🛡️ The Sentinel Stack (Full-Stack Observability)

> A microservices-style stack focused on monitoring and observability.

---

👁️ Project Overview

In modern DevOps, "deploying" is only half the battle. The real challenge lies in Observability: understanding why a system behaves the way it does.

The Sentinel Stack is an advanced microservices architecture that extends a standard web stack (Flask/Redis/Nginx) with a robust Metric Monitoring Layer. It focuses on Metric-based Observability by utilizing Prometheus for scraping time-series data and Grafana for high-fidelity visualization. This helps avoid the "black box" problem by exposing metrics like request latency, error rates, and traffic patterns.

---

🏗 Architecture & Data Flow

The system can be viewed in two parts: request handling and metric collection.

graph TD;
    subgraph "Traffic Plane"
        Client((User)) -->|HTTP:80| Nginx[Nginx Gateway];
        Nginx -->|Proxy Pass| Web[Flask App];
        Web <-->|Read/Write| Redis[(Redis DB)];
    end
    
    subgraph "Observability Plane"
        Prometheus[Prometheus Server] -- Scrape /metrics --> Web;
        Grafana[Grafana Dashboard] -- Query Data --> Prometheus;
    end

Loading

1. Ingress: Requests hit Nginx (Reverse Proxy).
2. Processing: Flask handles logic and exposes system metrics at /metrics.
3. Persistence: Redis maintains stateful data (visitor counts).
4. Collection: Prometheus scrapes metrics from Flask every 5 seconds.
5. Visualization: Grafana queries Prometheus to display real-time dashboards.

---

📊 Dashboard Preview

(This dashboard visualizes Request Rate, Error Rate, and Latency in real-time)

Note: The dashboard above tracks HTTP Status Codes, Response Times, and Redis Connectivity.

---

✨ Key Features

• Real-time Monitoring: Automatically tracks application performance, including Request Per Second (RPS) and 95th Percentile Latency.
• Metric-Driven Insights: Exposes metrics that make performance issues easier to spot.
• Containerized Ecosystem: The entire stack, including the monitoring infrastructure, is defined as code (docker-compose.yml) for reproducible deployments.
• Alerting Ready: Structure is prepared for setting up Prometheus Alertmanager rules (e.g., trigger alert if Error Rate > 5%).

---

🛠 Tech Stack

Component	Role	Technology
Orchestrator	Infrastructure Management	Docker Compose
Gateway	Reverse Proxy & Load Balancing	Nginx (Alpine)
Backend	Application Logic & Metric Exporter	Python Flask
Database	Key-Value Store	Redis
Collector	Time-Series Database & Scraping	Prometheus
Visualizer	Analytics & Dashboarding	Grafana

---

🚀 Setup & Execution Guide

1. Prerequisites

Ensure you have Docker and Docker Compose installed on your machine.

2. Launch the Stack

Clone the repository and spin up the containers:

git clone https://github.com/swnst/the-sentinel-stack.git
cd the-sentinel-stack
docker-compose up -d --build

3. Verify Services

Check if all containers are healthy:

docker-compose ps

4. Access the Interfaces

Service	URL	Credentials (Default)
Web App	http://localhost:80	N/A
Prometheus	http://localhost:9090	N/A
Grafana	http://localhost:3000	User: admin / Pass: admin

---

📂 Project Structure

.
├── app.py               # Flask App with Prometheus Exporter initialized
├── docker-compose.yml   # Full stack definition (5 Services)
├── Dockerfile           # Python application container spec
├── nginx.conf           # Reverse proxy configuration
├── prometheus.yml       # Prometheus scraping rules
├── requirements.txt     # Python dependencies
└── README.md            # Documentation