Complete MQTT v5.0 Platform

🚀 A complete MQTT v5.0 platform featuring both high-performance client library AND full-featured broker implementation - pure Rust, zero unsafe code

This project provides everything you need for MQTT v5.0 development:

Production-ready MQTT v5.0 broker (Mosquitto replacement)
High-performance client library with AWS IoT compatibility
Multiple transport support (TCP, TLS, WebSocket)
Comprehensive testing with network simulation and property-based testing

🏗️ Dual Architecture: Client + Broker

Component	Use Case	Key Features
MQTT Broker	Run your own MQTT infrastructure	TLS, WebSocket, Authentication, Bridging, Monitoring
MQTT Client	Connect to any MQTT broker	AWS IoT compatible, Auto-reconnect, Mock testing

📦 Installation

Library Crate

[dependencies]
mqtt5 = "0.4.1"

CLI Tool

cargo install mqttv5-cli

🚀 Quick Start

Start an MQTT Broker

use mqtt5::broker::{BrokerConfig, MqttBroker};

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    // Create broker with default configuration
    let mut broker = MqttBroker::bind("0.0.0.0:1883").await?;
    
    println!("🚀 MQTT broker running on port 1883");
    
    // Run until shutdown
    broker.run().await?;
    Ok(())
}

Connect a Client

use mqtt5::{MqttClient, QoS};

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let client = MqttClient::new("my-device-001");
    
    // Connect to your broker (or any MQTT broker)
    client.connect("mqtt://localhost:1883").await?;
    
    // Subscribe with callback
    client.subscribe("sensors/+/data", |msg| {
        println!("📧 {}: {}", msg.topic, String::from_utf8_lossy(&msg.payload));
    }).await?;
    
    // Publish a message
    client.publish("sensors/temp/data", b"25.5°C").await?;
    
    tokio::time::sleep(tokio::time::Duration::from_secs(5)).await;
    Ok(())
}

🚀 Command Line Interface (mqttv5)

Superior CLI tool that replaces mosquitto_pub, mosquitto_sub, and mosquitto with unified ergonomics:

Installation

# Install from crates.io
cargo install mqttv5-cli

# Or build from source
git clone https://github.com/fabriciobracht/mqtt-lib
cd mqtt-lib
cargo build --release -p mqttv5-cli

Usage Examples

# Start a broker (replaces mosquitto daemon)
mqttv5 broker --host 0.0.0.0:1883

# Publish a message (replaces mosquitto_pub)
mqttv5 pub --topic "sensors/temperature" --message "23.5"

# Subscribe to topics (replaces mosquitto_sub)  
mqttv5 sub --topic "sensors/+" --verbose

# Smart prompting when arguments are missing
mqttv5 pub
# ? MQTT topic › sensors/
# ? Message content › Hello World!
# ? Quality of Service level › ● 0 (At most once)

Key CLI Advantages

🎯 Unified interface - One command instead of mosquitto_pub/mosquitto_sub/mosquitto
🧠 Smart prompting - Guides users instead of showing walls of help text
✅ Input validation - Catches errors early with helpful suggestions
📝 Descriptive flags - --topic instead of -t, with short aliases available
🔄 Interactive & non-interactive - Works great for both humans and scripts

🎯 Why This Platform?

✅ Production-Ready Broker

Mosquitto replacement with better performance and memory usage
Multiple transports: TCP, TLS, WebSocket in a single binary
Built-in authentication: Username/password, file-based, bcrypt
Resource monitoring: Connection limits, rate limiting, memory tracking
Self-contained: No external dependencies (Redis, PostgreSQL, etc.)

✅ High-Performance Client

Pure Rust implementation: No FFI, no unsafe code
AWS IoT compatibility: Works seamlessly with AWS IoT Core
Zero-copy operations: Efficient memory usage with BeBytes
Direct async/await: Clean Rust async patterns
Comprehensive testing: Property-based tests and network simulation

📦 Broker Features

Core MQTT v5.0 Broker

✅ Full MQTT v5.0 compliance - All packet types, properties, reason codes
✅ Multiple QoS levels - QoS 0, 1, 2 with proper flow control
✅ Session persistence - Clean start, session expiry, message queuing
✅ Retained messages - Persistent message storage and retrieval
✅ Shared subscriptions - Load balancing across multiple clients
✅ Will messages - Last Will and Testament (LWT) support

Transport & Security

✅ TCP transport - Standard MQTT over TCP on port 1883
✅ TLS/SSL transport - Secure MQTT over TLS on port 8883
✅ WebSocket transport - MQTT over WebSocket for browsers/firewalls
✅ Certificate authentication - Client certificate validation
✅ Username/password authentication - File-based user management

