Is Docker Compose Fundamentals — Defining Multi-Container Applications free to learn on DevOps Network?

Yes - this topic, like everything on DevOps Network, is 100% free with no paywall or sign-up gate.

Docker Compose Fundamentals — Defining Multi-Container Applications

Q: What does the Docker Compose Fundamentals — Defining Multi-Container Applications topic cover?

Define and run multi-container applications with Docker Compose — covering the compose file schema, service configuration, and essential CLI commands.

Overview and What You Will Learn

Running a web application locally used to mean installing PostgreSQL, Redis, and RabbitMQ on your laptop, managing three different service configs, and hoping nothing conflicted with another project. Docker Compose replaces all of that with a single YAML file and a single command.

In this guide you will learn the complete Docker Compose file schema, how services, networks, and volumes work together, and every essential CLI command for managing a Compose application.

Core Principles

Bash

+------------------------------------------+
| docker-compose.yml                       |
|                                          |
| services:         <- what to run         |
|   api:            <- one container       |
|   postgres:       <- another container   |
|   redis:          <- another container   |
|                                          |
| networks:         <- how they connect    |
|   app-network:                           |
|                                          |
| volumes:          <- where data lives    |
|   postgres-data:                         |
+------------------------------------------+
         |
         | docker compose up -d
         |
         v
+------------------------------------------+
| Three containers running                 |
| Connected on app-network                 |
| Data persisted in postgres-data volume   |
| api reaches postgres by name             |
+------------------------------------------+

Detailed Step-by-Step Practical Lab

Milestone 1: The Complete docker-compose.yml Schema

YAML

version: "3.8"      # Compose file format version — use 3.8 for most projects
 
services:
 
  # ── Service Definition ─────────────────────────────
  payment-api:
 
    # What image to use (or build)
    image: registry.razorpay.in/payment-api:v3.1.0
 
    # Or build from a Dockerfile
    build:
      context: .              # Build context directory
      dockerfile: Dockerfile  # Which Dockerfile (default: Dockerfile)
      args:
        BUILD_ENV: production
 
    # Container name (otherwise: projectname_servicename_1)
    container_name: payment-api
 
    # Port mapping: host:container
    ports:
      - "8080:8080"
      - "127.0.0.1:9090:9090"  # Only localhost can access this port
 
    # Environment variables
    environment:
      NODE_ENV: production
      DB_HOST: postgres         # Use service name as hostname
      DB_PORT: "5432"
      DB_NAME: payments
 
    # Load env vars from a file (keeps secrets out of compose file)
    env_file:
      - .env
      - .env.production
 
    # Volume mounts
    volumes:
      - ./config:/app/config:ro              # Bind mount, read-only
      - payment-data:/app/data               # Named volume
      - type: tmpfs
        target: /tmp                         # In-memory storage
 
    # Which networks this service joins
    networks:
      - app-network
 
    # Service dependency ordering
    depends_on:
      postgres:
        condition: service_healthy    # Wait until postgres is healthy
      redis:
        condition: service_started    # Just wait until redis starts
 
    # Restart policy
    restart: unless-stopped
 
    # Resource limits
    deploy:
      resources:
        limits:
          cpus: "1.5"
          memory: 512M
        reservations:
          cpus: "0.25"
          memory: 128M
 
    # Health check
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8080/health"]
      interval: 30s
      timeout: 5s
      retries: 3
      start_period: 15s
 
    # Override the default command
    command: ["node", "dist/server.js", "--port", "8080"]
 
  # ── Database Service ───────────────────────────────
  postgres:
    image: postgres:15-alpine
    environment:
      POSTGRES_DB: payments
      POSTGRES_USER: api_user
      POSTGRES_PASSWORD_FILE: /run/secrets/db-password  # Secret from file
    volumes:
      - postgres-data:/var/lib/postgresql/data
      - ./db/init.sql:/docker-entrypoint-initdb.d/init.sql  # Init script
    networks:
      - app-network
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U api_user -d payments"]
      interval: 10s
      timeout: 5s
      retries: 5
 
