Send Zipkin traces to Promscale

Promscale natively supports the OpenTelemetry Protocol (OTLP) for traces. To ingest Zipkin traces to Promscale, use the OpenTelemetry Collector. The OpenTelemetry Collector converts the Zipkin traces to OpenTelemetry traces.

Send data using the OpenTelemetry Collector

You can configure the OpenTelemetry Collector to forward Zipkin traces to Promscale using Zipkin receiver and the OpenTelemetry exporter.

Here’s an example configuration file for the OpenTelemetry Collector to forward traces to Promscale after batch processing the data:

  2. receivers:
  4.   zipkin:
  6. exporters:
  8.   otlp:
  10.     endpoint: "<PROMSCALE_HOST>:<gRPC_PORT>"
  12.     tls:
  14.       insecure: true
  16.     sending_queue:
  18.       queue_size: 1000000
  20.     timeout: 10s
  22. processors:
  24.   batch:
  26.     send_batch_size: 4000
  28.     send_batch_max_size: 4000
  30.     timeout: 10s
  32. service:
  34.   pipelines:
  36.     traces:
  38.       receivers: [zipkin]
  40.       exporters: [otlp]
  42.       processors: [batch]

Where:

  • <PROMSCALE_HOST>: hostname of Promscale.
  • <gRPC_PORT>: gRPC port of Promscale. The default port is 9202.

If you are running the OTLP Collector and the Promscale Connector on a Kubernetes cluster, the endpoint parameter is similar to endpoint: "promscale-connector.default.svc.cluster.local:<PORT>"

The default port exposed by the OpenTelemetry Collector Zipkin receiver is 9411. This is the port where you should send your Zipkin traces.

For more information about OpenTelemetry Collector, see the OpenTelemetry documentation.