With Kamal 2, you can deploy web apps anywhere—from bare metal to cloud VMs. Kamal offers zero-downtime deploys, rolling restarts, asset bridging, remote builds, accessory service management, and everything else you need to deploy and manage your web app in production with Docker. Originally built for Rails apps, Kamal works with any web app that can be containerized.
Multipass provides Ubuntu VMs on demand for any workstation. With cloud-style VMs at your fingertips, you can launch instances of Ubuntu and initialize them with cloud-init metadata in the same way as AWS, Azure, Google Cloud, IBM, and Oracle. This allows you to simulate a cloud deployment directly on your workstation.
This post builds upon the concepts discussed in Advanced Practices in Spring Boot: Building a Modular Application with Docker, Zipkin, and 100% Code Coverage.
- Prerequisites
- 1. Installing Multipass and Creating VMs
- 2. Install Kamal 2
- 3. Build and Push Docker Image
- 4. Configuring Kamal 2
- 5. Deploying the Application
- 6. Verifying Deployment
- Conclusion
Prerequisites
Before we dive into the development process, ensure you have:
- Java 21 installed on your system. You can install it using SDKMAN and select Java 21.
- Docker and Docker Compose installed for setting up the local environment.
- Clone the project repository:
git clone https://github.com/dmakariev/examples.git
cd examples/spring-boot/bookstore
1. Installing Multipass and Creating VMs
Install Multipass:
brew install --cask multipass
Step 1: Create Base VM
The base-vm will be our template for setting up SSH and Docker correctly. We’ll use it to clone the other VMs.
multipass launch -n base-vm --cpus 2 --memory 2G --disk 10G
Step 2: Configure SSH Access in Base VM
Inside the VM:
multipass shell base-vm
sudo apt update && sudo apt install -y openssh-server
sudo systemctl enable ssh
sudo systemctl start ssh
On your local machine:
ssh-keygen -t rsa -b 4096
multipass transfer ~/.ssh/id_rsa.pub base-vm:/home/ubuntu/
Inside the VM:
multipass shell base-vm
mkdir -p ~/.ssh
cat ~/id_rsa.pub >> ~/.ssh/authorized_keys
chmod 600 ~/.ssh/authorized_keys
chmod 700 ~/.ssh
rm ~/id_rsa.pub
Restart SSH and test access:
sudo systemctl restart ssh
ssh ubuntu@<base-vm-ip>
Step 3: Install Docker in Base VM
sudo apt update && sudo apt install -y docker.io
sudo systemctl enable --now docker
sudo usermod -aG docker ubuntu
newgrp docker
Verify Docker installation:
docker --version
Step 4: Clone VMs
We will now create four additional VMs by cloning base-vm:
multipass clone base-vm --name addon-vm
multipass clone base-vm --name db-vm
multipass clone base-vm --name spring-vm
multipass clone base-vm --name web2-vm
Retrieve the list of running VMs:
multipass list
Expected output:
Name State IPv4 Image
addon-vm Running 192.168.67.10 Ubuntu 24.04 LTS
base-vm Stopped -- Ubuntu 24.04 LTS
db-vm Running 192.168.67.11 Ubuntu 24.04 LTS
spring-vm Running 192.168.67.9 Ubuntu 24.04 LTS
web2-vm Running 192.168.67.12 Ubuntu 24.04 LTS
The spring-vm and web2-vm will act as web instances of our application. In a later blog post, we will cover setting up a load balancer, simulating cloud-hosted solutions like DigitalOcean, AWS Lightsail, and Hetzner.
2. Install Kamal 2
If Ruby is not installed, set it up:
brew install openssl@3 libyaml gmp rust
curl https://mise.run | sh
echo 'eval "$(~/.local/bin/mise activate)"' >> ~/.zshrc
source ~/.zshrc
mise use -g ruby@3
Install Kamal:
gem install kamal
3. Build and Push Docker Image
Ensure your Spring Boot application includes the following LABEL in the Dockerfile:
LABEL service="bookstore"
Build and push the image to GHCR:
docker buildx build --platform linux/amd64,linux/arm64 -t ghcr.io/dmakariev/bookstore:1.0.0 --push .
4. Configuring Kamal 2
Create ./config/deploy.yml in the project directory:
service: bookstore
image: dmakariev/bookstore
servers:
web:
- 192.168.67.9
- 192.168.67.12
ssh:
user: ubuntu
proxy:
ssl: false
app_port: 80
healthcheck:
path: /actuator/health
interval: 3 # seconds
timeout: 40 # seconds
5. Deploying the Application
Run the following commands to deploy:
kamal setup -P --version=1.0.0
kamal deploy -P --version=1.0.0
To update:
kamal deploy --version=1.1.0
6. Verifying Deployment
Check logs:
kamal logs
Ensure the application is running:
curl http://<multipass-vm-ip>/actuator/health
Conclusion
By integrating Multipass, Docker, and Kamal 2, we achieve a robust deployment pipeline for a Spring Boot application. This approach facilitates local testing in a production-like environment while remaining efficient and manageable. Whether scaling up to a cloud provider or maintaining a local dev setup, this workflow ensures seamless deployment.
In an upcoming post, we will discuss setting up a load balancer to distribute traffic across the spring-vm and web2-vm, simulating cloud-hosted solutions.
Happy coding! 🚀