Docker Management

Docker Management SOP

When to Use

• User wants to write or improve a Dockerfile or docker-compose.yml
• User has a container that won't start, is crashing, or behaves unexpectedly
• User wants to reduce image size or improve build speed
• User asks about container networking, volumes, or environment variables

Part 1 — Dockerfile Best Practices

Golden Template (Node.js — multi-stage)

# ─── Stage 1: Dependencies ──────────────────────────────────────────
FROM node:20-alpine AS deps
WORKDIR /app

# Copy only package files first — maximizes layer caching
COPY package.json package-lock.json ./
RUN npm ci --only=production

# ─── Stage 2: Builder ───────────────────────────────────────────────
FROM node:20-alpine AS builder
WORKDIR /app

COPY package.json package-lock.json ./
RUN npm ci

COPY . .
RUN npm run build

# ─── Stage 3: Production Runtime ────────────────────────────────────
FROM node:20-alpine AS runner
WORKDIR /app

ENV NODE_ENV=production

# Security: run as non-root
RUN addgroup -S appgroup && adduser -S appuser -G appgroup
USER appuser

# Copy only what's needed at runtime
COPY --from=deps    /app/node_modules ./node_modules
COPY --from=builder /app/dist         ./dist

EXPOSE 3000
HEALTHCHECK --interval=30s --timeout=10s --start-period=15s --retries=3 \
  CMD wget -qO- http://localhost:3000/health || exit 1

CMD ["node", "dist/main.js"]

Golden Template (Python — FastAPI)

FROM python:3.12-slim AS base
WORKDIR /app
ENV PYTHONDONTWRITEBYTECODE=1 \
    PYTHONUNBUFFERED=1 \
    PIP_NO_CACHE_DIR=1

FROM base AS builder
COPY requirements.txt .
RUN pip install --prefix=/install -r requirements.txt

FROM base AS runner
COPY --from=builder /install /usr/local
COPY . .

RUN adduser --disabled-password --gecos "" appuser && chown -R appuser /app
USER appuser

EXPOSE 8000
HEALTHCHECK --interval=30s --timeout=10s CMD curl -f http://localhost:8000/health || exit 1
CMD ["uvicorn", "main:app", "--host", "0.0.0.0", "--port", "8000"]

Layer Caching Rules

1. Order layers from least to most frequently changing:
   • OS packages → dependency manifests → dependencies → source code → build output
2. COPY package.json . then RUN npm ci before COPY . . — source changes don't bust the dep layer
3. RUN commands that install packages should be combined into one RUN:

   # WRONG — 3 layers
   RUN apt-get update
   RUN apt-get install -y curl
   RUN rm -rf /var/lib/apt/lists/*
   
   # CORRECT — 1 layer
   RUN apt-get update && apt-get install -y curl && rm -rf /var/lib/apt/lists/*
   

Part 2 — .dockerignore

Always create a .dockerignore alongside your Dockerfile:

# Version control
.git
.gitignore

# Dependencies (reinstalled in container)
node_modules
.venv
__pycache__
*.pyc

# Build output (rebuilt in container)
dist
build
.next
out

# Dev tooling
.env
.env.*
*.test.*
*.spec.*
coverage/
.nyc_output

# Docker files themselves
Dockerfile*
docker-compose*

# OS / IDE
.DS_Store
.idea
.vscode
*.swp

Part 3 — Docker Compose Patterns

Standard Web App Stack

version: "3.9"

services:
  app:
    build:
      context: .
      target: runner          # target multi-stage build stage
    ports:
      - "3000:3000"
    environment:
      - NODE_ENV=production
      - DATABASE_URL=postgresql://postgres:secret@db:5432/mydb
    depends_on:
      db:
        condition: service_healthy
    restart: unless-stopped
    networks:
      - app-net

  db:
    image: postgres:16-alpine
    environment:
      POSTGRES_USER: postgres
      POSTGRES_PASSWORD: secret
      POSTGRES_DB: mydb
    volumes:
      - pg-data:/var/lib/postgresql/data
    healthcheck:
      test: ["CMD-SHELL", "pg_isready -U postgres"]
      interval: 10s
      timeout: 5s
      retries: 5
    networks:
      - app-net

  redis:
    image: redis:7-alpine
    volumes:
      - redis-data:/data
    healthcheck:
      test: ["CMD", "redis-cli", "ping"]
      interval: 10s
    networks:
      - app-net

volumes:
  pg-data:
  redis-data:

networks:
  app-net:
    driver: bridge

Dev Override (docker-compose.override.yml)

# Automatically merged with docker-compose.yml in dev
services:
  app:
    build:
      target: builder         # use dev stage with devDependencies
    volumes:
      - .:/app                # live reload via volume mount
      - /app/node_modules     # anonymous volume to preserve container modules
    environment:
      - NODE_ENV=development
    command: npm run dev
    ports:
      - "9229:9229"           # Node.js debugger port

Part 4 — Debugging Commands

Container Won't Start

# See why it exited
docker logs <container_name_or_id>
docker logs --tail 50 myapp

# Inspect the container state and config
docker inspect myapp | jq '.[0].State'

# Override entrypoint to get a shell inside the image
docker run --rm -it --entrypoint sh myimage:latest

Container Running but Misbehaving

# Get a shell in a running container
docker exec -it myapp sh       # alpine/slim images
docker exec -it myapp bash     # full debian/ubuntu images

# Watch real-time resource usage
docker stats myapp

# See all environment variables
docker exec myapp env

# Inspect network connectivity
docker exec myapp wget -qO- http://db:5432  # test service-to-service
docker network ls
docker network inspect app-net

Image Debugging

# Inspect image layers and sizes
docker image history myimage:latest

# Dive tool — interactive layer explorer (install separately)
dive myimage:latest

# Check image metadata
docker inspect myimage:latest | jq '.[0].Config'

# Build with no cache to force fresh layers
docker build --no-cache -t myimage:latest .

# Build a specific stage
docker build --target builder -t myimage-debug:latest .

Volume and Permission Issues

# List volumes
docker volume ls
docker volume inspect pg-data

# Check file ownership inside container
docker exec myapp ls -la /app

# Fix common permission error: EACCES / Permission denied
# → Add to Dockerfile:
RUN chown -R appuser:appgroup /app
USER appuser

Part 5 — Image Size Optimization Checklist

• Use alpine or slim base image variants
• Multi-stage build — only copy runtime artifacts to final stage
• RUN apt-get clean && rm -rf /var/lib/apt/lists/* after apt installs
• npm ci --only=production (not npm install) in production stage
• .dockerignore excludes node_modules, .git, coverage/, test files
• No dev tools (curl, vim, git) installed in the production stage
• Use COPY --chown to avoid a separate RUN chown layer

Benchmark: A typical Node.js service should be under 150MB; Python under 200MB.

# Check final image size
docker images myimage:latest --format "{{.Size}}"

Agent Instructions

1. When asked to write a Dockerfile, always ask: language/runtime, target environment (prod vs dev), any special requirements (GPU, native extensions)
2. Default to multi-stage builds for compiled or bundled apps
3. Always include a HEALTHCHECK and a non-root USER in production images
4. When debugging, start with docker logs before anything else — it reveals 90% of issues
5. Check depends_on + healthcheck when services fail to connect to each other
6. After writing a Dockerfile, run through the image optimization checklist mentally