Polling System with Kafka, Zookeeper, WebSockets, and PostgreSQL

This project is a polling system that supports multiple users voting on polls in real-time, with a resilient design to ensure no votes are lost, even in case of failures. The system utilizes Kafka and Zookeeper for messaging and fault tolerance, PostgreSQL for data storage, and WebSockets for real-time updates.

Features

• Create polls with options
• Vote on polls and store votes in Kafka
• Real-time poll updates via WebSockets
• Leaderboard feature to show the most voted options across all polls
• Fault-tolerant voting using Kafka and Zookeeper

Setup Instructions

1. Clone the repository

git clone https://github.com/rudra016/backend-assignment-abhiman.git
cd backend-assignment-abhiman

2. Install Dependencies

npm install

3. Set Up PostgreSQL Database

Create a database in postgresql named "polling" in pgAdmin, after creating run this query

-- Create polls table
CREATE TABLE polls (
    id SERIAL PRIMARY KEY,
    question TEXT NOT NULL,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

-- Create poll_options table
CREATE TABLE poll_options (
    id SERIAL PRIMARY KEY,
    poll_id INT REFERENCES polls(id) ON DELETE CASCADE,
    option_text TEXT NOT NULL,
    vote_count INT DEFAULT 0,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

-- Create votes table
CREATE TABLE votes (
    id SERIAL PRIMARY KEY,
    poll_id INT REFERENCES polls(id) ON DELETE CASCADE,
    option_id INT REFERENCES poll_options(id) ON DELETE CASCADE,
    created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP
);

-- Leaderboard view: Aggregates vote counts for each option across all polls
CREATE VIEW leaderboard AS
SELECT 
    p.id AS poll_id,
    o.option_text,
    o.vote_count,
    RANK() OVER (ORDER BY o.vote_count DESC) AS rank
FROM 
    polls p
JOIN 
    poll_options o ON p.id = o.poll_id
ORDER BY 
    o.vote_count DESC;

then, run this seperately from the above code.

ALTER TABLE polls ADD COLUMN options JSONB;

4. Edit db.js

// db.js
const { Pool } = require("pg");

const pool = new Pool({
  user: "postgres",
  host: "localhost",
  database: "polling",
  password: "your_password",
  port: 5432,
});

module.exports = pool;

5. Set Up Kafka with Zookeeper

# Start Zookeeper
zookeeper-server-start.sh config/zookeeper.properties

# Start Kafka
kafka-server-start.sh config/server.properties

In kafla folder, navigate to .\bin\windows\ and run this command in cmd

.\kafka-topics.bat --create --topic votes --bootstrap-server localhost:9092 --partitions 3 --replication-factor 1

6. Start the server

npm start

7. Testing APIs

Please connect your postman throught the localhost by postman desktop agent, then you'll be able to test the api's. (Postman doc is attached on top)

/polls -> create polls.
/polls/:id/vote -> vote on poll using specific id of poll.
/polls/:id -> get the votes on a specific polls.
/leaderboard/ -> get details of best polls with count.