Skip to content

TESTING.md

Latest commit

 

History

History
201 lines (144 loc) · 4.76 KB

TESTING.md

File metadata and controls

201 lines (144 loc) · 4.76 KB

API Testing Guide

This document provides an overview of the testing strategy and instructions for running different types of tests for the Cognitive Mesh Convener Backend API.

Table of Contents

Test Types

  1. Unit Tests: Test individual components in isolation
  2. Integration Tests: Test interactions between components
  3. Performance Tests: Test API performance under load
  4. Concurrency Tests: Test handling of concurrent requests
  5. Rate Limiting Tests: Test API rate limiting functionality

Prerequisites

  • Node.js 16+
  • npm or yarn
  • k6 (for performance testing)
  • Access to a running instance of the API

Running Tests

Install Dependencies

npm install

Run Unit Tests

npm test

Run Performance Tests

Run all performance tests:

npm run test:performance

Run specific test scenarios:

# Large payload tests
k6 run tests/performance/large-payloads.test.js

# Concurrency tests
k6 run tests/performance/concurrency.test.js

# Rate limiting tests
k6 run tests/performance/rate-limiting.test.js

Environment Variables

Configure tests using environment variables:

export BASE_URL=http://localhost:3000/v1
export AUTH_TOKEN=your-auth-token
export RATE_LIMIT=100  # requests per minute

Test Descriptions

Unit Tests

  • Location: __tests__/unit/
  • Purpose: Test individual functions and components in isolation
  • Coverage:
    • Input validation
    • Business logic
    • Error handling
    • Utility functions

Integration Tests

  • Location: __tests__/integration/
  • Purpose: Test API endpoints and their interactions
  • Coverage:
    • API endpoints
    • Database interactions
    • External service integrations
    • Authentication and authorization

Performance Testing

Large Payload Tests

Tests the API's ability to handle requests with large payloads.

Scenarios:

  • Small payloads (10KB)
  • Medium payloads (100KB)
  • Large payloads (1MB+)

Metrics Tracked:

  • Response times
  • Success rates
  • Error rates
  • Memory usage

Concurrency Tests

Tests how the API handles multiple simultaneous requests.

Scenarios:

  • Low concurrency (10 concurrent users)
  • Medium concurrency (50 concurrent users)
  • High concurrency (100+ concurrent users)

Metrics Tracked:

  • Requests per second
  • Error rates
  • Response time percentiles
  • System resource usage

Rate Limiting Tests

Tests the API's rate limiting functionality.

Scenarios:

  • Requests within rate limit
  • Requests exceeding rate limit
  • Rate limit reset behavior
  • Multiple clients with different rate limits

Metrics Tracked:

  • Rate limit headers
  • 429 responses
  • Retry-after headers
  • Successful vs. rate-limited requests

Test Data

Test data is generated dynamically for most tests. For specific test cases, you can find test data in the test-data/ directory.

Continuous Integration

Tests are automatically run on pull requests and merges to the main branch. The CI pipeline includes:

  1. Linting
  2. Unit tests
  3. Integration tests
  4. Performance tests (on schedule)
  5. Security scanning

Best Practices

  1. Isolate Tests: Each test should be independent and not rely on the state from other tests.
  2. Clean Up: Always clean up test data after tests complete.
  3. Use Mocks: Mock external services to make tests more reliable and faster.
  4. Test Edge Cases: Include tests for edge cases and error conditions.
  5. Performance Baselines: Establish performance baselines and monitor for regressions.
  6. Documentation: Keep test documentation up to date.
  7. Security: Include security tests in your test suite.

Troubleshooting

Common Issues

  1. Tests Failing:

    • Check the test logs for specific error messages
    • Verify that all services are running
    • Check for network connectivity issues

  2. Performance Test Failures:

    • Check server resource usage
    • Verify network latency
    • Check for external service dependencies

  3. Rate Limiting Issues:

    • Verify rate limit headers in responses
    • Check for proper 429 responses
    • Verify rate limit reset behavior

Getting Help

If you encounter issues not covered in this guide, please:

  1. Check the project's issue tracker
  2. Review the API documentation
  3. Contact the development team

License

This testing documentation is part of the Cognitive Mesh project and is licensed under the MIT License.