Why Choose This Project?

Modern cloud-native applications are complex and distributed, making them prone to unexpected failures. Chaos engineering helps teams proactively test system resilience by intentionally injecting failures into production or staging environments. Using tools like Gremlin or Chaos Monkey, this project teaches students how to design fault-tolerant, self-healing systems and ensure high availability.

What You Get

Ability to simulate failures like server crashes, network latency, and CPU/memory spikes
Automated chaos experiments integrated into CI/CD pipelines
Dashboards to monitor system behavior under stress
Alerts for failures, anomalies, and degraded performance
Logging of experiments for audit and analysis

Key Features

Technology Stack

Chaos Engineering Tools:

• Gremlin (SaaS or on-prem) or Chaos Monkey (Netflix OSS)

Infrastructure Layer:

• Kubernetes (EKS, GKE, AKS) or cloud VMs (AWS EC2, Azure, GCP Compute Engine)

• Dockerized microservices

Monitoring Layer:

• Prometheus / Grafana for metrics and dashboards

• CloudWatch / Azure Monitor / GCP Monitoring (optional)

CI/CD Layer (Optional):

• Jenkins / GitLab CI / GitHub Actions

Cloud Services Used

• AWS / Azure / GCP – Host applications or clusters

• Cloud Monitoring – Track system behavior during chaos tests

• Gremlin SaaS – Orchestrate chaos experiments

• Cloud Storage – Store experiment logs and reports

Working Flow

1. Environment Selection – Choose the target cluster, nodes, or services for chaos testing.

2. Define Chaos Experiments – Configure failures to inject: CPU spikes, memory exhaustion, network latency, or pod termination.

3. Execute Experiments – Run chaos tests in a controlled or automated manner.

4. Monitoring & Metrics – Observe metrics, logs, and system health during the experiment.

5. Alerting – Notify teams if performance degradation or failures occur.

6. Rollback / Recovery Verification – Ensure system self-heals or CI/CD triggers rollback mechanisms.

7. Reporting & Analysis – Document results for resilience verification and improvement.

Main Modules

• Chaos Definition Module – Configures failure types, targets, and schedule

• Execution Module – Runs experiments safely and in controlled environments

• Monitoring Module – Collects metrics during chaos experiments

• Alerting Module – Sends notifications for anomalies and degraded performance

• Analysis & Reporting Module – Logs experiment outcomes and generates dashboards

Security Features

• Role-based access control for who can run chaos experiments

• Isolation of chaos tests to prevent unintended production impact

• Audit logging for all chaos actions and results

• Encrypted communication between tool agents and central server