Spring PetClinic sample application for Kubernetes

The famous Spring PetClinic sample application is now available as a Kubernetes native application leveraging microservices.

All dependencies / code which are not required to run this application with Kubernetes have been removed: for example you don't need to use service discovery with Spring Cloud Netflix since the Kubernetes platform already provides one such implementation.

Moreover, container images for the Spring PetClinic are now built using Cloud Native Buildpacks. You will not find any Dockerfile in this repository: the pack CLI is used to build secure and optimized container images for you.

Building this application

You need a JDK 8+ to build this application:

$ ./mvnw clean package

Pre-built container images are available, so that you can start deploying this app to your favorite Kubernetes cluster. In case you'd like to build your own images, please follow these instructions.

Read this guide to deploy the pack CLI to your workstation.

Many buildpack implementations are available: for best results, use Paketo buildpacks:

$ pack set-default-builder gcr.io/paketo-buildpacks/builder:base

You're ready to build container images with no Dockerfile!

Use the provided Makefile to build container images:

$ make all DOCKER_PREFIX=myrepo

Running this application locally

There is no need to run an Eureka server or anything else: this application is ready to run on your workstation.

Start the gateway:

$ java -jar spring-petclinic-api-gateway/target/spring-petclinic-api-gateway-VERSION.jar

Start the customers service:

java -jar spring-petclinic-customers-service/target/spring-petclinic-customers-service-VERSION.jar

Start the vets service:

$ java -jar spring-petclinic-vets-service/target/spring-petclinic-vets-service-VERSION.jar

Start the visits service:

$ java -jar spring-petclinic-visits-service/target/spring-petclinic-visits-service-VERSION.jar

Using your browser, go to http://localhost:8080 to access the application.

Enabling Wavefront when running locally

This application includes Wavefront integration. Enable profile wavefront to get access to a free Wavefront trial, in order to get metrics / traces from the application.

Set the system property spring.profiles.active=wavefront to enable wavefront profile.

For example:

$ java -jar -Dspring.profiles.active=wavefront spring-petclinic-visits-service/target/spring-petclinic-visits-service-VERSION.jar

As the application starts, a link to your application dashboard will be displayed in the console output:

Your existing Wavefront account information has been restored from disk.

To share this account, make sure the following is added to your configuration:

	management.metrics.export.wavefront.api-token=6fe0383b-0338-4664-883e-2642824b968c
	management.metrics.export.wavefront.uri=https://wavefront.surf

Use this link to get access to metrics / traces.

Deploying this application to Kubernetes

This application relies on a MySQL database to persist data: you first need to deploy a MySQL instance.

Run these commands to create a MySQL database:

$ kubectl create ns spring-petclinic
$ helm upgrade pdb bitnami/mysql -n spring-petclinic -f k8s/services/mysql/values.yml --version 6.14.4 --install

If you want to enable Wavefront integration, you need to deploy a proxy first. Follow this guide to deploy a Wavefront proxy for Kubernetes.

You may want to reuse this Wavefront proxy configuration:

$ kubectl create ns wavefront
$ helm upgrade wavefront wavefront/wavefront -f k8s/services/wavefront/values.yml --set wavefront.url=https://vmware.wavefront.com --set wavefront.token=wavefront-api-token --set clusterName=k8s-cluster-name -n wavefront --install --version 1.2.6

Make sure you set the Wavefront API token and the Kubernetes cluster name.

Edit file k8s/wavefront/configmap.yml and set the Kubernetes cluster name:

data:
  # Don't forget to reuse the same K8s cluster name used by the Wavefront proxy.
  WAVEFRONT_APPLICATION_CLUSTER: dev01

Deploy this configuration file to your cluster:

$ kubectl apply -f k8s/wavefront

It's time to bind the application to your Wavefront space. Create a Secret by setting the Wavefront API token:

$ kubectl -n spring-petclinic create secret generic app-wavefront --from-literal=MANAGEMENT_METRICS_EXPORT_WAVEFRONT_API-TOKEN=wavefront-api-token

You're almost there!

Deploy the application to your cluster:

$ kubectl apply -f k8s

The application is not publicly accessible: you need to create a Kubernetes service. Depending on your cluster configuration, you may have to use an Ingress route or a LoadBalancer to expose your application.

Run this command to use an ingress route (edit file k8s/ingress/ingress.yml first to set the route):

$ kubectl apply -f k8s/ingress

Run this command to use a Kubernetes managed load balancer:

$ kubectl apply -f k8s/loadbalancer

Congratulations: you're done!