🦦 OtterMQ

OtterMQ is a high-performance message broker written in Go, aiming for compatibility with the AMQP 0.9.1 protocol and interoperability with RabbitMQ clients. It includes a broker backend and a modern management UI built with Vue 3 + Quasar.

This site tracks project status and documentation for users and contributors. For full source code and issue tracking, visit the GitHub repository:

• Repository: https://github.com/ottermq/ottermq
• Docker images and compose files are available in the repo

Quick links

• AMQP 0.9.1 Support Status: Status Matrix
• REST API Swagger: served by the broker at /docs when running
• Project README: https://github.com/ottermq/ottermq#readme
• Roadmap: https://github.com/ottermq/ottermq/blob/main/ROADMAP.md

Highlights

• Complete AMQP 0.9.1 Basic Class - All message operations implemented
• AMQP-style exchanges and queues (direct, fanout, topic)
• RabbitMQ client compatibility tested with multiple clients
• Quality of Service (QoS) with prefetch limits
• Message acknowledgments, rejections, and recovery
• Pluggable persistence layer (JSON today; Memento WAL planned)
• Built-in management UI (Vue + Quasar)
• Supports multiple virtual hosts for tenant isolation
• Low memory footprint with smart buffering and backpressure

RabbitMQ Extensions

• Dead Letter Exchange (DLX): Route failed/rejected messages for retry or logging
• Message TTL and Expiration: Time-To-Live for messages with per-message and per-queue TTL support

Metrics & Observability

OtterMQ includes built-in metrics collection for monitoring broker performance in real-time.

Available Metrics

Broker-Level

• Total publish rate, delivery rate, ack/nack rates (messages/second)
• Active connections, channels, queues, exchanges (counts)
• Total message count, consumer count

Queue Metrics (per queue)

• Message rate (enqueue/sec), delivery rate, ack rate
• Current depth (ready + unacked messages)
• Consumer count

Channel Metrics (per channel)

• Publish rate, deliver rate, ack rate, unroutable rate
• Channel state (running, idle, flow, closing)
• Unacked message count, prefetch settings

Exchange Metrics (per exchange)

• Publish rate, delivery/routing rate
• Message counts

Access via Management API

All metrics are available via REST endpoints:

# Get all channels with live metrics
curl http://localhost:3000/api/channels | jq

# Response includes rates:
{
  "channels": [{
    "number": 1,
    "connection_name": "127.0.0.1:54321",
    "state": "running",
    "publish_rate": 125.4,
    "deliver_rate": 120.8,
    "ack_rate": 118.2
  }]
}

Time-series data: Metrics include 5-minute history by default (configurable), enabling trend analysis and charting in the management UI.

Configuration

Metrics are enabled by default with minimal performance overhead. Configure via environment variables:

OTTERMQ_ENABLE_METRICS=true              # Enable/disable
OTTERMQ_METRICS_WINDOW_SIZE=5m           # Retention window
OTTERMQ_METRICS_MAX_SAMPLES=60           # History samples
OTTERMQ_METRICS_SAMPLES_INTERVAL=5       # Sample frequency (seconds)

See Configuration Reference for all available settings.

Future: Prometheus exporter and Grafana dashboards are planned for advanced monitoring workflows.

Getting started

• Build and run using make build-all then make run
• Development: make run-dev for the broker and quasar dev for the UI in ottermq_ui/

See the repository README for detailed steps and configuration options.

Contributing documentation

Status pages live in this /docs folder and are deployed from the pages branch via GitHub Pages. When implementing or changing AMQP features (classes/methods), please keep the Status Matrix up to date.