The article demonstrates how to automate a canary analysis and deployment for an application using Spinnaker, Prometheus, and HashiCorp Consul service mesh on Kubernetes. It shows how to deploy Consul and Prometheus to collect metrics, configure the service mesh, and set up Spinnaker for automated canary analysis. The example uses a simple application with two services, "ui" and "web", where all traffic from the "ui" service goes to the baseline version of the "web" service. A new version of the "web" service is released as a canary release, and Spinnaker analyzes the metrics from Consul's proxy metrics and generates a score based on the error rate. If the analysis passes, Spinnaker increases traffic to the canary version; if it fails, it rolls back to the baseline version. The pipeline also controls traffic splitting between the baseline and canary versions using Consul's service splitter. The example uses Grafana for visualization of request metrics and demonstrates how to configure Spinnaker canary analysis with various parameters, such as lifetime, delay, and interval.