Dgraph Dg-Auth-Proxy

A lightweight authentication proxy for Dgraph databases that adds an authentication layer between clients and your Dgraph instance.

Features

• Acts as a secure middleware between clients and Dgraph
• Simple configuration via environment variables
• Docker-ready with multi-platform support
• Built with modern Node.js (using H3 for HTTP handling)

Prerequisites

• Node.js (v18+ recommended)
• Yarn (v4+)
• Docker (for containerized deployment)
• A Dgraph instance (local or remote)

Installation

Local Development

1. Clone the repository:

   git clone https://github.com/ezyostudio/dgraph-dg-auth-proxy.git
   cd dgraph-dg-auth-proxy

2. Install dependencies:

   yarn install

3. Create a .env file with your Dgraph configuration:

   DGRAPH_URL=https://your-dgraph-instance.url
   DGRAPH_AUTH_TOKEN=your-dgraph-auth-token

4. Start the development server:

   yarn dev

Docker Deployment

The included docker-compose.yml sets up:

• Dgraph Zero service
• Dgraph Alpha service
• This auth proxy service

1. Replace the placeholder tokens in docker-compose.yml with your actual tokens:

• your-admin-token (Dgraph admin API key)
• your-client-token (Dgraph client API key, also set in proxy environment)

2. Start the services:

   docker-compose up -d

Configuration

Environment Variables

Variable	Description	Required	Default
PORT	Port to listen on	No	80
HOST	Host to bind to	No	0.0.0.0
DGRAPH_URL	URL to your Dgraph instance	Yes	-
DGRAPH_AUTH_TOKEN	Authentication token for Dgraph	Yes	-

Deployment

Docker Image

The project includes a GitHub Actions workflow that automatically builds and pushes a Docker image to GitHub Container Registry when a new release is published.

To manually build and push:

2. Pull the image

   docker pull ghcr.io/ezyostudio/dgraph-dg-auth-proxy:latest
   docker run -p 3000:3000 -e PORT=3000 -e DGRAPH_URL=http://<dgraph-alpha-url> -e DGRAPH_AUTH_TOKEN=<client-token> ghcr.io/ezyostudio/dgraph-dg-auth-proxy

Cloudflare Workers

The project includes a wrangler.toml configuration for deploying to Cloudflare Workers. To deploy:

1. Install Wrangler:

   npm install -g wrangler

2. Deploy:

   wrangler deploy

Development

• Dev server: yarn dev (watches for changes)
• Production server: yarn start

Dependabot

The project uses Dependabot to automatically check for dependency updates daily for both npm packages and GitHub Actions.

License

MIT