Table of Contents

• What You Will Build
• What You Will Need
• Create a Spring Boot Application
• Containerize the Application
• Deploy the Application to Kubernetes

This guide walks you through the process of deploying a Spring Boot application on Kubernetes. You can choose from many ways to do things with Spring Boot and Kubernetes. The intention of this guide is to get you going as quickly as possible, not to discuss all the alternatives or go into all the details of how you get to production.

Note

There are some interactive tutorials that complement and extend the content of this guide on Katacoda/springguides. If you follow those tutorials, all the code will be running in the cloud from your browser. Alternatively, you can create your own cluster, install all the tools you need locally, and copy-paste from the guides. Getting Started with Spring Boot on Kubernetes: The same material as this guide, but running in your browser. Install Kubernetes: A guide to installing Kubernetes locally by using Kind. You can use it to get setup on your laptop if you prefer to run the tutorials there. Kubernetes Probes with Spring Boot: A guide to liveness and readiness probes with Spring Boot.

What You Will Build

Kubernetes is an open-source system for automating the deployment, scaling, and management of containerized applications. It groups containers that make up an application into logical units for easy management and discovery. In this guide, we build and deploy a simple Spring boot application.

Note	You can also find a Getting Started Guide and a Topical Guide on Docker, which cover some of the background on building a container image.

What You Will Need

You will need a Linux or Linux-like command line. Command line examples in this guide work on Linux, a MacOS terminal with a shell, or WSL on Windows.

You will also need a Kubernetes cluster and the command line tool Kubectl. You can create a cluster locally by using Kind (on Docker) or Minikube. Alternatively, you can use a cloud provider, such as Google Cloud Platform, Amazon Web Services, or Microsoft Azure. Before proceeding further, verify that you can run kubectl commands from the shell. The following example uses kind:

$ kubectl cluster-info
Kubernetes master is running at https://127.0.0.1:46253
KubeDNS is running at https://127.0.0.1:46253/api/v1/namespaces/kube-system/services/kube-dns:dns/proxy

To further debug and diagnose cluster problems, use 'kubectl cluster-info dump'.

You should also run the following command:

$ kubectl get all
NAME                 TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)   AGE
service/kubernetes   ClusterIP   10.43.0.1    <none>        443/TCP   7m13s

Create a Spring Boot Application

First, we create a Spring Boot application. If you have one you prefer to use already in github, you could clone it in the terminal (git and java are installed already). Alternatively, you can create an application from scratch by using start.spring.io:

curl https://start.spring.io/starter.tgz -d dependencies=webflux,actuator | tar -xzvf -

You can then build the application:

./mvnw install

Note	It will take a couple of minutes the first time, but, once the dependencies are all cached, it will be fast.

Then you can see the result of the build. If the build was successful, you should see a JAR file similar to the following:

ls -l target/*.jar
-rw-r--r-- 1 root root 19463334 Nov 15 11:54 target/demo-0.0.1-SNAPSHOT.jar

The JAR is executable:

$ java -jar target/*.jar

The applicaiton has some built-in HTTP endpoints because of the actuator dependency we added when we downloaded the project. You should see output similar to the following in the logs on startup:

...
2019-11-15 12:12:35.333  INFO 13912 --- [           main] o.s.b.a.e.web.EndpointLinksResolver      : Exposing 2 endpoint(s) beneath base path '/actuator'
2019-11-15 12:12:36.448  INFO 13912 --- [           main] o.s.b.web.embedded.netty.NettyWebServer  : Netty started on port(s): 8080
...

Then you can curl the endpoints in another terminal:

$ curl localhost:8080/actuator | jq .
{
  "_links": {
    "self": {
      "href": "http://localhost:8080/actuator",
      "templated": false
    },
    "health-path": {
      "href": "http://localhost:8080/actuator/health/{*path}",
      "templated": true
    },
    "health": {
      "href": "http://localhost:8080/actuator/health",
      "templated": false
    },
    "info": {
      "href": "http://localhost:8080/actuator/info",
      "templated": false
    }
  }
}

To complete this step, press Ctrl+C to stop the application.

Containerize the Application

There are multiple options for containerizing a Spring Boot application. As long as you are already building a Spring Boot jar file, you only need to call the plugin directly. The following command uses Maven:

$ ./mvnw spring-boot:build-image

The following command uses Gradle:

$ ./gradlew bootBuildImage

You can run the container locally:

$ docker run -p 8080:8080 demo:0.0.1-SNAPSHOT

Then you can check that it works in another terminal:

$ curl localhost:8080/actuator/health

Finish by stopping the container.

You cannot push the image unless you authenticate with Dockerhub (docker login), but there is already an image there that should work. If you were authenticated, you could:

$ docker tag demo:0.0.1-SNAPSHOT springguides/demo
$ docker push springguides/demo

In real life, the image needs to be pushed to Dockerhub (or some other accessible repository) because Kubernetes pulls the image from inside its Kubelets (nodes), which are not usually connected to the local docker daemon. For the purposes of this scenario, you can omit the push and use the image that is already there.

Note	For testing, there are workarounds that make docker push work with an insecure local registry (for instance) but that is out of scope for this guide.

Deploy the Application to Kubernetes

Now you have a container that runs and exposes port 8080, so all you need to make Kubernetes run it is some YAML. To avoid having to look at or edit YAML, for now, you can ask kubectl to generate it for you. The only thing that might vary here is the --image name. If you deployed your container to your own repository, use its tag instead of this one:

$ kubectl create deployment demo --image=springguides/demo --dry-run -o=yaml > deployment.yaml
$ echo --- >> deployment.yaml
$ kubectl create service clusterip demo --tcp=8080:8080 --dry-run -o=yaml >> deployment.yaml

You can take the YAML generated above and edit it if you like, or you can apply it as is:

$ kubectl apply -f deployment.yaml
deployment.apps/demo created
service/demo created

Check that the application is running:

$ kubectl get all
NAME                             READY     STATUS      RESTARTS   AGE
pod/demo-658b7f4997-qfw9l        1/1       Running     0          146m

NAME                 TYPE        CLUSTER-IP      EXTERNAL-IP   PORT(S)    AGE
service/kubernetes   ClusterIP   10.43.0.1       <none>        443/TCP    2d18h
service/demo         ClusterIP   10.43.138.213   <none>        8080/TCP   21h

NAME                   READY     UP-TO-DATE   AVAILABLE   AGE
deployment.apps/demo   1/1       1            1           21h

NAME                              DESIRED   CURRENT   READY     AGE
replicaset.apps/demo-658b7f4997   1         1         1         21h
d

Tip	Repeat kubectl get all until the demo pod shows its status as Running.

Now you need to be able to connect to the application, which you have exposed as a Service in Kubernetes. One way to do that, which works great at development time, is to create an SSH tunnel:

$ kubectl port-forward svc/demo 8080:8080

Then you can verify that the app is running in another terminal:

$ curl localhost:8080/actuator/health
{"status":"UP"}