  # ── Cache Service ──────────────────────────────────
  redis:
    image: redis:7-alpine
    command: redis-server --appendonly yes  # Enable persistence
    volumes:
      - redis-data:/data
    networks:
      - app-network
 
# ── Networks ─────────────────────────────────────────
networks:
  app-network:
    driver: bridge
 
# ── Volumes ──────────────────────────────────────────
volumes:
  postgres-data:    # Docker creates and manages this
  redis-data:
  payment-data:

Milestone 2: Essential Compose CLI Commands

Bash

# Start all services in background
docker compose up -d
 
# Start and rebuild images first
docker compose up -d --build
 
# Stop and remove containers (keeps volumes and networks)
docker compose down
 
# Stop, remove containers, AND remove named volumes (DELETES DATA)
docker compose down -v
 
# View status of all services
docker compose ps
# NAME          IMAGE          STATUS          PORTS
# payment-api   payment-api    Up 5 minutes    0.0.0.0:8080->8080/tcp
# postgres      postgres:15    Up 5 minutes (healthy)
# redis         redis:7        Up 5 minutes
 
# View logs from all services
docker compose logs
 
# Follow logs from a specific service
docker compose logs -f payment-api
 
# Run a command in a running service container
docker compose exec postgres psql -U api_user -d payments
 
# Run a one-off command (starts a new container)
docker compose run --rm payment-api npm run migrate
 
# Scale a service (run multiple instances)
docker compose up -d --scale payment-api=3
 
# Restart a specific service
docker compose restart payment-api
 
# Pull latest images for all services
docker compose pull
 
# Build all services that have a build context
docker compose build
 
# See resource usage of running services
docker compose top

Milestone 3: Compose v1 vs Compose v2

Bash

# Compose v1: separate binary (docker-compose)
# Installed separately, uses Python, slower
docker-compose up -d   # Old way
 
# Compose v2: built into Docker CLI as a plugin
# Faster, Go-based, installed with Docker Desktop and Docker Engine 20.10+
docker compose up -d   # New way (space, not hyphen)
 
# Check which version you have
docker compose version
# Docker Compose version v2.23.0
 
# The compose file format is the same — only the binary changed
# Always use 'docker compose' (v2) on modern systems

Common Mistakes

Mistake	Consequence	Fix
Using `depends_on` without health checks	Service starts before dependency is ready	Add `condition: service_healthy` with a healthcheck
Hardcoding secrets in compose file	Secrets in version control	Use `env_file` with `.env` in `.gitignore`
No resource limits	One service starves others	Always set `deploy.resources.limits`
`docker compose down -v` by accident	All database data deleted permanently	Know exactly what `-v` does before using it
Using `container_name` with scaling	Conflict when scaling to multiple instances	Remove `container_name` for services you want to scale

Tip
Use docker compose config to see the fully resolved compose file with all variable substitutions applied. This is the fastest way to verify that your .env file variables are being read correctly and that all service configurations look as expected.

Remember
docker compose down keeps your volumes (your data is safe). docker compose down -v deletes your volumes (your data is gone permanently). This is the most common cause of accidental data loss with Docker Compose.

JSON

{
  "title": "Docker Compose for Local Development — Full Stack in One Command",
  "slug": "docker-compose-local-development",
  "cluster": "docker",
  "description": "Build a complete local development stack with Docker Compose — API server, PostgreSQL, Redis, and a message queue — with hot reloading and proper config.",
  "primaryKeyword": "docker compose local development"
}

Docker Compose for Local Development — Full Stack in One Command

Overview and What You Will Learn

Every developer at CRED or Zerodha runs a complex application stack locally. Instead of installing PostgreSQL, Redis, and RabbitMQ on every developer's laptop — and dealing with version conflicts, OS differences, and "works on my machine" bugs — the team uses a single docker compose up command that gives every developer the exact same environment.

Core Principles

The ideal local development Compose file has three properties: hot reloading (code changes appear instantly without rebuild), isolated data (each developer has their own database), and production parity (the same images and config as production, with minor development overrides).

