Skip to content

Telemetry & Monitoring

DNS-collector exposes extensive internal telemetry metrics, allowing you to monitor the health, performance, CPU usage, memory allocations, and active goroutines of the DNS-collector process itself.

Why Enable Telemetry?

By enabling process telemetry, you gain real-time visibility into:

  • Ingress and egress message rates per pipeline worker.
  • Dropped and discarded message counts (crucial for buffer sizing and bottleneck detection).
  • System resource usage (CPU, memory allocations, garbage collection cycles, active goroutines count).

How to Enable Telemetry

To start exposing Prometheus metrics, enable telemetry in the global section of your configuration:

global:
  telemetry:
    enabled: true
    web-listen: ":9165"
    web-path: "/metrics"
    prometheus-prefix: "dnscollector"

Once enabled, you can access the raw Prometheus metrics at http://localhost:9165/metrics and point your Prometheus scraper to this endpoint.

Grafana Dashboard

To help visualize the internal metrics of the DNS-collector binary, a pre-configured Grafana dashboard is available:

  • Go Runtime Exporter Dashboard — Tracks CPU usage, memory allocations, garbage collection cycles, and active goroutines of the DNS-collector process.

You can download the raw dashboard configuration directly:

Download JSON

Dashboard Preview

Grafana Exporter Dashboard

Process Performance Metrics

In addition to standard Go runtime metrics, the telemetry endpoint exposes internal pipeline metrics to track performance and detect bottlenecks:

Pipeline & Worker Throughput

  • dnscollector_worker_ingress_traffic_total

    Total number of DNS messages received by each pipeline worker. This reflects successful handoff to the next internal processing stage.

  • dnscollector_worker_egress_traffic_total

    Total number of DNS messages successfully forwarded out by each worker. Note: This does not guarantee delivery to remote external systems.

  • dnscollector_worker_discarded_traffic_total

    The number of messages dropped when worker output channels/buffers are full (e.g., due to slow loggers or network bottlenecks).

  • dnscollector_policy_forwarded_total

    Total number of DNS messages forwarded after matching policy rules.

  • dnscollector_policy_dropped_total

    Total number of DNS messages dropped by active policy rules.