Docker Configuration

This document covers the Docker containerization setup for the signals trading platform. The Docker configuration packages the built application artifacts into a deployable container image for production environments.

For information about the build process that creates the artifacts used by Docker, see Build Process.

Container Architecture

The signals platform uses a single Docker container that packages the entire application stack, including both the backend API server and frontend static assets.

Dockerfile Configuration

The container configuration is defined in the root Dockerfile which implements a multi-stage approach optimized for production deployment.

Base Image and Runtime

The container uses the official Node.js 22 Alpine Linux image as its base runtime:

Configuration	Value	Purpose
Base Image	node:22-alpine	Lightweight Node.js runtime
Working Directory	/app	Application root directory
Additional Tools	curl	Health check and debugging utility

File System Layout

The container copies specific directories from the build context:

Build Context Exclusions

The .dockerignore file ensures only production-ready artifacts are included in the Docker build context:

Excluded Directory	Reason
src	Source code (already compiled to build/)
test	Test files not needed in production
scripts	Build scripts not needed at runtime
apps	Frontend source code (already built to build/)
tools	Development utilities
docs	Documentation files
docker	Docker-related files
config	Build configuration files

This exclusion strategy minimizes the Docker build context size and ensures no development artifacts leak into production containers.

Container Deployment Process

The Docker deployment process involves building the image and publishing it to a container registry.

Platform-Specific Publishing Scripts

The platform provides publishing scripts for different operating systems:

Windows Publishing (scripts/win/publish.bat:1-3):

• Builds Docker image with tag tripolskypetr/signal-app
• Pushes to container registry using Docker Hub

Linux Publishing (scripts/linux/publish.sh:1-3):

• Identical functionality to Windows script
• Uses bash shell for Linux/macOS environments

Both scripts use the same Docker image tag and registry destination, ensuring consistent deployment across platforms.

Runtime Configuration

Application Startup

The container starts the application using the Node.js runtime with the compiled backend bundle:

CMD ["node", "./build/index.mjs"]
Copy

This command launches the main application entry point, which includes:

• Hono web server initialization
• IoC container setup and service registration
• Signal processing job scheduling
• Frontend asset serving

Container Health and Monitoring

The container includes curl utility installation for health check capabilities:

Integration with Build System

The Docker configuration integrates seamlessly with the build system documented in Build Process:

1. Build Phase: The copy-build.ts script consolidates all application artifacts into the build/ directory
2. Docker Phase: The Dockerfile copies these pre-built artifacts into the container
3. Runtime Phase: The container serves both the API backend and frontend applications

This separation ensures that Docker containers contain only production-ready, optimized code without any development dependencies or source files.