GopherProxy & Sentinel: A Distributed Service Mesh Lite

PROJECT VERSION: 1.0 - DEC 31, 2025

GopherProxy & Sentinel is a production-grade, health-aware Load Balancer and Service Discovery engine built entirely in Go. It simulates a cloud-native environment locally, featuring a decoupled Data Plane (Proxy) and Control Plane (Sentinel), managed via Redis, and fully observable through Prometheus and Grafana.

Architecture Overview

The system follows a three-tier distributed architecture:

1. The Data Plane (GopherProxy): A high-concurrency Go reverse proxy that routes traffic to healthy backends and enforces security via Rate Limiting.
2. The Control Plane (Sentinel): An automated monitoring agent that pings backend ports and updates the shared Service Registry.
3. The Service Registry (Redis): A persistent "Source of Truth" where backends are registered and de-duplicated.
4. The Observability Stack: Prometheus scrapes metrics from the Proxy, and Grafana visualizes the traffic flow and backend health.

---

Architecture Diagram

Key Features

• Decoupled Service Discovery: Uses Redis Sets to manage dynamic backend registration without restarting the proxy.
• High-Performance Concurrency: Leverages sync.RWMutex for efficient multi-reader access and sync/atomic for thread-safe load balancing.
• Security & Hardening:
  • Rate Limiting: Implements a Token Bucket algorithm to prevent local DDoS.
  • Graceful Shutdown: Handles OS signals (SIGINT, SIGTERM) to ensure zero-drop connection closing.
  • Isolation: Runs as a non-privileged gopheruser inside the container.
• Observability (SRE): Exports custom Prometheus metrics including Request Counters and Healthy Backend Gauges.
• Cloud-Native Optimization: Multi-stage Docker build using ARG and Static Linking, resulting in a tiny 13.4MB image.

---

Tech Stack

• Language: Go (Standard Library, httputil, context)
• Registry: Redis
• Monitoring: Prometheus & Grafana
• Deployment: Docker, Docker Compose
• Networking: HTTP Reverse Proxy, TCP Dialing

---

Getting Started

Prerequisites

• Docker & Docker Compose
• Python (to run mock backend servers)

1. Build and Start the Cluster

docker-compose up --build

2. Start Local Backends (Microservices)

Open two separate terminals:

# Terminal A
python -m http.server 8081
# Terminal B
python -m http.server 8082

3. Generate Traffic

# Observe the Rate Limiter (2 req/sec)
for i in {1..50}; do curl -I http://localhost:8080; done

---

License

This project is licensed under the MIT License - see the LICENSE file for details.