The following tutorials explain some of the basic concepts in iter8 and provide examples of basic building blocks to get started with your own experiments.
This tutorial shows you how iter8 can be used to perform canary releases by gradually shifting traffic to a canary version of a microservice. This tutorial has 5 parts, which are supposed to be tried in order. Here you will learn: how to perform a canary rollout with iter8; how to set different success criteria for iter8 to analyze canary releases and determine success or failure; how to have iter8 immediately stop an experiment as soon as a criterion is not met; how to use your own custom metrics in success criteria for canary analyses; and how iter8 can be used for canary releases of both internal and user-facing services.
In iter8 the versions of a service being compared can be specified using deployment names or using service names. Other tutorials showed how to specify different versions using Kubernetes deployment names. In this tutorial, we learn how to do a canary rollout of an application when different versions are indicated by different Kubernetes service names. In this tutorial, we again consider the user facing service productpage of the bookinfo application and we learn how to create an iter8 Experiment that specifies the baseline and candidate versions using Kubernetes services.