How to Monitor Envoy Metrics with Prometheus, Grafana, and Nacos

In modern cloud‑native architectures, Envoy serves as a high‑performance edge and service‑mesh proxy, acting as the "gatekeeper" for inter‑service communication. Effective monitoring of Envoy has become essential for ensuring system stability.

1. Enable Envoy admin interface

admin:
  address:
    socket_address:
      address: 0.0.0.0
      port_value: 9901

Adding the above configuration exposes Envoy's management API on port 9901.

2. Verify metrics data

After restarting Envoy, you can access http://<em>host</em>:9901/stats/prometheus to see the raw metrics.

3. Register Envoy with Nacos

$ nacos-service-register register -s 172.139.20.100:8848/nacos -u jiaxzeng -p /retaIkk6uNjaQTdqp5bgg== --namespaceId 46f891fb-c1a2-4b67-a0cb-b9d3888ac9be -i 
'{"serviceName": "envoy", "port": 9901, "ip": "172.139.20.170", "metadata": { "__meta_nacos_service_name": "envoy", "metrics_path": "/stats/prometheus" } }'
[INFO] envoy service register 172.139.20.170 instance succeed

This registers the Envoy instance in Nacos so that Prometheus can discover it.

4. Let Prometheus scrape the metrics

Prometheus is configured to query Nacos for services named envoy and scrape the /stats/prometheus endpoint.

Verify that Prometheus receives the metrics.

5. Import Grafana dashboard

Download the official Envoy dashboard JSON from GitHub and import it into Grafana.

6. Add monitoring panels

After importing, the dashboard displays key Envoy metrics such as request rates, latency, and error ratios.

Conclusion

Envoy is more than a proxy; it is a "data probe" in modern application architectures. Proper observability enables rapid fault response and proactive risk detection, shifting operations from firefighting to fire‑prevention. Incorporate Envoy monitoring into your standard delivery process to ensure every service’s traffic is visible, measurable, and controllable.