OpenTelemetry Host Monitoring extension
- Extension
Monitor OpenTelemetry-instrumented hosts with auto-generated entity topology, metric visualizations, and alerts for faster infrastructure analysis.
Get started
Overview
Monitor your hosts with OpenTelemetry using integrated metric visualizations, topology, and alerts. This extension automatically generates entities for hosts and their running processes, presenting telemetry data through intuitive interfaces designed for quick analysis. By correlating metrics, logs, and spans to host and process entities, you gain full context and a comprehensive view of your infrastructure.
Use cases
- Track performance, health, and availability of OpenTelemetry-monitored infrastructure.
- Analyze trends and baselines for capacity planning.
- Raise alerts on resource saturation, network errors, and other infrastructure issues.
- Get a dedicated, pre-configured view for infrastructure monitoring.
Requirements
This extension depends on telemetry data that is pushed to Dynatrace from OpenTelemetry via an OpenTelemetry Collector.
Activation and setup
-
Deploy the OpenTelemetry Collector.
- Follow the instructions on the OpenTelemetry Collector to deploy the Collector.
- Use the reference configuration.
- Ensure the Collector is running and telemetry data is reported to Dynatrace correctly.
-
Activate the OpenTelemetry Host Monitoring extension.
Details
The OpenTelemetry Collector captures telemetry data from your infrastructure and pushes metrics to Dynatrace via the OTel API.
The extension displays telemetry data captured by the OpenTelemetry Collector and adds Dynatrace-specific context to all signals (metrics, logs, and spans) for analysis and correlation.
For information on how to use the extension within Dynatrace, see Monitor hosts that send OpenTelemetry data to Dynatrace.
This extension does not include any feature sets. Users control what data the Collector sends to Dynatrace based on the Collector configuration.
Licensing and costs
All ingested OpenTelemetry data (logs, metrics, and spans) is charged according to your rate card, see License Dynatrace or Dynatrace classic licensing.
The use of this extension does not itself incur any additional costs.
Based on our measurements using the reference configuration, we estimate the following metrics ingest volume:
- Approximately 4400 metric data points per host per hour.
- Approximately 400 datapoints per process per hour.
Both these numbers will vary depending on the available metrics. For more details, see Limitations.
Limitations
- The metric
system.processes.createdis currently available only on Linux and BSD operating systems. - The metric
process.disk.iorequires running the Collector with privileged access. Not doing so will prevent the metric from being captured.
