Troubleshoot the Elastic Agent

Use the topics in this section to troubleshoot issues with the Elastic Agent.

If you're not sure where to start, review the Collector's logs for error messages and validate your configuration using the --dry-run option. For more detailed diagnostics, refer to Enable debug logging.

• Collector out of memory: Diagnose and resolve out-of-memory issues in the Elastic Agent using Go's Performance Profiler.

• Insufficient resources in Kubernetes: Troubleshoot resource allocation issues when running the Elastic Agent in Kubernetes environments.

• Collector doesn't start: Resolve startup failures caused by invalid configuration, port conflicts, or missing components.

• Missing or incomplete traces due to Collector sampling: Troubleshoot missing or incomplete traces caused by sampling configuration.

• APM services missing due to misconfigured elasticapm connector: Resolve missing APM services and metrics caused by placing the elasticapm connector under processors: instead of connectors:.

• Collector doesn't propagate client metadata: Learn why the Collector doesn't extract custom attributes and how to propagate such values using Elastic OTel SDKs.

• Export failures when sending telemetry data: Resolve export failures caused by sending_queue overflow and Elasticsearch exporter timeouts.

• ResourceExhausted errors and decompression limits in Collector-to-Collector pipelines: Troubleshoot ResourceExhausted errors caused by gRPC message size limits, decompression limits, memory pressure, or backpressure in Collector-to-Collector pipelines.

• Enable debug logging: Learn how to enable debug logging for the Elastic Agent in supported environments.

• Elastic OTel SDKs troubleshooting: For end-to-end issues that may involve both the Collector and SDKs.

• Troubleshoot Elastic OpenTelemetry: Overview of all Elastic OpenTelemetry troubleshooting resources.

For in-depth details on troubleshooting, refer to the contrib OpenTelemetry Collector troubleshooting documentation.