🏥 MediWatch – Hospital Infrastructure Reliability Platform

A production-grade DevOps project demonstrating automated infrastructure, CI/CD pipelines, monitoring, and self-healing systems for hospital applications.

---

📋 Table of Contents

• Problem Statement
• Solution Overview
• Screenshots
• Tech Stack
• Features
• Project Structure
• Quick Start
• Deployment
• Monitoring
• Testing
• Author
• License

---

🎯 Problem Statement

Hospitals rely on critical systems like patient management, lab processing, and appointment scheduling. However:

• ⏱️ Average downtime: 45+ minutes
• 🚨 Manual failure detection
• 🏥 Delays in patient care
• 📉 No automated recovery

---

💡 Solution Overview

MediWatch provides:

• ✅ Real-time failure detection
• ✅ Auto rollback when error rate > 5%
• ✅ Recovery time reduced to < 2 minutes
• ✅ Zero-downtime deployments

---

📸 Screenshots

🏗️ Architecture Diagram

---

🐳 Docker Services Running

---

📜 Service Logs

---

❤️ Health Checks

Service 1

Service 2

Service 3

---

🔗 API Testing

---

📊 Prometheus Metrics

---

🛠️ Tech Stack

Category	Technologies
Backend	Python 3.11, Flask
Database	PostgreSQL 15
Containerization	Docker, Docker Compose
Orchestration	Kubernetes (K3s)
CI/CD	Jenkins
IaC	Ansible
Monitoring	Prometheus, Grafana
Cloud	AWS (EC2, ECR, S3)
Version Control	Git, GitHub
Testing	pytest

---

✨ Features

🔄 CI/CD Pipeline

• Linting with flake8
• Testing with pytest
• Docker image build & push to AWS ECR
• Kubernetes rolling deployment
• GitHub webhook integration

---

🎯 Microservices Architecture

• Patient Service
• Lab Service
• Appointment Service

Each includes:

• /health
• /metrics

---

📊 Monitoring

• Prometheus (15s scrape interval)
• Grafana dashboards:
  • Uptime
  • Requests/sec
  • P95 latency
  • Pod restarts
• Real-time alerting

---

🔧 Self-Healing

• Auto-remediation script
• Rollback triggers:
  • Error rate > 5%
  • Health check failure
  • High pod restarts

---

⚡ High Availability

• Kubernetes HPA enabled
• CPU-based auto-scaling (70%)
• 1–4 replicas per service
• Zero-downtime rolling updates

---

📁 Project Structure

mediwatch/
├── services/
│   ├── patient-service/
│   │   ├── app.py
│   │   ├── requirements.txt
│   │   ├── Dockerfile
│   │   └── tests/
│   │       └── test_patient.py
│   ├── lab-service/
│   │   ├── app.py
│   │   ├── requirements.txt
│   │   ├── Dockerfile
│   │   └── tests/
│   │       └── test_lab.py
│   └── appointment-service/
│       ├── app.py
│       ├── requirements.txt
│       ├── Dockerfile
│       └── tests/
│           └── test_appointment.py
├── jenkins/
│   └── Jenkinsfile
├── ansible/
│   ├── provision.yml
│   └── inventory.ini
├── kubernetes/
│   ├── patient-deployment.yaml
│   ├── lab-deployment.yaml
│   ├── appointment-deployment.yaml
│   ├── postgres-deployment.yaml
│   ├── configmap.yaml
│   └── monitoring/
│       ├── prometheus-config.yaml
│       ├── prometheus-deployment.yaml
│       ├── grafana-deployment.yaml
│       └── alertmanager-config.yaml
├── remediation/
│   ├── auto-remediation.py
│   └── requirements.txt
├── monitoring/
│   ├── docker-compose.yml
│   └── prometheus.yml
├── scripts/
│   ├── setup.sh
│   ├── health-check.sh
│   ├── rollback.sh
│   ├── cleanup.sh
│   └── init-db.sql
├── docker-compose.yml
└── README.md

This structure follows a microservices-based architecture with clear separation of services, infrastructure, CI/CD, and monitoring.

---

🚀 Quick Start

git clone https://github.com/tanikush/MediWatch
cd MediWatch
docker-compose up -d
cat scripts/init-db.sql | docker exec -i mediwatch-postgres-1 psql -U postgres
docker ps

🔍 Health Check
curl http://localhost:5001/health
curl http://localhost:5002/health
curl http://localhost:5003/health

🌐 Deployment (AWS + Kubernetes)

ansible-playbook ansible/provision.yml -i ansible/inventory.ini
ssh -i ~/.ssh/mediwatch-key.pem ubuntu@YOUR_EC2_IP

git clone https://github.com/tanikush/MediWatch.git
cd MediWatch

docker-compose build

kubectl apply -f kubernetes/
kubectl apply -f kubernetes/monitoring/

📊 Monitoring

Prometheus

http://localhost:9090
http://localhost:30090

Grafana

http://localhost:3000
http://localhost:30300

Login: admin / admin

🧪 Testing

pytest services/*/tests/ -v
docker exec -it mediwatch-patient-service-1 pytest tests/ -v

👩‍💻 Author

Tanisha Kushwah
DevOps Engineer | Cloud Enthusiast

GitHub: https://github.com/tanikush
LinkedIn: https://www.linkedin.com/in/tanisha-kushwah-280944284/

📄 License

This project is licensed under the MIT License.

---

## ✅ Project Complete - April 2026

- 3 Hospital Microservices ✅
- Docker & Docker Compose ✅
- Jenkins CI/CD Pipeline ✅
- Prometheus + Grafana Monitoring ✅
- GitHub Integration ✅

Built by Tanisha Kushwah


---