Complete Local Development Stack

YAML

# docker-compose.yml — base configuration
version: "3.8"
 
services:
  api:
    build:
      context: .
      dockerfile: Dockerfile.dev      # Development dockerfile with dev tools
    ports:
      - "3000:3000"
      - "9229:9229"                   # Node.js debugger port
    volumes:
      - ./src:/app/src                # Bind mount for hot reloading
      - /app/node_modules             # Keep container's node_modules
    environment:
      NODE_ENV: development
      DB_HOST: postgres
      REDIS_HOST: redis
      AMQP_URL: amqp://rabbitmq:5672
    depends_on:
      postgres:
        condition: service_healthy
      redis:
        condition: service_started
    command: ["npm", "run", "dev"]    # Dev command with nodemon/ts-node
 
  postgres:
    image: postgres:15-alpine
    ports:
      - "5432:5432"                   # Expose for local DB GUI (TablePlus, etc.)
    environment:
      POSTGRES_DB: myapp_dev
      POSTGRES_USER: developer
      POSTGRES_PASSWORD: devpassword123
    volumes:
      - postgres-dev-data:/var/lib/postgresql/data
      - ./db/seed.sql:/docker-entrypoint-initdb.d/seed.sql
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U developer -d myapp_dev"]
      interval: 5s
      timeout: 3s
      retries: 5
 
  redis:
    image: redis:7-alpine
    ports:
      - "6379:6379"                   # Expose for redis-cli on host
 
  rabbitmq:
    image: rabbitmq:3-management-alpine
    ports:
      - "5672:5672"                   # AMQP port
      - "15672:15672"                 # Management UI — http://localhost:15672
    environment:
      RABBITMQ_DEFAULT_USER: developer
      RABBITMQ_DEFAULT_PASS: devpassword123
    volumes:
      - rabbitmq-dev-data:/var/lib/rabbitmq
 
  mailhog:
    image: mailhog/mailhog           # Email capture for development
    ports:
      - "1025:1025"                  # SMTP port
      - "8025:8025"                  # Web UI — http://localhost:8025
    # Send emails to localhost:1025, view them in browser at :8025
 
volumes:
  postgres-dev-data:
  rabbitmq-dev-data:

Dockerfile

# Dockerfile.dev — development-specific
FROM node:20-alpine
WORKDIR /app
 
# Install development tools
RUN apk add --no-cache git curl
 
COPY package.json package-lock.json ./
RUN npm install                      # Install ALL deps including devDependencies
 
# Source is mounted via bind mount — not copied here
EXPOSE 3000 9229
CMD ["npm", "run", "dev"]           # nodemon watches for changes

Hot Reloading Setup

YAML

# The critical volume mounts for hot reloading:
volumes:
  - ./src:/app/src         # Your source code mounted into container
  - /app/node_modules      # Anonymous volume — keeps container's node_modules
                           # Without this, host's node_modules overrides container's
                           # (Different OS = different native modules)

TEXT

// package.json
{
  "scripts": {
    "dev": "nodemon --watch src --ext ts,js,json --exec ts-node src/index.ts"
  }
}

Bash

# Start the full dev stack
docker compose up -d
 
# Watch API logs while developing
docker compose logs -f api
 
# Run database migrations
docker compose exec api npm run migrate
 
# Reset the database
docker compose exec postgres psql -U developer -c "DROP DATABASE myapp_dev;"
docker compose exec postgres psql -U developer -c "CREATE DATABASE myapp_dev;"
docker compose restart postgres
 
# Open a postgres shell
docker compose exec postgres psql -U developer -d myapp_dev

Override File for Development

YAML

# docker-compose.override.yml
# Auto-merged with docker-compose.yml when both are present
version: "3.8"
 
services:
  api:
    environment:
      DEBUG: "*"              # Enable all debug logging
      LOG_LEVEL: debug
    volumes:
      - ./tests:/app/tests   # Mount test files for running tests inside container

Tip
Expose all service ports in your local development Compose file (postgres 5432, redis 6379, rabbitmq 15672) so you can connect with GUI tools like TablePlus, RedisInsight, and the RabbitMQ management UI. Remove these port exposures in production compose files — external access is not needed.

