chaos-engineering
Installation
SKILL.md
Chaos Engineering
Purpose
Design and execute chaos engineering experiments to validate system resilience using fault injection, steady-state hypothesis testing, and controlled blast radii.
Architecture Decision Trees
Tool Selection: Chaos Mesh vs Litmus vs Gremlin vs AWS FIS
| Tool | Platform | Installation | Experiment Types | Cost |
|---|---|---|---|---|
| Chaos Mesh | Kubernetes | Helm, easy | Pod kill, network, stress, DNS, HTTP | Free (CNCF) |
| Litmus | Kubernetes | Operator + CLI | Pod/container kill, network, node drain, chaos charts | Free (CNCF) |
| Gremlin | Multi-platform | Agent-based | CPU, memory, IO, network, DNS, process, shutdown | Paid (freemium) |
| AWS FIS | AWS-only | AWS service | EC2 stop, RDS failover, ECS task stop, ASG | Pay-per-action |
| Toxiproxy | App-level | Sidecar/proxy | Network latency, disconnects, throttling | Free |
| Pumba | Docker | Container | Container kill, network, pause | Free |