Skip to content

l3montree-dev/kubernetes-developer-workshop

BranchesTags

Repository files navigation

Kubernetes for Developers Workshop 🚀

Caution

All material provides only an introduction to containerization & K8s. It is not intended for production use and does not cover all best practices. Please dig especially into security topics before deploying anything to production.

This workshop is intended for developers who want to learn how to deploy applications to Kubernetes.

Tip

Check if your local environment is setup correctly by running the following command:

git clone [email protected]:l3montree-dev/kubernetes-developer-workshop.git
cd kubernetes-developer-workshop
./check_env.sh

If the script notifies you of any missing dependencies, please install them before proceeding.

Table of Contents
  1. Prerequisites
  2. Color-Application Architecture Overview
  3. Section 1 Containerize the Application
  4. Section 2 Kubernetes Deployment
  5. Section 3: Helm Charts

Prerequisites

Color-Application Architecture Overview

flowchart TD
    A[React Frontend] -->|Fetches Color| B[Spring-Boot Color-API]
    B -->|Returns| A
    B-->|Logs Request| C[PostgreSQL]
Loading

Section 1 Containerize the Application

Part 1.1 Understanding Containers

  1. Explore the example Spring Boot application in the color-api directory. It provides an endpoint that returns a random hex color.

  2. Review the Dockerfile.

  3. Build the container image with the following command:

    docker build -t spring-boot-demo:latest color-api

  4. Inspect the image, which is essentially a tar archive:

    docker save spring-boot-demo:latest -o spring-boot-demo.tar

  5. Run the inspect_layers.sh script:

    • It extracts the tar archive into the extracted_image folder.
    • It extracts each layer tar archive in the extracted_image/blobs folder into the layers folder.
    • It merges the layers into a single file system in the merged_layers folder, resolving conflicts by replacing files as needed.

  6. Examine the merged_layers folder to see the file system of the container image.

Part 1.1.1 Understanding Multi-Stage Builds (Optional)

  1. Modify the Dockerfile as instructed in the comments to create a multi-stage build. Rebuild and inspect the image again.

Part 1.2 Build and Run the Frontend Application Container

  1. Explore the example Next.js application in the nextjs-frontend directory. It fetches the randomly generated color from the color-api and displays it.
  2. Build and run the container: 
    docker build -t nextjs-frontend:latest nextjs-frontend
docker run -p 3000:3000 nextjs-frontend:latest
  3. Open http://localhost:3000 in your browser.

Part 1.3 Connecting Containers Using docker-compose

  1. Review the docker-compose.yaml file.
  2. Note that it adds PostgreSQL to the deployment and configures the COLOR_API_URL environment variable.
  3. Build and start the containers using the following command. Use the image references from GitHub to avoid building the images yourself: 
    docker compose up
  4. Open http://localhost:3000 in your browser. It should now return a color from the color-api.

Section 2 Kubernetes Deployment

Part 2.1 Installing and Starting Minikube

Follow the instructions at Minikube Start.

brew install minikube && minikube start

Enable the storage provisioner and ingress controller:

minikube addons enable storage-provisioner
minikube addons enable ingress

Part 2.2 Create the "workshop" Namespace in Minikube

Set the kubectl alias to the Minikube cluster:

alias kubectl="minikube kubectl --"

Create the namespace:

kubectl create namespace workshop

Alternatively, install kubectl by following the official documentation.

Part 2.3 PostgreSQL Installation Using Helm

Add the Bitnami repository:

helm repo add bitnami https://charts.bitnami.com/bitnami

Install PostgreSQL:

helm install -n workshop postgresql bitnami/postgresql
# or
helm install postgresql -n workshop oci://registry-1.docker.io/bitnamicharts/postgresql

Inspect the running deployment to ensure it started successfully:

kubectl get pods -n workshop

NAME                        READY   STATUS    RESTARTS       AGE
postgresql-0                1/1     Running   1 (2m5s ago)   15m

Part 2.4 Creating Kubernetes Deployment Files

  • Deploy the COLOR-API

    The Next.js frontend deployment files are already provided as an example in the kube/nextjs-frontend directory.

    kubectl apply -n workshop -f kube/nextjs-frontend

    Create the color-api deployment files and apply them to the cluster:

    kubectl apply -n workshop -f kube/color-api

    Hints:

    • You need a deployment and a service.
    • Use the image ghcr.io/l3montree-dev/kubernetes-developer-workshop/color-api:latest.
    • The application runs on port 8080.
    • Required environment variables:
      • DB_URL: jdbc:postgresql://postgresql:5432/postgres
      • DB_USER: postgres
      • DB_PASS: ? (check the secret created by the PostgreSQL Helm chart)

  • Configure the connection between the frontend and the color-api using the COLOR_API_URL environment variable.