Advanced Features

✅ Broker-to-broker bridging - Connect multiple broker instances
✅ Resource monitoring - $SYS topics, connection metrics, rate limiting
✅ Hot configuration reload - Update settings without restart
✅ Storage backends - File-based or in-memory persistence
✅ ACL (Access Control Lists) - Fine-grained topic permissions

Performance & Scalability

✅ High concurrency - Handle 10,000+ concurrent connections
✅ Connection pooling - Efficient resource reuse
✅ Optimized routing - Fast topic matching and message delivery
✅ Memory monitoring - Prevent resource exhaustion attacks
✅ Rate limiting - Per-client message and bandwidth limits

📦 Client Features

Core MQTT v5.0 Client

✅ Full MQTT v5.0 protocol compliance - All MQTT 5.0 features implemented
✅ Callback-based message handling - Simple, intuitive API with automatic message routing
✅ AWS IoT SDK Compatible - Subscribe returns (packet_id, qos) like Python paho-mqtt
✅ Automatic reconnection - Built-in exponential backoff and session recovery
✅ Client-side message queuing - Handles offline scenarios gracefully

Transport & Connectivity

✅ Certificate loading from bytes - Load TLS certificates from memory (PEM/DER formats)
✅ WebSocket transport - MQTT over WebSocket for browsers and firewall-restricted environments
✅ TLS/SSL support - Secure connections with certificate validation
✅ Session persistence - Survives disconnections with clean_start=false

Testing & Development

✅ Mockable Client Interface - MqttClientTrait enables testing without real brokers
✅ Comprehensive property testing - 29 property-based tests ensuring robustness
✅ Flow control - Respects broker receive maximum limits
✅ Zero-copy message handling - Efficient memory usage with BeBytes

🚦 Advanced Broker Configuration

Multi-Transport Broker

use mqtt_v5::broker::{BrokerConfig, TlsConfig, WebSocketConfig};

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let config = BrokerConfig::default()
        // TCP on port 1883
        .with_bind_address("0.0.0.0:1883".parse()?)
        // TLS on port 8883
        .with_tls(
            TlsConfig::new("certs/server.crt".into(), "certs/server.key".into())
                .with_ca_file("certs/ca.crt".into())
                .with_bind_address("0.0.0.0:8883".parse()?)
        )
        // WebSocket on port 8080
        .with_websocket(
            WebSocketConfig::default()
                .with_bind_address("0.0.0.0:8080".parse()?)
                .with_path("/mqtt")
        )
        .with_max_clients(10_000);

    let mut broker = MqttBroker::with_config(config).await?;
    
    println!("🚀 Multi-transport MQTT broker running:");
    println!("  📡 TCP:       mqtt://localhost:1883");
    println!("  🔒 TLS:       mqtts://localhost:8883");
    println!("  🌐 WebSocket: ws://localhost:8080/mqtt");
    
    broker.run().await?;
    Ok(())
}

Broker with Authentication

use mqtt_v5::broker::{BrokerConfig, AuthConfig, AuthMethod};

#[tokio::main]
async fn main() -> Result<(), Box<dyn std::error::Error>> {
    let auth_config = AuthConfig {
        allow_anonymous: false,
        password_file: Some("users.txt".into()),
        auth_method: AuthMethod::Password,
        auth_data: None,
    };
    
    let config = BrokerConfig::default()
        .with_bind_address("0.0.0.0:1883".parse()?)
        .with_auth(auth_config);

    let mut broker = MqttBroker::with_config(config).await?;
    broker.run().await?;
    Ok(())
}

Broker Bridging

use mqtt_v5::broker::bridge::{BridgeConfig, BridgeDirection};
use mqtt_v5::QoS;

// Connect two brokers together
let bridge_config = BridgeConfig::new("edge-to-cloud", "cloud-broker:1883")
    // Forward sensor data from edge to cloud
    .add_topic("sensors/+/data", BridgeDirection::Out, QoS::AtLeastOnce)
    // Receive commands from cloud to edge  
    .add_topic("commands/+/device", BridgeDirection::In, QoS::AtLeastOnce)
    // Bidirectional health monitoring
    .add_topic("health/+/status", BridgeDirection::Both, QoS::AtMostOnce);

// Add bridge to broker (broker handles connection management)
// broker.add_bridge(bridge_config).await?;

🧪 Testing Support

Unit Testing with Mock Client

use mqtt_v5::{MockMqttClient, MqttClientTrait, PublishResult, QoS};

#[tokio::test]
async fn test_my_iot_function() {
    // Create mock client
    let mock = MockMqttClient::new("test-device");
    
    // Configure mock responses
    mock.set_connect_response(Ok(())).await;
    mock.set_publish_response(Ok(PublishResult::QoS1Or2 { packet_id: 123 })).await;
    
    // Test your function that accepts MqttClientTrait
    my_iot_function(&mock).await.unwrap();
    
    // Verify the calls
    let calls = mock.get_calls().await;
    assert_eq!(calls.len(), 2); // connect + publish
}

