Docker & Deployment

Docker & Deployment

Gova uses a containerized architecture to ensure consistency across development and production environments. The backend consists of a FastAPI application, a PostgreSQL database with vector support, and a Kafka message broker for handling moderation events.

Prerequisites

Before deploying, ensure the following are installed on your host machine:

• Docker (Engine 20.10+)
• Docker Compose (v2.0+)
• PgVector: The database requires the pgvector extension for storing and querying moderation embeddings.

---

Local Development

For development, use the provided docker-compose.yaml to spin up the entire stack including the API, database, and message brokers.

1. Clone the repository and enter the directory:

   git clone https://github.com/JadoreThompson/gova-backend.git
   cd gova-backend
   

2. Configure Environment Variables: Create a .env file in the root directory. See the Configuration section for required variables.

3. Start the services:

   docker-compose up --build
   

   The API will be accessible at http://localhost:8000 by default.

---

Configuration

Gova relies on environment variables to manage third-party integrations (Discord, Stripe, Kafka).

| Variable | Description | | :--- | :--- | | DATABASE_URL | PostgreSQL connection string (must support PgVector). | | KAFKA_BOOTSTRAP_SERVERS | Address of the Kafka broker(s). | | STRIPE_PRICING_PRO_PRICE_ID | The Stripe Price ID for the "Pro" subscription tier. | | DOMAIN / SUB_DOMAIN | The base domain (e.g., gova.chat) and subdomain for callbacks. | | SCHEME | Connection protocol (http or https). | | CUSTOMER_SUPPORT_EMAIL | Destination for messages sent via the Public Contact form. | | ENCRYPTION_KEY | Key used by the EncryptionService to secure OAuth payloads. |

---

Production Deployment

The Dockerfile is a multi-stage build optimized for production. It separates the build environment from the runtime environment to minimize the image size and attack surface.

Build Stages

1. Builder: Installs build dependencies and compiles Python wheels.
2. Runtime: A slim Python image containing only the necessary packages and the src code.

Running the API Runner

The container executes runners/api_runner.py to start the Uvicorn server. In a production environment, you should ensure the following:

• Database Migrations: Run migrations before starting the container to ensure the schema (including uuidv7 defaults and JSONB columns) is up to date.
• Kafka Topics: Ensure the topic defined in KAFKA_MODERATOR_EVENTS_TOPIC exists and has appropriate retention settings.
• SSL/TLS: While the FastAPI app runs via Uvicorn, it is recommended to use a reverse proxy (like Nginx or Traefik) to handle SSL termination.

---

Scaling Components

The system is designed to scale horizontally:

• API Instances: You can run multiple instances of the APIRunner. Load balance them using a round-robin or least-connections strategy.
• Moderation Workers: The moderation logic is event-driven. You can scale workers consuming from the Kafka moderator_events topic to handle higher message volumes across Discord guilds.
• Database: Ensure PostgreSQL is configured to handle the maximum connection pool size defined in your SQLAlchemy async_session settings.