Deploying Full-Stack Node.js Applications with Docker Compose on Azure

How I Deployed a Full-Stack Bookstore with Docker Compose on Azure (Capstone Project)

Vivian Chiamaka Okose deployed The EpicBook, a full-stack online bookstore, using a multi-container architecture on Azure. The system leverages an internal backend network to ensure the MySQL database remains inaccessible from the public internet.

Why This Matters

While ideal models suggest seamless container orchestration, technical reality often involves troubleshooting environment-specific failures like YAML corruption during heredoc pastes or database seeding errors. This project highlights the critical role of healthchecks and named volumes in production, demonstrating how data for 54 books and 53 authors was preserved across container destruction cycles.

Key Insights

• Multi-stage Docker builds optimize production runtimes by using a builder stage to install dependencies and a clean alpine stage for the final image (Okose, 2026).
• Service healthchecks are mandatory for orchestration; using the ‘service_healthy’ condition in docker-compose prevents the application from starting before the MySQL database is ready.
• Network isolation via bridge drivers separates the Nginx-facing frontend network from the sensitive backend network where the database resides.
• Data persistence is validated through the use of named volumes, ensuring that 54 book records survive a ‘docker-compose down’ and ‘up’ sequence.
• Nginx acts as a reverse proxy using Docker’s internal DNS to resolve service names like ‘app:8080’ without hardcoded IP addresses.

Working Examples

Multi-stage Dockerfile for production Node.js environments.

FROM node:18-alpine AS builder
WORKDIR /app
COPY package*.json ./
RUN npm ci --only=production

FROM node:18-alpine
WORKDIR /app
COPY --from=builder /app/node_modules ./node_modules
COPY . .
EXPOSE 8080
CMD ["node", "server.js"]

Docker Compose snippet showing healthcheck-dependent service orchestration.

services:
  db:
    image: mysql:8.0
    networks:
      - backend_network
    healthcheck:
      test: ["CMD", "mysqladmin", "ping", "-h", "localhost"]
      interval: 10s
      retries: 5
  app:
    build: .
    depends_on:
      db:
        condition: service_healthy
    networks:
      - frontend_network
      - backend_network

Practical Applications

• Seed databases by using ‘docker cp’ to move SQL files into the container before executing with ‘source’ to avoid redirection errors. Pitfall: Using stdin redirection which may fail if the file is not present inside the container filesystem.
• Utilize Python scripts to write configuration files in automated environments to avoid YAML corruption. Pitfall: Using shell heredocs which can be broken by special characters or formatting issues during terminal pastes.

References:

• https://dev.to/vivian_okose/how-i-deployed-a-full-stack-bookstore-with-docker-compose-on-azure-capstone-project-11dc
• https://github.com/vivianokose/the-epicbook_capstone.git