Remember
The anonymous volume trick (- /app/node_modules) is critical on macOS and Windows. Without it, your macOS node_modules (compiled for macOS) override the container's node_modules (compiled for Linux), and native modules fail at runtime.

JSON

{
  "title": "Docker Compose Health Checks and Dependency Ordering",
  "slug": "docker-compose-healthchecks-dependencies",
  "cluster": "docker",
  "description": "Configure service health checks and correct startup ordering in Docker Compose to prevent applications from starting before their dependencies are ready.",
  "primaryKeyword": "docker compose health check"
}

Docker Compose Health Checks and Dependency Ordering

Overview and What You Will Learn

The most common Docker Compose failure pattern: your API starts, tries to connect to PostgreSQL, gets connection refused because PostgreSQL is still initialising, crashes, and gets stuck in a restart loop. The fix is health checks combined with depends_on condition: service_healthy.

Core Principles

◈ DIAGRAM

WRONG: depends_on without health check
  postgres starts -> api starts immediately
  postgres still initialising (takes 3-5 seconds)
  api tries to connect -> connection refused
  api crashes -> restart loop
 
CORRECT: depends_on with service_healthy
  postgres starts
  postgres health check runs every 5s
  postgres becomes healthy (pg_isready returns 0)
  THEN api starts
  api connects successfully -> no crash

Health Check Configuration

YAML

version: "3.8"
 
services:
  api:
    image: registry.razorpay.in/payment-api:v3.1.0
    depends_on:
      postgres:
        condition: service_healthy    # Wait until postgres passes health check
      redis:
        condition: service_healthy
      rabbitmq:
        condition: service_healthy
 
  postgres:
    image: postgres:15-alpine
    environment:
      POSTGRES_DB: payments
      POSTGRES_USER: api_user
      POSTGRES_PASSWORD: secret
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U api_user -d payments"]
      interval: 5s          # Run health check every 5 seconds
      timeout: 3s           # Health check must complete in 3 seconds
      retries: 5            # Fail after 5 consecutive failures
      start_period: 10s     # Wait 10s after start before first check
                            # Prevents false failures during initialisation
 
  redis:
    image: redis:7-alpine
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 5s
      timeout: 3s
      retries: 3
 
  rabbitmq:
    image: rabbitmq:3-alpine
    healthcheck:
      test: ["CMD", "rabbitmq-diagnostics", "ping"]
      interval: 10s
      timeout: 5s
      retries: 5
      start_period: 30s     # RabbitMQ takes longer to start

Health Check Command Examples

YAML

# HTTP endpoint check (requires curl in image)
healthcheck:
  test: ["CMD", "curl", "-f", "http://localhost:8080/health"]
 
# Use wget if curl is not available (Alpine images)
healthcheck:
  test: ["CMD", "wget", "--no-verbose", "--tries=1",
         "--spider", "http://localhost:8080/health"]
 
# PostgreSQL readiness
healthcheck:
  test: ["CMD-SHELL", "pg_isready -U postgres"]
 
# MySQL readiness
healthcheck:
  test: ["CMD", "mysqladmin", "ping", "-h", "localhost"]
 
# Redis ping
healthcheck:
  test: ["CMD", "redis-cli", "ping"]
 
# MongoDB readiness
healthcheck:
  test: ["CMD", "mongosh", "--eval", "db.adminCommand('ping')"]
 
# Simple TCP port check
healthcheck:
  test: ["CMD-SHELL", "nc -z localhost 8080 || exit 1"]
 
# Custom application check via exit code
healthcheck:
  test: ["CMD-SHELL", "node /app/scripts/healthcheck.js || exit 1"]

Monitoring Health Status

Bash

# Check health status of all services
docker compose ps
# NAME       STATUS
# api        Up 2 minutes (healthy)
# postgres   Up 3 minutes (healthy)
# redis      Up 3 minutes (healthy)
 
