Loading

No logs, metrics, or traces visible in Kibana

Stack Serverless Observability

If the EDOT Collector or SDKs appear to be running, but you see no logs, metrics, or traces in the Kibana UI, try to use these solutions to identify and resolve the issue.

  • Collector process is consuming CPU/memory but no telemetry data is visible in Kibana
  • APM services don’t show up in Kibana
  • Dashboards appear empty or partially loaded
  • The Collector is running without crash logs

This issue is typically caused by one or more of the following:

  • Incorrect export endpoint
  • Missing or invalid API key/token
  • Network issues, such as proxy misconfigurations
  • TLS verification failures
  • Misconfigured pipelines or disabled signals
  • Incomplete setup — to capture all telemetry data, you must use the EDOT Collector together with an appropriate EDOT SDK

Use the following checks to identify and fix common configuration or connectivity issues that can prevent telemetry data from reaching Kibana.

Make sure the Collector is configured to send data to the correct Elastic endpoint.

To confirm:

  1. Open Kibana for your deployment.
  2. In the top right, select Add data.
  3. Select Application > OpenTelemetry.
  4. In the APM Agents panel, locate OpenTelemetry.
  5. Ensure the endpoint in your Collector configuration matches exactly.

If you’re using the managed OTLP endpoint, confirm the region and cluster ID are correct.

Ensure the Collector or SDK includes an API key in the Authorization header:

headers:
  Authorization: ApiKey <your-api-key>

If you’re using environment variables, confirm the key is set correctly in the runtime context.

Common log messages include:

permanent error: rpc error: code = Unavailable desc = connection refused

Also look for:

  • TLS handshake failures
  • Invalid character errors, which may indicate proxy or HTML redirect instead of JSON

Increase verbosity using --log-level=debug for deeper insights. <!--Refer to [Enable debug logging] for more information.-->

You can validate connectivity using curl:

curl -v https://<endpoint> -H "Authorization: ApiKey <your-key>"

Or use telnet or nc to verify port 443 is reachable.

Check that each pipeline is defined properly in your configuration:

service:
  pipelines:
    traces:
      receivers: [otlp]
      processors: [...]
      exporters: [otlp]

If only logs are configured, metrics and traces will not be sent.