Observability

WebMQ exposes built-in health checks and Prometheus metrics for monitoring.

Health check

When enabled, the server exposes a health check endpoint that verifies RabbitMQ connectivity and WebSocket server status.

Configuration

const server = new WebMQServer({
  rmqUrl: 'amqp://localhost',
  exchange: 'my_app',
  port: 8080,
  healthEndpoint: '/health'   // Default: /health
});

Response

{
  "healthy": true,
  "rabbitMQQueues": 5,
  "websockets": 42
}

• HTTP 200 when healthy
• HTTP 503 when unhealthy (RabbitMQ disconnected or consecutive channel failures)

Disable

Omit healthEndpoint from the options (or set to empty string).

Prometheus metrics

WebMQ exposes standard Prometheus metrics for monitoring message throughput, connections, and errors.

Configuration

const server = new WebMQServer({
  rmqUrl: 'amqp://localhost',
  exchange: 'my_app',
  port: 8080,
  metricsEndpoint: '/metrics'  // Default: /metrics
});

Available metrics

Metric	Type	Description
webmq_websocket_connections	Gauge	Current active WebSocket connections
webmq_rabbitmq_consumers	Gauge	Current RabbitMQ consumers
webmq_websocket_messages_received_total	Counter	Messages received from WebSocket clients
webmq_websocket_to_rabbitmq_publishes_total	Counter	Messages published to RabbitMQ
webmq_rabbitmq_to_websocket_publishes_total	Counter	Messages forwarded from RMQ to WebSocket
webmq_rabbitmq_bindings	Gauge	Current RabbitMQ bindings
webmq_rabbitmq_messages_acked_total	Counter	Messages acknowledged to RabbitMQ
webmq_rabbitmq_messages_nacked_total	Counter	Messages negatively acknowledged (requeued)
webmq_rabbitmq_consecutive_channel_failures	Gauge	Consecutive RabbitMQ channel failures
webmq_errors_total	Counter	Errors by type
process_*	—	Default Prometheus process metrics (enabled automatically)

Prometheus scrape config

# prometheus.yml
scrape_configs:
  - job_name: 'webmq'
    scrape_interval: 15s
    static_configs:
      - targets: ['localhost:8080']

Grafana queries

# Messages per second
rate(webmq_websocket_to_rabbitmq_publishes_total[1m])

# Active connections over time
webmq_websocket_connections

# Error rate by type
rate(webmq_errors_total[5m])

# RabbitMQ health
webmq_rabbitmq_consecutive_channel_failures

Logging

Both frontend and backend support configurable log levels:

// Frontend
import WebMQClient from 'webmq-frontend';
const client = new WebMQClient({ url: 'ws://localhost:8080', sessionId: 'abc' });
client.logLevel = 'DEBUG';  // 'DEBUG' | 'INFO' | 'WARNING' | 'ERROR' | 'SILENT'

// Backend
const server = new WebMQServer({
  rmqUrl: 'amqp://localhost',
  exchange: 'my_app',
  port: 8080,
  logLevel: 'INFO'  // Same levels as frontend
});
// Or set after construction:
server.logLevel = 'DEBUG';

Level hierarchy: SILENT < ERROR < WARNING < INFO < DEBUG.

Benchmarking

The project includes a benchmarking suite at benchmarks/ for measuring throughput and latency under load. See the benchmarks README for details.