Incident Simulator
Platform for simulating real production incidents.
Screenshots
The Problem
Engineers rarely experience realistic system outages during development, making them less prepared for real production failures.
The Solution
Incident Simulator creates controlled outages such as latency spikes and database failures.
Implementation Details
Theory is one thing; debugging a live system while the alerts are firing is another. The Incident Simulator is a safe space to fail.
Controlled Chaos
The 플랫폼 uses custom Docker images that can be "poisoned" via an API call. You can inject high latency, simulate 500 errors, or kill the primary database node to see how the system (and you) reacts.
Game Days
I've used this tool to run "SRE Game Days" where teams have to diagnose and fix a simulated outage within a specific time limit. It's been instrumental in improving incident response times.