Backend Deployment

Relevant source files

The following files were used as context for generating this wiki page:

• README.md
• backend/main.py
• backend/requirements.txt
• backend/database.py
• backend/api/questions.py
• backend/settings.py

The backend for the QA-Dashboard is a Python-based application built using the FastAPI framework. It is responsible for serving RESTful API endpoints, managing WebSocket connections for real-time updates, handling user authentication, and interacting with a PostgreSQL database. The backend also integrates with AI services for features like answer suggestions.

Deployment involves setting up the required infrastructure (database, Python environment), configuring the application using environment variables, and running it with an ASGI server like Uvicorn. The project documentation suggests deployment targets such as AWS Elastic Beanstalk or a combination of AWS Lambda and API Gateway.

Sources: README.md, backend/main.py

Architecture Overview

The backend is a monolithic service that communicates with a Next.js frontend, a PostgreSQL database, and external AI services.

This diagram illustrates the high-level components and their interactions during deployment.

Prerequisites

Before deploying the backend, the target environment must meet the following prerequisites.

System Requirements

• Python 3.8+: The runtime environment for the FastAPI application.
• PostgreSQL: A relational database is required. The project is configured for AWS RDS PostgreSQL.
• Git: For cloning the repository.

Sources: README.md

Python Dependencies

The application's Python dependencies are listed in requirements.txt. They must be installed in a virtual environment using pip.

Package	Description
fastapi	The core web framework for building the API.
sqlalchemy	ORM for database interaction with PostgreSQL.
psycopg2-binary	PostgreSQL adapter for Python.
python-jose[cryptography]	Implements JWTs for authentication.
passlib[bcrypt]	Used for hashing and verifying passwords.
python-dotenv	Manages environment variables from a .env file.
uvicorn[standard]	ASGI server to run the FastAPI application.
langchain / langchain-google-genai	Framework for integrating with Large Language Models (LLMs).
faiss-cpu	Library for efficient similarity search, used by the AI service.
pydantic[email]	Used for data validation and settings management.

Sources: backend/requirements.txt, README.md

Configuration

The application is configured using environment variables, which are typically stored in a .env file in the backend/ directory for local development. For production deployment, these variables must be set in the hosting environment (e.g., Elastic Beanstalk configuration, Lambda environment variables).

Variable	Description	Example Value
DB_HOST	The endpoint/hostname of the PostgreSQL database.	your-rds-endpoint.rds.amazonaws.com
DB_PORT	The port number for the database connection.	5432
DB_NAME	The name of the database to use.	qa_dashboard
DB_USER	The username for the database connection.	admin
DB_PASSWORD	The password for the database user.	your-secure-password
SECRET_KEY	A secret key for signing JWTs and other security functions.	your-secret-key-change-in-production
DEBUG	Toggles debug mode. Should be False in production.	True
JWT_ALGORITHM	The algorithm used for JWT signing.	HS256
JWT_EXPIRATION_HOURS	The number of hours a JWT token is valid.	24
WEBHOOK_URL	Endpoint for webhook notifications.	https://your-webhook-endpoint.com/webhook
OPENAI_API_KEY	Optional API key for OpenAI integration.	your-openai-key
GOOGLE_API_KEY	API key for Google Generative AI (Gemini).	your-google-api-key

Sources: README.md, backend/services/ai_service.py:15, backend/database.py:12

Database Setup

The backend requires a PostgreSQL database. The schema is defined using SQLAlchemy ORM models.

Database Schema

The database consists of three main tables: users, questions, and answers.

This ER diagram shows the relationships between the database tables.

Initialization

On application startup, the init_db() function is called. This function uses SQLAlchemy's Base.metadata.create_all(bind=engine) to create all defined tables if they do not already exist in the database.

Sources: backend/main.py:14, backend/database.py:83-84

Application Startup

The application is launched via an ASGI server like Uvicorn, pointing to the app instance in main.py.

uvicorn main:app --reload

Sources: README.md

Startup Lifecycle (lifespan manager)

The main.py file defines a lifespan asynchronous context manager that executes tasks during application startup and shutdown.

The following flowchart illustrates the sequence of events during application startup.

Middleware Configuration

The FastAPI application is configured with several middleware components:

• CORSMiddleware: Allows cross-origin requests from specified origins (e.g., http://localhost:3000), which is necessary for the frontend to communicate with the backend.
• TrustedHostMiddleware: Restricts which hosts the application will respond to.

Sources: backend/main.py:48-58

API Routers

The application's endpoints are organized into routers which are included in the main app instance.

• auth.router: Handles user registration and login.
• questions.router: Manages creating, retrieving, and updating questions, and includes the WebSocket endpoint.
• answers.router: Manages creating and retrieving answers.

Sources: backend/main.py:60-62

Deployment Strategies

The README.md file suggests several strategies for deploying the backend to a production environment.

• AWS Elastic Beanstalk: A PaaS offering that automates the deployment, scaling, and management of the application.
• AWS Lambda + API Gateway: A serverless approach where the FastAPI application is wrapped in a handler function and exposed via API Gateway.

In both scenarios, the PostgreSQL database would be hosted on AWS RDS, and the application configuration would be managed through the respective service's environment variable settings.

Sources: README.md