Docker Manager

Container lifecycle management with production-ready patterns.

When to Use

• User mentions Docker, containers, images, or Dockerfiles
• User wants to containerize an application
• User needs docker-compose orchestration
• User asks about container debugging or optimization

Procedure

Building Images

1. Analyze the project to determine base image and dependencies
2. Create a multi-stage Dockerfile for minimal final image
3. Build: docker build -t name:tag .
4. Verify: docker images | grep name

Running Containers

1. Run: docker run -d --name my-app -p 8080:3000 name:tag
2. Check logs: docker logs -f my-app
3. Exec into: docker exec -it my-app /bin/sh

Docker Compose

1. Create docker-compose.yml with service definitions
2. Start: docker compose up -d
3. Monitor: docker compose logs -f
4. Scale: docker compose up -d --scale web=3

Cleanup

1. Stop containers: docker stop $(docker ps -q)
2. Remove containers: docker container prune
3. Remove images: docker image prune -a
4. Remove volumes: docker volume prune

Best Practices

• Use .dockerignore to exclude unnecessary files
• Pin base image versions (node:20-alpine, not node:latest)
• Use multi-stage builds to reduce final image size
• Run as non-root user in production
• Use HEALTHCHECK for container health monitoring

Pitfalls

• Never store secrets in Dockerfiles or images
• Avoid running as root in production containers
• Watch for large context sizes slowing builds
• Handle signal propagation for graceful shutdown

Verification

• Container running: docker ps | grep name
• Health check passing: docker inspect --format='{{.State.Health}}' name
• Logs clean: docker logs --tail 50 name