Custom metric metadata

To add more context to data points and their dimensions, your custom metric can carry additional useful information, such as the unit of measurement, display name, and value ranges.

You can provide such information via custom metric metadata. Metadata is stored independently from data points and tied together by the metric key. You can push data points and set metadata in any order.

You cannot provide metadata for built-in or calculated metrics; metadata is supported only for custom ingested metrics.

Available parameters

The following parameters are available for metric metadata.

Parameter

Type

Description

displayName

string

The name of the metric in the user interface.

description

string

A short description of the metric.

unit

string

The unit of the metric.

Find the possible values in the description of the unit field for built-in metrics.

tags

string[]

A list of tags applied to the metric.

metricProperties

A list of the metric's properties.

dimensions

A list of the metric's dimensions.

The MetricProperties object

Properties of a metric.

Parameter

Type

Description

minValue

integer

The minimum allowed value of the metric.

maxValue

integer

The maximum allowed value of the metric.

rootCauseRelevant

boolean

Whether (true or false) the metric is related to a root cause of a problem.

A root-cause relevant metric represents a strong indicator for a faulty component.

impactRelevant

boolean

Whether (true or false) the metric is relevant to a problem's impact.

An impact-relevant metric is highly dependent on other metrics and changes because an underlying root-cause metric has changed.

valueType

string

The type of the metric's value. You have these options:

  • score: A score metric is a metric where high values indicate a good situation, while low values indicate trouble. An example of such a metric is a success rate.
  • error: An error metric is a metric where high values indicate trouble, while low values indicate a good situation. An example of such a metric is an error count.

latency

integer

The reporting delay of the metrics, in minutes.

The delay caused by constraints of cloud vendors or other third-party data sources that leads to a latency in data ingest on the Dynatrace side.

The MetricDimensions object

A dimension of the metric.

Parameter

Type

Description

key

string

The key of the dimension to be used in the ingestion protocol.

displayName

string

The name of the dimension in the user interface.

Set metric metadata

Use the POST an object call of the Settings API to provide metadata for your metric. Use the following parameters in the payload:

Field
Value
scope
metric-{your-metric-key}
schemaId
builtin:metric.metadata
value
The desired set of metadata. See the available fields above.
[
  {
    "scope": "metric-business.shop.revenue",
    "schemaId": "builtin:metric.metadata",
    "value": {
      "displayName": "Total revenue",
      "description": "Total store revenue by region, city, and store",
      "unit": "Unspecified",
      "tags": ["KPI", "Business"],
      "metricProperties": {
        "maxValue": 1000000,
        "minValue": 0,
        "rootCauseRelevant": false,
        "impactRelevant": true,
        "valueType": "score",
        "latency": 1
      },
      "dimensions": [
        {
          "key": "city",
          "displayName": "City name"
        },
        {
          "key": "country",
          "displayName": "Country name"
        },
        {
          "key": "region",
          "displayName": "Sales region"
        },
        {
          "key": "store",
          "displayName": "Store #"
        }
      ]
    }
  }
]

Alternatively, you can:

View metric metadata

You can retrieve the metadata of a metric via the GET metric descriptor call of the Metrics v2 API or via the Metrics browser.

Related topics