🛡️ Auto-Healing Microservices System

A self-healing microservice architecture with Docker, Flask, Prometheus, and Grafana

📦 System Architecture

🚀 Quick Start

# Clone and deploy
git clone https://github.com/Dasmat13/autohealing-microservices.git
cd autohealing-microservices
docker-compose up -d --build

🌐 Access Endpoints

Component	URL	Credentials
Service A	http://localhost/a	-
Service B	http://localhost/b	-
Prometheus	http://localhost:9090	-
Grafana	http://localhost:3000	admin:admin

🔧 Key Features

• Self-healing: Watchtower auto-restarts failed containers
• Monitoring: Prometheus scrapes metrics every 15s
• Visualization: Pre-configured Grafana dashboards
• Load Balancing: Nginx distributes traffic

📊 Grafana & Prometheus — Quick Setup Notes

1. Open Grafana: http://localhost:3000 (admin/admin)
2. Add Prometheus data source: URL → http://prometheus:9090 and click Save & Test
3. Import dashboards:
   • Docker metrics dashboard: Grafana dashboard ID 893 (requires cadvisor scraping)
   • Or import custom dashboard JSON (service metrics) — see grafana/flask-dashboard.json
4. If Grafana shows "No data": verify Prometheus targets at http://localhost:9090/targets

📷 Prometheus Dashboard Screenshot

📷 Grafana Dashboard Screenshot

🛠️ Troubleshooting

Common fixes:

# Rebuild single service
docker-compose up -d --build service-a

# Check container health
docker ps --filter "health=unhealthy"

📚 Documentation

• Prometheus Configuration
• Flask Service Setup

📜 License

MIT © 2023 Dasmat13