# Incident Management Platform

A distributed incident and change management platform built using **Python, Spring Boot, React, and Docker**.

This project focuses on **real-world backend engineering problems** such as
incident detection, error analysis, trend tracking, and safe automation with
human approval.

The goal is to demonstrate **clean architecture, service boundaries, and production-style workflows**, not toy examples.

---

## 🧱 System Architecture

UI (React) ↓ API (Spring Boot) ↓ Engine (Python / Flask)

Each service is independently deployable and communicates over HTTP.
All components are containerized and orchestrated using Docker Compose.

---

## 🔧 Tech Stack

- **Frontend:** React
- **Backend API:** Java, Spring Boot
- **Incident Engine:** Python, Flask
- **Infrastructure:** Docker, Docker Compose

---

## 🧠 Key Capabilities

- Asynchronous incident processing (non-blocking APIs)
- Structured error classification
- Confidence scoring for diagnosis reliability
- Error frequency and trend tracking
- Dependency and version conflict detection
- Approval-based decision flow for safe automation
- Clear separation of responsibilities across services

---

## ▶️ Running the Project Locally

### Prerequisites
- Docker
- Docker Compose

### Start all services
```bash
docker compose up --build

Services

• UI: http://localhost:3000
• API: http://localhost:8080
• Engine: http://localhost:5001

---

🧪 Example API Calls

Approve an Incident (API Service)

curl -X POST http://localhost:8080/incident/approve \
  -H "Content-Type: application/json" \
  -d '{"id":"INC-101"}'

Submit Incident for Analysis (Engine Service)

curl -X POST http://localhost:5001/analyze \
  -H "Content-Type: application/json" \
  -d '{
    "error_type": "ZeroDivisionError",
    "message": "division by zero",
    "stacktrace": ["File calc.py line 10 return a / b"]
  }'

The engine processes incidents asynchronously and logs the analysis result internally for audit and follow-up actions.

---

🛡 Design Principles

• Clear service boundaries
• Defensive automation (no blind fixes)
• Human-in-the-loop approvals
• Stateless APIs with stateful analysis tracking
• Simple, extensible error knowledge base
• Production-oriented defaults

---

📂 Project Structure

services/
├── api/        # Spring Boot backend
├── engine/     # Python incident analysis engine
└── ui/         # React frontend

---

🚀 Planned Enhancements

• Email-based approval workflow
• Persistent storage using PostgreSQL
• Authentication and authorization
• Event-driven communication (Kafka / RabbitMQ)
• Metrics, monitoring, and observability

---