JwtSecurity.Lab

A distributed .NET lab demonstrating secure JWT authentication, stateful TCP connection management, and real-time observability.

---

🏗️ Architecture

The system is divided into two primary domains communicating via a message bus (MassTransit/RabbitMQ):

• Validator.Server: A high-concurrency TCP gateway. It manages stateful device connections using ConcurrentDictionary, performs JWT validation, and emits telemetry events.
• Analytics.Worker: An event-driven service that consumes authentication events and updates real-time metrics (Prometheus/Grafana).
• Simulator: A multi-threaded, distributed telemetry simulator that tests both valid traffic and security-sensitive scenarios.

---

🛠️ Key Technical Features

• Stateful Connection Management: Uses ConcurrentDictionary for thread-safe session tracking, ensuring accurate ActiveConnection monitoring without race conditions.
• Event-Driven Observability: Decouples validation logic from telemetry collection using an asynchronous message bus, ensuring that monitoring overhead never impacts connection latency.
• Real-time Metrics Pipeline: Implements the Gauge vs. Counter pattern to provide both instantaneous system snapshots (e.g., current active users) and historical throughput analysis (e.g., auth attempts).
• Resilience & Graceful Shutdown: Robust error handling and lifecycle management that ensures metrics are always consistent and connections are disposed of cleanly.

---

🏗 Architecture Diagram

     flowchart LR
    subgraph Client ["Client Layer"]
        dev[Devices / Simulator]
    end

    subgraph Gateway ["Gateway Layer"]
        val[Validator.Server]
    end

    subgraph Bus ["Communication Layer"]
        mq[RabbitMQ / MassTransit]
    end

    subgraph Analytics ["Analytics Layer"]
        an[Analytics.Worker]
    end

    subgraph Mon ["Observability Stack"]
        prom[(Prometheus)]
        graf[Grafana Dashboard]
    end

    dev -->|TCP/JWT| val
    val -->|Publish Events| mq
    mq -->|Consume| an
    an -->|Metrics| prom
    prom -->|Query| graf

Loading

---

📊 Observability Dashboard (Grafana)

The dashboard is designed to provide a comprehensive view of system health, utilizing three distinct visualization patterns:

• Stats Panels: For immediate status of throughput and active connections.
• Time Series: To track trends and traffic spikes over time.
• Pie Chart: To identify root causes by grouping failure reasons (e.g., token expiration vs. signature mismatch).

Metrics Exposed:

Metric	Type	Purpose
active_connections_total	Gauge	Tracks real-time stateful TCP connections
auth_success_total	Counter	Monitors overall successful throughput.
auth_failures_total	Counter	Categorized by reason and device_id for security auditing.
auth_attempts_total	Counter	Tracks total system load and traffic volume.

---

🚀 Getting Started

1. Start the Infrastructure: Ensure you have RabbitMQ, Prometheus, and Grafana running. A docker-compose.yml is provided in the root to spin these up instantly:

docker compose up -d

2. Launch the Services: Run the following applications locally

   • Run the Validator.Server to start the TCP gateway.
   • Run the Analytics.Worker to begin processing events.

3. Run the Simulator: Launch Simulator to inject traffic and observe the dashboard in real-time..

---

Built with: NET 10 | MassTransit | RabbitMQ | Prometheus | Grafana | Clean Architecture | Async-First Design