import chaos

Testing is a tricky business. Testing full stack apps is even trickier. You have to deal with frontend, backend, database, network, and more. First, of course, you unit test your components, functions, and modules in isolation. Then you write integration tests to ensure they play nicely together. You might even add a few end-to-end tests for the entire application to simulate real user interactions.

My new article as a freeCodeCamp News author just got published!

Modern systems often generate vast streams of high-volume, high-cardinality, or continuously changing data. Performing exact queries - like counting unique users on a website, tracking event frequencies, or checking membership in massive sets - can be slow, memory-intensive, or even infeasible on traditional relational or NoSQL databases.

Chaos testing usually brings Netflix or Amazon to mind: production services deliberately stressed to find weaknesses. For smaller teams, or solo developers, that kind of chaos feels extreme or impractical.

Introducing chaos-fetch: a TypeScript/ESM library for injecting network chaos (latency, failures, drops, etc.) into fetch requests. Designed for programmatic use, chaos-fetch provides a flexible middleware system for simulating adverse network conditions in client-side code.