# If a service is unhealthy:
# docker compose ps shows: Up 2 minutes (unhealthy)
 
# See health check output for a container
docker inspect postgres-container --format \
  '{{json .State.Health}}' | jq
# {
#   "Status": "healthy",
#   "FailingStreak": 0,
#   "Log": [
#     {
#       "Start": "2024-01-15T09:00:00Z",
#       "End": "2024-01-15T09:00:00.123Z",
#       "ExitCode": 0,
#       "Output": "/var/run/postgresql:5432 - accepting connections"
#     }
#   ]
# }
 
# If health check is failing, the Output field shows why
docker inspect unhealthy-container \
  --format '{{(index .State.Health.Log 0).Output}}'
# FATAL: database "payments" does not exist
# -> Database not created yet — check POSTGRES_DB env var

Tip
Always set start_period on health checks for services that take time to initialise (like PostgreSQL, RabbitMQ, Elasticsearch). Without start_period, health check failures during the normal startup period count toward the failure threshold and can mark the service as unhealthy before it even had a chance to start properly.

Common Mistake
Using depends_on without condition: service_healthy. The basic depends_on: postgres only waits for the container to start — not for PostgreSQL to be ready to accept connections. PostgreSQL takes 2-5 seconds after the container starts to become ready. Without service_healthy, your API will almost certainly try to connect before PostgreSQL is ready.

JSON

{
  "title": "Docker Compose in CI/CD Pipelines — GitHub Actions Integration",
  "slug": "docker-compose-ci-cd",
  "cluster": "docker",
  "description": "Use Docker Compose in GitHub Actions CI/CD pipelines to run integration tests against real service dependencies — databases, caches, and message queues.",
  "primaryKeyword": "docker compose ci cd"
}

Docker Compose in CI/CD Pipelines — GitHub Actions Integration

Overview and What You Will Learn

Integration tests that mock the database are not real integration tests. Real integration tests hit a real PostgreSQL database, a real Redis cache, and a real message queue — the same services your application uses in production. Docker Compose makes running these real dependencies in GitHub Actions free and reliable.

Core Principles

Bash

GitHub Actions runner:
  ubuntu-latest has Docker pre-installed
  
Your CI workflow:
  1. docker compose up -d (start postgres, redis, etc.)
  2. Wait for services to be healthy
  3. Run your integration tests against real services
  4. docker compose down (cleanup)
  
Result: integration tests that test real behavior
        not mocked behavior

Complete GitHub Actions Workflow

YAML

# .github/workflows/ci.yml
name: CI Pipeline
 
on:
  push:
    branches: [main, develop]
  pull_request:
    branches: [main]
 
jobs:
  integration-tests:
    runs-on: ubuntu-latest
 
    steps:
      - name: Checkout
        uses: actions/checkout@v4
 
      - name: Set up Docker Buildx
        uses: docker/setup-buildx-action@v3
 
      - name: Start services with Docker Compose
        run: |
          docker compose -f docker-compose.ci.yml up -d
          echo "Services started"
      - name: Wait for services to be healthy
        run: |
          echo "Waiting for postgres..."
          timeout 60 bash -c \
            'until docker compose -f docker-compose.ci.yml exec -T postgres \
              pg_isready -U api_user -d payments; do sleep 2; done'
          
          echo "Waiting for redis..."
          timeout 30 bash -c \
            'until docker compose -f docker-compose.ci.yml exec -T redis \
              redis-cli ping | grep -q PONG; do sleep 1; done'
          
          echo "All services ready"
 
      - name: Run database migrations
        run: |
          docker compose -f docker-compose.ci.yml exec -T api \
            npm run migrate
      - name: Run integration tests
        run: |
          docker compose -f docker-compose.ci.yml exec -T api \
            npm run test:integration
        env:
          CI: true
 
      - name: Collect logs on failure
        if: failure()
        run: |
          docker compose -f docker-compose.ci.yml logs > ci-logs.txt 2>&1
      - name: Upload logs as artifact on failure
        if: failure()
        uses: actions/upload-artifact@v4
        with:
          name: ci-logs
          path: ci-logs.txt
 
      - name: Teardown services
        if: always()      # Run even if tests failed
        run: |
          docker compose -f docker-compose.ci.yml down -v
          # -v removes volumes — clean slate for next run

