CDC with Debezium, Kafka, Postgres, Docker

Overview

This Python script is designed to generate simulated financial transactions and insert them into a PostgreSQL database. It's particularly useful for setting up a test environment for Change Data Capture (CDC) with Debezium. The script uses the faker library to create realistic, yet fictitious, transaction data and inserts it into a PostgreSQL table.

System Architecture

Prerequisites

Before running this script, ensure you have the following installed:

• Python 3.9+
• psycopg2 library for Python
• faker library for Python
• PostgreSQL server running locally or accessible remotely
• Docker and Docker Compose installed on your machine.
• Basic understanding of Docker, Kafka, and Postgres.

Installation

1. Install Required Python Libraries:

   You can install the required libraries using pip:

   pip install psycopg2-binary faker
   

Services in the Compose File

• Zookeeper: A centralized service for maintaining configuration information, naming, providing distributed synchronization, and providing group services.
• Kafka Broker: A distributed streaming platform that is used here for handling real-time data feeds.
• Confluent Control Center: A web-based tool for managing and monitoring Apache Kafka.
• Debezium: An open-source distributed platform for change data capture.
• Debezium UI: A user interface for managing and monitoring Debezium connectors.
• Postgres: An open-source relational database.

Getting Started

1. Clone the Repository: Ensure you have this Docker Compose file in your local system. If it's part of a repository, clone the repository to your local machine.

2. Navigate to the Directory: Open a terminal and navigate to the directory containing the Docker Compose file.

3. Run Docker Compose: Execute the following command to start all services defined in the Docker Compose file:

   docker-compose up -d
   

   This command will download the necessary Docker images, create containers, and start the services in detached mode.

4. Verify the Services: Check if all the services are up and running:

   docker-compose ps
   

   You should see all services listed as 'running'.

5. Accessing the Services:

   • Kafka Control Center is accessible at http://localhost:9021.
   • Debezium UI is accessible at http://localhost:8080.
   • Postgres is accessible on the default port 5432.

6. Shutting Down: To stop and remove the containers, networks, and volumes, run:

   docker-compose down
   

Customization

You can modify the Docker Compose file to suit your needs. For example, you might want to persist data in Postgres by adding a volume for the Postgres service.

Note

This setup is intended for development and testing purposes. For production environments, consider additional factors like security, scalability, and data persistence.