypolt (@ gregsypolt) is Director of Quality Engineering at Gannett | USA Today Network, a Fixate IO Contributor, and co-founder of Quality Element. He is creditworthy for test automation solutions, test coverage (from unit to end-to-end), and uninterrupted integration across all Gannett | USA Today Network ware, and has helped alter the testing approach from manual to machine-driven testing across respective merchandise at Gannett | USA Today Network. To determine improvements and testing gaps, he acquit a face-to-face audience survey process to interpret all product development and deployment process, testing strategies, tooling, and interactive in-house training programme.