Part 2.5 Deploy the Entire Application

Kubernetes is idempotent, so you can apply the entire configuration:

kubectl apply -n workshop -R -f kube

Check the running pods:

kubectl get pods -n workshop

NAME                         READY   STATUS    RESTARTS      AGE
color-api-5c8bc7f9d6-2t7df   1/1     Running   0             23m
color-api-5c8bc7f9d6-72qnw   1/1     Running   0             23m
frontend-68bcf87ffd-n7wtp    1/1     Running   0             39m
postgresql-0                 1/1     Running   1 (49m ago)   63m

Use Minikube to view the frontend:

minikube tunnel

The frontend should be available at http://127.0.0.1.

Section 3: Helm Charts

Part 3.1 Create a Helm Chart for the color-api

Create a new Helm chart:

helm create color-application

Remove everything except Chart.yaml and values.yaml:

rm -rf color-application/templates/* color-application/charts

Clear the values.yaml file:

echo "" > color-application/values.yaml

Copy the contents of the kube folder into the templates folder of the Helm chart:

cp -r kube/* color-application/templates/

Modify the color-application/nextjs-frontend/ingress.yaml file to use the path prefix /helm.

Install the Helm chart:

helm upgrade --install -n workshop-helm --create-namespace color-application color-application

Check the running pods in the workshop-helm namespace:

kubectl get pods -n workshop-helm

NAME                         READY   STATUS                       RESTARTS   AGE
color-api-564cf76456-nchc7   0/1     CreateContainerConfigError   0          9m56s
color-api-564cf76456-w2dkr   0/1     CreateContainerConfigError   0          9m56s
frontend-68bcf87ffd-6pkz9    1/1     Running                      0          9m56s

The color-api pods are not starting because the database is missing. Add the PostgreSQL dependency to Chart.yaml:

dependencies:
  - name: postgresql
    version: 16.4.9
    repository: https://charts.bitnami.com/bitnami

Install the Helm chart again:

helm upgrade --install -n workshop-helm --create-namespace color-application color-application

If it fails, install the dependencies first:

helm dependency update color-application

Apply the Helm chart again:

helm upgrade --install -n workshop-helm --create-namespace color-application color-application

Check the running pods:

kubectl get pods -n workshop-helm

NAME                             READY   STATUS                       RESTARTS   AGE
color-api-564cf76456-khh2s       0/1     CreateContainerConfigError   0          52s
color-api-564cf76456-w2dkr       0/1     CreateContainerConfigError   0          16m
color-application-postgresql-0   1/1     Running                      0          108s
frontend-68bcf87ffd-6pkz9        1/1     Running                      0          16m

The color-api pods are still not starting. Configure the environment variables correctly:

env:
  - name: DB_PASS
    valueFrom:
      secretKeyRef:
        name: color-application-postgresql
        key: postgres-password

Check if the DB_URL is correctly configured. Find the service name with:

kubectl get services -n workshop-helm

Apply the Helm chart again:

helm upgrade --install -n workshop-helm --create-namespace color-application color-application

Check the running pods and visit the frontend.

Part 3.2 Configure Helm Values for Replicas

Set the number of replicas in values.yaml:

# values.yaml
replicaCount: 2

Use the replicaCount value in the color-application/color-api/deployment.yaml:

spec:
  replicas: {{ .Values.replicaCount }}

Apply the Helm chart again:

helm upgrade --install -n workshop-helm --create-namespace color-application color-application

Check the running pods:

kubectl get pods -n workshop-helm

NAME                             READY   STATUS    RESTARTS   AGE
color-api-86f8b7bb7d-29mt7       1/1     Running   0          8m48s
color-api-86f8b7bb7d-2lzql       1/1     Running   0          8m46s
color-api-86f8b7bb7d-rl78x       1/1     Running   0          15s
color-application-postgresql-0   1/1     Running   0          10m
frontend-68bcf87ffd-vvv5v        1/1     Running   0          10m

Visit the frontend. You should see three different colors after reloading a few times.

Part 3.3 Package the Helm Chart

Package the Helm chart:

helm package color-application

The packaged Helm chart is in the color-application-0.1.0.tgz file. Install the packaged Helm chart:

helm install -n workshop-helm-packaged --create-namespace color-application color-application-0.1.0.tgz --set replicaCount=1

The installation fails due to a conflicting path in the ingress. Maybe this can be made configurable as well?

Miscellaneous

Building and Publishing Docker Images

Build and push the Docker images:

docker buildx build --push --platform linux/arm64/v8,linux/amd64 -t ghcr.io/l3montree-dev/kubernetes-developer-workshop/nextjs-frontend:latest nextjs-frontend
docker buildx build --push --platform linux/arm64/v8,linux/amd64 -t ghcr.io/l3montree-dev/kubernetes-developer-workshop/color-api:latest color-api

About

No description, website, or topics provided.

Resources

Readme

License

MIT license
Activity
Custom properties

Stars

3 stars

Watchers

0 watching

Forks

0 forks
Report repository

Releases

No releases published

Packages

 
 
 

Contributors 2

  •  
  •  

Languages