YAML

# docker-compose.ci.yml — CI-specific compose file
version: "3.8"
 
services:
  api:
    build: .
    environment:
      NODE_ENV: test
      DB_HOST: postgres
      DB_PORT: "5432"
      DB_NAME: payments_test
      DB_USER: api_user
      DB_PASSWORD: testpassword
      REDIS_HOST: redis
 
  postgres:
    image: postgres:15-alpine
    environment:
      POSTGRES_DB: payments_test
      POSTGRES_USER: api_user
      POSTGRES_PASSWORD: testpassword
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U api_user -d payments_test"]
      interval: 5s
      timeout: 3s
      retries: 5
    # No volume — data does not need to persist in CI
    # Faster than using a volume, clean state every run
 
  redis:
    image: redis:7-alpine
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 3s
      retries: 5
    # No volume — clean state every run

Caching Docker Layers in CI

YAML

# Significantly speeds up CI builds by caching built layers
- name: Build with cache
  uses: docker/build-push-action@v5
  with:
    context: .
    push: false
    load: true
    tags: myapp:test
    cache-from: type=gha       # Read cache from GitHub Actions cache
    cache-to: type=gha,mode=max  # Write new layers to cache

Common Mistakes

Mistake	Consequence	Fix
Not waiting for health checks	Tests fail intermittently — race condition	Use `timeout bash -c 'until condition; do sleep 1; done'`
Using volumes in CI	State bleeds between runs	No volumes in CI compose — in-memory or ephemeral storage
Not cleaning up with `always()`	Runner disk fills up with old containers and volumes	Always use `if: always()` on teardown step
Running tests directly on runner	Inconsistent dependencies between runners	Run tests inside the compose service container

Tip
Use docker compose -f docker-compose.ci.yml exec -T service command (note the -T flag) for non-interactive commands in CI. Without -T, GitHub Actions may hang waiting for TTY input that never comes.

Remember
Add if: always() to your teardown step. If tests fail and you do not tear down, the next CI run may fail because ports are already in use from the previous run's containers still running on the same GitHub Actions runner.

JSON

{
  "title": "Docker Compose Production Patterns — What to Use and What to Avoid",
  "slug": "docker-compose-production-patterns",
  "cluster": "docker",
  "description": "Understand what Docker Compose is appropriate for in production, its limitations compared to Kubernetes, and patterns for small-team production deployments.",
  "primaryKeyword": "docker compose production"
}

Docker Compose Production Patterns — What to Use and What to Avoid

Overview and What You Will Learn

Docker Compose is excellent for local development and CI. In production it is appropriate for specific situations — small teams, single-host deployments, internal tools — and completely wrong for others. Understanding where the line is prevents you from building on the wrong foundation.

When Compose Is Appropriate in Production

TEXT

Compose is GOOD for production when:
  * Single host deployment (one server)
  * Team of 1-5 engineers
  * Non-critical internal tools
  * Staging / preview environments
  * Startups with simple infrastructure needs
 
Compose is NOT appropriate when:
  * Multiple hosts needed (high availability)
  * Self-healing required (container fails = outage)
  * Rolling updates with zero downtime required
  * Horizontal scaling across nodes required
  * Production traffic is business-critical

Production-Ready Compose Configuration

YAML

# docker-compose.prod.yml
version: "3.8"
 
services:
  api:
    image: registry.razorpay.in/payment-api:v3.1.0
    restart: unless-stopped           # Restart on crash, not on manual stop
    ports:
      - "127.0.0.1:8080:8080"        # Only localhost — nginx proxies externally
    environment:
      NODE_ENV: production
    env_file:
      - .env.production               # Production secrets from file
    logging:
      driver: "json-file"
      options:
        max-size: "100m"              # Log rotation — prevents disk fill
        max-file: "3"                 # Keep 3 log files
    deploy:
      resources:
        limits:
          cpus: "1.5"
          memory: 512M
    healthcheck:
      test: ["CMD", "curl", "-f", "http://localhost:8080/health"]
      interval: 30s
      timeout: 5s
      retries: 3
 
