Extraction stages in OpenPipeline

Latest Dynatrace
Explanation
3-min read
Published Apr 13, 2026

This article describes extraction in OpenPipeline and provides an overview of the extraction stages and processors available in pipelines.

Get familiar with OpenPipeline stage and processor concepts before you begin. To learn more, see Processing in OpenPipeline.

Overview

You can extract new telemetry from incoming records and create records of a different type than the source records. For example, you can extract metrics from spans and process those metrics in OpenPipeline.

All extraction stages follow the same logic:

The matching condition specifies from which records new telemetry should be extracted.
You can extract fields or dimensions for the new telemetry and choose to extract all fields or dimensions, specific fields or dimensions, or exclude specific fields.

Important fields in extraction

Dynatrace version 1.337

Important fields do not apply to the Davis problems and Davis events configuration scopes.

Certain fields must be present in extracted telemetry to ensure that permissions, cost allocation, and analytics continue to work correctly and to avoid misconfiguration.

Important fields include:

dt.security_context: Enforces record-level permissions.
dt.cost.costcenter and dt.cost.product: Required for cost and billing attribution.
Primary Grail fields: Provide entity and context information. When primary fields are updated in the Semantic Dictionary, the changes are reflected in OpenPipeline the next time you create or update a processor.
dt.smartscape.*: Identify Smartscape entities and relationships. These fields support wildcard extraction.
primary_tags.*: Provide metadata tags for compliance and operational workflows. These fields support wildcard extraction.

How important fields are handled

Depending on how you configure field extraction in a processor, important fields are handled as follows:

Extract all fields or dimensions: All important fields are included automatically. No manual configuration is required.
Extract specific fields or dimensions: OpenPipeline offers a predefined list of important fields that you can choose to include in your configuration.
Exclude specific fields: No fields are excluded by default. If you manually add a field to the exclusion list and it is an important field, Dynatrace informs you.
Existing processors created with Dynatrace version 1.336 and earlier: Important fields are not added retroactively. We recommend that you manually add the relevant fields or dimensions to your existing processors.

Important fields are appended at the end of the extracted field list. Keeping important fields is a best practice, but you can remove them.

Wildcard extraction

dt.smartscape.* and primary_tags.* support two extraction modes.

Catch-all (dt.smartscape.*, primary_tags.*): All fields matching the prefix are extracted. This is the default. New fields introduced later are included automatically. If the limit of fields is exceeded, Dynatrace removes such fields in order, from the last until it re-enters the limit.
Specific fields: Only the listed fields are extracted (for example, dt.smartscape.host or primary_tags.team). This keeps dimensions stable; in case new Smartscape or primary tag fields are added, existing time series are preserved.

Available extraction stages and processors

Stage Name	Processors
Metric extraction	Counter metric (`counterMetric`), Sampling aware counter metric (`samplingAwareCounterMetric`), Value metric (`valueMetric`), Histogram metric (`histogramMetric`), Sampling aware value metric (`samplingAwareValueMetric`), Sampling aware histogram metric (`samplingAwareHistogramMetric`)
Smartscape node extraction	Smartscape node (`smartscapeNode`)
Smartscape edge extraction	Smartscape edge (`smartscapeEdge`)
Data extraction	Business event (`bizevent`), Software development lifecycle event (`sdlcEvent`)
Davis extraction	Davis event (`davis`)

API parameters

See the Settings API builtin:openpipeline.<configuration-scope>.pipelines schema reference for a complete overview of available API parameters.

Use cases

Extract metrics to produce custom latency, throughput, or error metrics.
Extract business events from logs to support domain-focused analysis and dashboards.
Derive SDLC events from CI/CD logs or generic events to track deployments and releases.
Extract custom topology.
Generate Davis events for incident correlation.