Cloudera Docs

Understanding telemetry data export in Cloudera Observability

Cloudera Observability exports collected telemetry data by using any exporter that the Cloudera distribution of OpenTelemetry collector supports. The exporter specification describes the encoding, transport, and delivery of telemetry data between telemetry sources, intermediate nodes such as collectors, and telemetry backends. This capability allows you to integrate Cloudera Observability with your existing services.

Cloudera supported exporters:

  • AWS EMF Exporter
  • Azure Monitor Exporter
  • Datadog Exporter
  • FileExporter
  • Load Balancing Exporter
  • Prometheus Exporter
  • Prometheus Remote Write Exporter
  • Syslog Exporter
  • Splunk HTTP Event Collector (HEC) Exporter
  • Debug Exporter
  • nop (no-operation) exporter
  • OTLP (OpenTelemetry Protocol) Exporter
  • OTLP HTTP Exporter

To allow data delivery to the backend, Cloudera Observability supports multiple components from the OpenTelemetry Collector. You can route infrastructure metrics, application metrics, and workload information logs to the analytics engines. Exporters currently export these logs and metrics.

Cloudera Observability collects and exports infrastructure and application or service metrics of the Cloudera Data Services on premises (Cloudera Data Engineering and Cloudera AI) environment. The OpenTelemetry Collector periodically scrapes running and completed jobs by calling internal APIs to gather status data.

For Cloudera AI, the exported logs include workload details, execution status, creator metadata, and assigned runtime configurations. For Cloudera Data Engineering, the collector generates payloads covering job types, such as Spark or Airflow, job status, access control lists (ACLs), catalog IDs, and execution parameters.