Skip to content

Latest commit

 

History

History
52 lines (35 loc) · 2.04 KB

README.md

File metadata and controls

52 lines (35 loc) · 2.04 KB

Kafka Idempotent Consumer

Introduction

The Kafka Idempotent Consumer project offers a robust solution for consuming messages from Kafka topics idempotently, ensuring that messages are processed exactly once. This project is particularly useful for scenarios where duplicate processing of messages is unacceptable, such as in financial transactions or event logging. It leverages the native kafka-clients library, avoiding the complexity of integrating with major frameworks.

Key Features

  • Framework-Independent: Utilizes kafka-clients for a lightweight, framework-independent implementation.
  • Lifecycle Management: Features a Lifecycle interface for graceful startup and shutdown of system components.
  • Manual Offset Committing: Employs manual offset committing post-message processing for precise consumption control.
  • Idempotency Assurance: Tracks processed messages using an internal data structure to prevent duplicates.

Quick Start with Docker Compose

To simplify the development and testing process, we've included a Docker Compose setup that spins up a three-node Kafka cluster with a single Zookeeper instance and Kafka-UI for easy monitoring. Kafka-UI will be accessible on localhost:8080.

Prerequisites

  • Docker and Docker Compose installed on your machine.

Running the Kafka Cluster

  1. Navigate to the Project Directory: Change into the project directory where the docker-compose.yml file is located.

    cd kafka-idempotent-consumer

  2. Start the Kafka Cluster: Use Docker Compose to start the services defined in docker-compose.yml.

    docker-compose up -d

    This command starts the Kafka cluster, Zookeeper, and Kafka-UI as background services.

  3. Access Kafka-UI: Open your web browser and go to http://localhost:8080 to access Kafka-UI, where you can monitor your Kafka cluster and topics.

  4. Shutting Down: To stop and remove the containers, networks, and volumes created by Docker Compose, run:

    docker-compose down