Deploying Kotlin Application on Kubernetes: Basic Guide

Deploying Kotlin on Kubernetes: A Practical Basic Guide

Kubernetes (K8s) has become the standard for orchestrating containerized applications. If you develop server-side applications in Kotlin (using Ktor, Spring Boot, or http4k), sooner or later you will face the question of deploying them in a cluster. In this guide, we will walk through the process from packaging a Kotlin application into a Docker image to running it in Kubernetes using ConfigMap and Ingress. All examples are working and ready for adaptation.

1. Preparing the Kotlin Application and Docker Image

For the example, let's take a simple application on Ktor (can be replaced with any other framework). Our application will listen on port 8080 and return a JSON with the version.

// src/main/kotlin/com/example/Application.ktimport io.ktor.server.engine.*import io.ktor.server.netty.*import io.ktor.server.routing.*import io.ktor.server.response.*

fun main() {    embeddedServer(Netty, port = 8080) {        routing {            get("/") {                call.respond(mapOf("app" to "kotlin-on-k8s", "version" to "1.0.0"))            }        }    }.start(wait = true)}

Let's build the application into a fat JAR using Gradle (the shadow plugin). After the build, create a Dockerfile:

# DockerfileFROM openjdk:17-jdk-slimWORKDIR /appCOPY build/libs/app-all.jar app.jarEXPOSE 8080CMD ["java", "-jar", "app.jar"]

Build the image and push it to a registry (e.g., Docker Hub):

docker build -t youruser/kotlin-app:1.0.0 .docker push youruser/kotlin-app:1.0.0

2. Creating Kubernetes Manifests

Now let's describe the resources for deployment. Create three files: deployment.yaml, service.yaml, and configmap.yaml.

2.1. ConfigMap for Configuration

Let's externalize application settings (e.g., logging level or connection string) into a ConfigMap:

# configmap.yamlapiVersion: v1kind: ConfigMapmetadata:  name: app-configdata:  LOG_LEVEL: "DEBUG"  APP_MODE: "production"

2.2. Deployment

In the Deployment manifest, specify the image, port, replicas, and inject variables from the ConfigMap:

# deployment.yamlapiVersion: apps/v1kind: Deploymentmetadata:  name: kotlin-appspec:  replicas: 2  selector:    matchLabels:      app: kotlin-app  template:    metadata:      labels:        app: kotlin-app    spec:      containers:        - name: app          image: youruser/kotlin-app:1.0.0          ports:            - containerPort: 8080          envFrom:            - configMapRef:                name: app-config          livenessProbe:            httpGet:              path: /              port: 8080            initialDelaySeconds: 10            periodSeconds: 5

2.3. Service

The Service will balance traffic between pods:

# service.yamlapiVersion: v1kind: Servicemetadata:  name: kotlin-app-servicespec:  selector:    app: kotlin-app  ports:    - protocol: TCP      port: 80      targetPort: 8080  type: ClusterIP

3. Deploying to the Cluster and Configuring Ingress

Apply the manifests:

kubectl apply -f configmap.yamlkubectl apply -f deployment.yamlkubectl apply -f service.yaml

Check that the pods are running:

kubectl get podskubectl logs <pod-name>

Now make the application accessible from outside. Create an Ingress (assuming an Ingress Controller, e.g., NGINX, is installed in the cluster):

# ingress.yamlapiVersion: networking.k8s.io/v1kind: Ingressmetadata:  name: kotlin-app-ingressspec:  rules:    - host: kotlin.example.com      http:        paths: