Microservices · Lesson 11

Containerizing with Docker

Why Containers Help

A container packages an application with its runtime assumptions into an isolated, portable unit. Docker images make the same service artifact runnable on a developer machine, CI worker, or production platform. Containers do not remove the need for security, monitoring, or resource limits.

A Container-Friendly Java Process

A Java service should write logs to standard output, read configuration from the environment, and handle termination signals so the platform can manage it.

public class ContainerProcess {
    public static void main(String[] arguments) throws InterruptedException {
        String serviceName = System.getenv().getOrDefault("SERVICE_NAME", "course-service");
        System.out.println(serviceName + " is ready");
        Thread.sleep(100);
        System.out.println(serviceName + " stopped cleanly");
    }
}

A Dockerfile commonly selects a small trusted JRE image, copies the built artifact, exposes the service port, and uses java -jar as the entrypoint.

Container Concerns

Common Mistake and Practices

Common Mistake: Baking credentials or local configuration into an image. Image layers can persist in registries and expose values to anyone who can pull them.

• Scan images for vulnerabilities.
• Run as a non-root user.
• Pin base images and rebuild regularly.
• Keep images small and disposable.
• Send logs and metrics outside the container.

Real-World Use Case

Each academy service is built into an immutable image and deployed with environment-specific settings. A failed release can be rolled back by selecting the previous image digest.