// Your production code uses the trait
async fn my_iot_function<T: MqttClientTrait>(client: &T) -> Result<(), Box<dyn std::error::Error>> {
    client.connect("mqtt://broker").await?;
    client.publish_qos1("telemetry", b"data").await?;
    Ok(())
}

☁️ AWS IoT Support

The client library includes AWS IoT compatibility features:

use mqtt_v5::{MqttClient, ConnectOptions};
use std::time::Duration;

// AWS IoT endpoint detection and connection handling
let client = MqttClient::new("aws-iot-device-12345");

// Connect to AWS IoT endpoint (automatically detects AWS IoT and optimizes connection)
client.connect("mqtts://abcdef123456.iot.us-east-1.amazonaws.com:8883").await?;

// Subscribe returns (packet_id, qos) tuple for compatibility
let (packet_id, qos) = client.subscribe("$aws/things/device-123/shadow/update/accepted", |msg| {
    println!("Shadow update accepted: {:?}", msg.payload);
}).await?;

// AWS IoT topic validation prevents publishing to reserved topics
use mqtt_v5::validation::namespace::NamespaceValidator;

let validator = NamespaceValidator::aws_iot().with_device_id("device-123");

// This will succeed - device can update its own shadow
client.publish("$aws/things/device-123/shadow/update", shadow_data).await?;

// This will be rejected - device cannot publish to shadow response topics
// client.publish("$aws/things/device-123/shadow/update/accepted", data).await?; // Error!

Key AWS IoT features:

Endpoint detection: Automatically detects AWS IoT endpoints and optimizes connection behavior
Topic validation: Built-in validation for AWS IoT topic restrictions and limits
ALPN support: TLS configuration with AWS IoT ALPN protocol support
Certificate loading: Load client certificates from bytes (PEM/DER formats)
SDK compatibility: Subscribe method returns (packet_id, qos) tuple like other AWS SDKs

🛠️ Development & Building

Prerequisites

Rust 1.82 or later
cargo-make (cargo install cargo-make)

Quick Setup

# Clone the repository
git clone https://github.com/fabriciobracht/mqtt-lib.git
cd mqtt-lib

# Install development tools and git hooks
./scripts/install-hooks.sh

# Run all CI checks locally (MUST pass before pushing)
cargo make ci-verify

Available Commands

# Development
cargo make build          # Build the project  
cargo make build-release  # Build optimized release version
cargo make test           # Run all tests
cargo make fmt            # Format code
cargo make clippy         # Run linter

# CI/CD  
cargo make ci-verify      # Run ALL CI checks (must pass before push)
cargo make pre-commit     # Run before committing (fmt + clippy + test)

# Examples (use raw cargo for specific targets)
cargo run --example simple_broker           # Start basic broker
cargo run --example broker_with_tls         # TLS-enabled broker
cargo run --example broker_with_websocket   # WebSocket-enabled broker
cargo run --example broker_bridge_demo      # Broker bridging demo

# Benchmarks (use raw cargo for specific targets)
cargo bench --bench broker_performance      # Broker performance tests
cargo bench --bench mqtt_benchmarks         # Core MQTT benchmarks

Testing

# Generate test certificates (required for TLS tests)
./scripts/generate_test_certs.sh

# Run unit tests (fast)
cargo make test-fast

# Run all tests including integration tests
cargo make test

# Run specific test suites (use raw cargo for specific targets)
cargo test --test broker_performance_tests
cargo test --test connection_pool_performance

🏗️ Architecture

This project follows modern Rust async patterns:

Design Principles

Direct async methods for all operations (no indirection)
Shared state via Arc<RwLock<T>> (no message passing)
Zero-copy operations where possible
Resource efficiency with connection pooling and buffer reuse

📊 Performance

The broker is designed for high performance:

10,000+ concurrent connections on modest hardware
Low memory footprint with connection pooling
Fast topic matching with optimized routing algorithms
Zero-copy message processing where possible
Comprehensive benchmarking suite for performance validation

🔐 Security

Security is built-in, not bolted-on:

TLS 1.2+ support with certificate validation
Username/password authentication with bcrypt hashing
Access Control Lists (ACL) for fine-grained permissions
Rate limiting to prevent DoS attacks
Resource monitoring to prevent resource exhaustion
Client certificate authentication for mutual TLS

📄 License

This project is licensed under either of

Apache License, Version 2.0, (LICENSE-APACHE or http://www.apache.org/licenses/LICENSE-2.0)
MIT license (LICENSE-MIT or http://opensource.org/licenses/MIT)

at your option.

🤝 Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

📚 Documentation

Architecture Overview - System design and principles
Broker Configuration - Complete config reference
Authentication Guide - Security setup
Deployment Guide - Production deployment
API Documentation - Complete API reference

Built with ❤️ in Rust. One reliable state machine.

mqtt5 0.4.4