Prometheus Integration

RobustMQ has built-in Prometheus metrics export. Metrics are always exposed via the Admin HTTP API's GET /metrics, sharing http_port (default 58080) — there's no separate enable switch or port to configure.

Verify Metrics Export

Replace 58080 below with the http_port configured in your config/server.toml:

# Check metrics endpoint
curl http://localhost:58080/metrics

# Verify metrics data
curl http://localhost:58080/metrics | grep mqtt_

Configure Prometheus

Add RobustMQ as a scrape target in Prometheus configuration:

# prometheus.yml
scrape_configs:
  - job_name: 'robustmq'
    static_configs:
      - targets: ['localhost:58080']
    scrape_interval: 15s
    metrics_path: /metrics

Cluster Configuration

For multi-node deployment (replace with each node's actual http_port):

scrape_configs:
  - job_name: 'robustmq-cluster'
    static_configs:
      - targets:
        - 'robustmq-node1:58080'
        - 'robustmq-node2:58080'
        - 'robustmq-node3:58080'

Available Metrics

RobustMQ exports the following types of metrics:

• MQTT Protocol Metrics: Packet send/receive, connection management, authentication statistics
• Performance Metrics: Request latency, processing time, queue depth
• Business Metrics: Session count, topic statistics, message processing
• System Metrics: Network connections, thread pools, error statistics

Common Queries

# Current connection count
mqtt_connections_count

# MQTT packet receive rate
rate(mqtt_packets_received[5m])

# Request processing latency P95
histogram_quantile(0.95, rate(request_total_ms_bucket[5m]))

# Authentication failure rate
rate(mqtt_auth_failed[5m])

Troubleshooting

Metrics Endpoint Inaccessible

# Check port listening (http_port defaults to 58080)
netstat -tlnp | grep 58080

# Check the configured http_port
grep "http_port" config/server.toml

Prometheus Cannot Scrape

# Check network connectivity
telnet robustmq-host 58080

# View Prometheus targets status
curl http://prometheus:9090/api/v1/targets

With the above configuration, RobustMQ monitoring metrics will be automatically collected by Prometheus for alerting, visualization, and performance analysis.