Skip to content

docker.md

Latest commit

 

History

History
241 lines (175 loc) · 5.07 KB

docker.md

File metadata and controls

241 lines (175 loc) · 5.07 KB

Docker Integration

DoMD provides seamless Docker integration, allowing you to run commands in isolated containers with automatic environment detection. This document covers all aspects of using Docker with DoMD.

Table of Contents

Basic Configuration

The .dodocker File

Create a .dodocker file in your project root to configure Docker execution:

# .dodocker
test:
  image: python:3.9-slim  # Base image to use
  description: Run Python tests  # Optional description
  workdir: /app  # Working directory in container
  volumes:  # Volume mappings
    ~/.cache/pip:/.cache/pip  # Cache directory
    .:/app  # Mount current directory
  environment:  # Environment variables
    PYTHONPATH: /app

Running Commands

Run commands with Docker (automatically detected from .dodocker):

# Run tests in Docker
poetry run domd run test

# Force local execution (bypass Docker)
poetry run domd run --no-docker test

# List available Docker commands
poetry run domd list --docker

Advanced Configuration

Multiple Command Configurations

You can configure multiple commands with different settings:

# .dodocker
format:
  image: python:3.9-slim
  description: Format Python code
  volumes:
    .:/app
  command: "black ."  # Override default command

test:
  image: node:16-slim
  description: Run JavaScript tests
  volumes:
    .:/app
    /app/node_modules  # Anonymous volume for dependencies
  environment:
    NODE_ENV: test

Using Docker Compose

For more complex setups, you can use Docker Compose:

# docker-compose.yml
version: '3.8'
services:
  app:
    build: .
    volumes:
      - .:/app
    environment:
      - PYTHONPATH=/app
    working_dir: /app

Then reference it in your .dodocker file:

test:
  compose: docker-compose.yml
  service: app
  command: pytest tests/

Volume Mounting

DoMD handles volume mounting with smart defaults:

  • Current directory is mounted to /app by default
  • Common cache directories are automatically mounted
  • Environment-specific paths are preserved

Custom Volume Mounts

test:
  image: python:3.9-slim
  volumes:
    .:/app  # Mount current directory
    ~/.cache/pip:/.cache/pip  # Cache directory
    /tmp:/tmp  # Mount system temp
    /data  # Anonymous volume

Environment Variables

Passing Environment Variables

You can pass environment variables to Docker containers:

# Pass environment variables from host
DOMD_ENV=production poetry run domd run test

# Or define them in .dodocker
test:
  environment:
    NODE_ENV: test
    DEBUG: 1
    DATABASE_URL: postgres://user:pass@db:5432/mydb

Environment File

For sensitive data, use an environment file:

test:
  env_file: .env.test
  environment:
    - NODE_ENV=test

Debugging

Debug Mode

To debug Docker-related issues:

# Show Docker commands being executed
DOMD_DEBUG=1 poetry run domd run test

# Get shell in container with the same environment
poetry run domd shell test

Logs and Inspection

# View container logs
docker logs <container_id>

# Inspect container configuration
docker inspect <container_id>

# View running processes
docker top <container_id>

Best Practices

  1. Use Specific Image Tags

    • Prefer specific version tags over latest
    • Example: python:3.9-slim instead of just python

  2. Minimize Image Size

    • Use -slim or -alpine variants when possible
    • Clean up package caches in the same RUN instruction

  3. Cache Dependencies

    • Mount package caches as volumes
    • Use multi-stage builds for production images

  4. Security

    • Never run as root in containers
    • Use .dockerignore to exclude sensitive files
    • Keep images updated with security patches

Troubleshooting

Common Issues

Permission Denied Errors

If you see permission errors with mounted volumes:

  1. Make sure the container user has proper permissions
  2. Set the correct user/group in your Dockerfile:
ARG USER_ID=1000
ARG GROUP_ID=1000

RUN groupadd -g $GROUP_ID appuser && \
    useradd -u $USER_ID -g appuser -m appuser

USER appuser

Missing Dependencies

If a command works locally but fails in Docker:

  1. Check if all required system packages are installed in the image
  2. Verify the working directory is set correctly
  3. Ensure environment variables are properly passed

Network Issues

For network-related problems:

# Check network connectivity inside container
docker run --rm --network host busybox ping google.com

# Inspect container network
docker network inspect bridge

Getting Help

If you encounter issues:

  1. Run with DOMD_DEBUG=1 for verbose output
  2. Check the GitHub Issues
  3. Join our Discord Community for support