  nginx:
    image: nginx:1.25-alpine
    restart: unless-stopped
    ports:
      - "80:80"
      - "443:443"
    volumes:
      - ./nginx/nginx.conf:/etc/nginx/nginx.conf:ro
      - ./nginx/ssl:/etc/nginx/ssl:ro
      - nginx-logs:/var/log/nginx
 
  postgres:
    image: postgres:15-alpine
    restart: unless-stopped
    volumes:
      - postgres-data:/var/lib/postgresql/data
    logging:
      driver: "json-file"
      options:
        max-size: "50m"
        max-file: "5"
 
volumes:
  postgres-data:
    driver: local
  nginx-logs:

Deploying Updates Without Downtime

Compose does not have built-in rolling updates. This is the pattern for zero-downtime updates:

Bash

# Pull new images
docker compose pull
 
# Recreate only the service that changed (api, not postgres)
docker compose up -d --no-deps api
# --no-deps: do not recreate dependencies (postgres, redis)
# This pulls new api image and recreates only the api container
 
# Note: there IS a brief downtime during the container recreation
# For true zero downtime, use Kubernetes or Traefik with multiple replicas
 
# For simple single-host zero-downtime:
# Run nginx in front, bring up new container on different port,
# update nginx config, reload nginx (zero downtime), remove old container

Compose vs Kubernetes Decision

Need	Compose	Kubernetes
Local development	Excellent	Overkill
Single host production	Good	Overkill
Multi-host / HA	Cannot do it	Designed for this
Self-healing	No (manual restart)	Yes (automatic)
Rolling updates (zero downtime)	Manual workaround	Built-in
Horizontal autoscaling	No	Built-in (HPA)
Service mesh / mTLS	No	Yes (Istio, Linkerd)
Complexity	Low	High
Ops cost	Very low	Significant

Tip
Use unless-stopped restart policy (not always) for production services. always restarts even when you intentionally stop the container for maintenance. unless-stopped restarts on crash but respects manual docker stop.

Common Mistake
Using Compose in production for business-critical services without understanding its limitations. If the server goes down, all your containers go down. If a container crashes and the restart fails (e.g., port conflict), the service is down until you SSH in and fix it manually. Plan for this reality before choosing Compose over Kubernetes for production.

Docker Compose Fundamentals — Defining Multi-Container Applications

Overview and What You Will Learn

Core Principles

Detailed Step-by-Step Practical Lab

Milestone 1: The Complete docker-compose.yml Schema

Milestone 2: Essential Compose CLI Commands

Milestone 3: Compose v1 vs Compose v2

Common Mistakes

Docker Compose for Local Development — Full Stack in One Command

Overview and What You Will Learn

Core Principles

Complete Local Development Stack

Hot Reloading Setup

Override File for Development

Docker Compose Health Checks and Dependency Ordering

Overview and What You Will Learn

Core Principles

Health Check Configuration

Health Check Command Examples

Monitoring Health Status

Docker Compose in CI/CD Pipelines — GitHub Actions Integration

Overview and What You Will Learn

Core Principles

Complete GitHub Actions Workflow

Caching Docker Layers in CI

Common Mistakes

Docker Compose Production Patterns — What to Use and What to Avoid

Overview and What You Will Learn

When Compose Is Appropriate in Production

Production-Ready Compose Configuration

Deploying Updates Without Downtime

Compose vs Kubernetes Decision

Resources

Docker Compose vs Kubernetes: When Not to Migrate

How to Reduce Your Docker Image Size by 80 Percent

Explore More in Docker Compose and Multi-Container Applications

Docker Volumes and Persistent Storage — Volumes, Bind Mounts, and tmpfs

Docker Compose for Local Development — Full Stack in One Command

Docker Compose Health Checks and Dependency Ordering

Docker Compose in CI/CD Pipelines — GitHub Actions Integration

Frequently Asked Questions

Is Docker Compose Fundamentals — Defining Multi-Container Applications free to learn on DevOps Network?

What does the Docker Compose Fundamentals — Defining Multi-Container Applications topic cover?