Load Testing Tool

A powerful load-testing tool designed to simulate heavy load scenarios via json config through CLI. Logging and summarizing API and DB metrics. Currently supports MongoDB (more databases to be added soon). Contributions are welcome

Features

• Concurrent HTTP Requests: Simulates multiple users sending requests.
• MongoDB Monitoring: Tracks database performance during load tests.
• Logging: Detailed logs for analysis and debugging.
• Customizable Load Scenarios: Configure request rates and payloads.

Installation

Install via npm:

npm install -g load_test_util

Or via yarn:

yarn global add load_test_util

Usage

Basic Example

sample_config.json

{
  "logDir": "load_test_logs",
  "prefix": "load_test",
  "mongoDBUrl": <mongo connection string>,
  "apiUrl": "https://example.com/api",
  "method": "POST",
  "payload": {
    "sample": 1,
    "data": 2
  },
  "startUsers": 200,
  "maxUsers": 2000,
  "incrementBy": 100,
  "durationPerStep": 30,
  "rampUpTime": 5
}

load-test --config /path/to/sample_config.json
Load test started. Logging to: load_test_logs/load_test_2025-04-01_21-35-53-568.txt
Starting load test...

Ramping up to 200 users...

Configuration Options

logDir: Directory to store logs.
prefix: Prefix for log file names.
mongoDBUrl: MongoDB connection string with admin user for capturing serverStatus metrics (optional).
apiUrl: The API endpoint to test.
method: HTTP method (e.g., GET, POST).
payload: JSON payload for the request (used for POST or PUT methods).
startUsers: Number of users to start with.
maxUsers: Maximum number of concurrent users.
incrementBy: Number of users to add in each step.
durationPerStep: Duration (in seconds) for each step.
rampUpTime: Time (in seconds) to wait before ramping up users.

Logging

Logs are stored in logDir. You can customize logging directory via config json.

License

MIT License. See LICENSE for details.