Skip to content

Latest commit

 

History

History
72 lines (47 loc) · 2.81 KB

README.md

File metadata and controls

72 lines (47 loc) · 2.81 KB

API Load Testing Guide

This guide provides instructions on how to perform load testing on various API endpoints using Locust, an open-source load testing tool. The testing script supports multiple environments and use cases for our APIs.

Prerequisites

Before running the tests, ensure you have the following prerequisites:

  • Python 3.x installed
  • Locust installed (pip install locust)
  • An API key for authenticating requests

Setting Up the Environment

  1. API Key: Ensure the API key is stored securely. It's recommended to use an environment variable to store the API key. You can set it in the shell like this:

    export API_KEY='your_actual_api_key_here'

  2. Install Dependencies: If not already installed, you can install Locust using pip:

    pip install locust

Configuration

Edit the api_config.json file to update the API endpoints, paths, and request data for different user classes. You can also update the config file for future use cases.

Running the Tests

Run Locust using the command below, specifying the user class to simulate based on the configurations defined in api_config.json:

  • Without Web UI (headless mode):

    locust -f api_load_tests/api_load_test.py --headless -u <number_of_users> -r <spawn_rate> --run-time <time> --html <report_name>.html --csv <csv_name> <UserClassName>

    Where:

    • <number_of_users> is the number of concurrent users.
    • <spawn_rate> is the rate per second in which users will spawn.
    • <time> is the duration for which the test will run (e.g., 1m for one minute).
    • <report_name>.html is the name of the file to save the HTML report.
    • <csv_name> is the prefix for the CSV files that will store the test results.
    • <UserClassName> is the name of the user class to simulate

    Example:

    locust -f api_load_tests/api_load_test.py --headless -u 60 -r 1 --run-time 2m --html report.html --csv report BetaEmailTypeUser

  • With Web UI:

    Start Locust without the headless flag to open the web interface:

    locust -f api_load_tests/api_load_test.py

    This will start a web server on http://localhost:8089 where you can configure the number of users, spawn rate, and view real-time statistics.

Analyzing the Results

After the test runs, review the HTML report and CSV files for detailed metrics about response times, request rates, and failure rates. This will help in assessing the performance and stability of the API endpoints.

Additional Notes

  • Modify the JSON configuration file as necessary to adapt to new testing scenarios.
  • Ensure the API key and endpoint configurations are up to date before running tests.
  • Use the environment variable API_KEY to securely pass the API credentials during tests.