With Agent v7.17+, the DogStatsD Mapper feature allows you to convert parts of a metric name submitted to DogStatsD to tags using mapping rules with wildcard and regex patterns. For example it allows you to transform the metric:

• airflow.job.duration.<JOB_TYPE>.<JOB_NAME>

into the metric airflow.job.duration with two associated tags:

• job_type:<JOB_TYPE>
• job_name:<JOB_NAME>.

To create a mapping rule:

1. Open your datadog.yaml file.
2. Add a mapping rule configuration block under the dogstatsd_mapper_profiles parameter.

Mapping rule configuration

A mapping rule block has the following layout:

dogstatsd_mapper_profiles:
    - name: '<PROFILE_NAME>'
      prefix: '<PROFILE_PREFIX>'
      mappings:
          - match: '<METRIC_TO_MATCH>'
            match_type: '<MATCH_TYPE>'
            name: '<MAPPED_METRIC_NAME>'
            tags:
                '<TAG_KEY>': '<TAG_VALUE_TO_EXPAND>'

With the following placeholders:

Wildcard match pattern

The wildcard match pattern matches dot-separated metric names using * as wildcard. The metric name must be only composed of alphanumeric, ., and _ characters for this pattern to work. Groups extracted can then be expanded with one of the following:

• $n format: $1, $2, $3, etc.
• ${n} format: ${1}, ${2}, ${3}, etc.

For instance, if you have the metric custom_metric.process.value_1.value_2 with the following mapping group configuration:

dogstatsd_mapper_profiles:
    - name: my_custom_metric_profile
      prefix: custom_metric.
      mappings:
          - match: 'custom_metric.process.*.*'
            match_type: wildcard
            name: custom_metric.process
            tags:
                tag_key_1: '$1'
                tag_key_2: '$2'

It would send the metric custom_metric.process to Datadog with the tags tag_key_1:value_1 and tag_key_2:value_2.

Regex match pattern

The regex match pattern matches metric names using regex patterns. Compared to the wildcard match pattern, it allows to define captured groups that contain .. Groups extracted can then be expanded with one of the following:

• $n format: $1, $2, $3, etc.
• ${n} format: ${1}, ${2}, ${3}, etc.

For instance, if you have the metric custom_metric.process.value_1.value.with.dots._2 with the following mapping group configuration:

dogstatsd_mapper_profiles:
    - name: my_custom_metric_profile
      prefix: custom_metric.
      mappings:
          - match: 'custom_metric\.process\.([\w_]+)\.(.+)'
            match_type: regex
            name: custom_metric.process
            tags:
                tag_key_1: '$1'
                tag_key_2: '$2'

It would send the metric custom_metric.process to Datadog with the tags tag_key_1:value_1 and tag_key_2:value.with.dots._2.

Expand group in metric name

For the regex and wildcard match type, group collected can be expanded as tags value with an associated tag key as see above, but can also be used in the metric name parameter. For instance, if you have the metric custom_metric.process.value_1.value_2 with the following mapping group configuration:

dogstatsd_mapper_profiles:
    - name: my_custom_metric_profile
      prefix: custom_metric.
      mappings:
          - match: 'custom_metric.process.*.*'
            match_type: wildcard
            name: 'custom_metric.process.prod.$1.live'
            tags:
                tag_key_2: '$2'

It would send the metric custom_metric.process.prod.value_1.live to Datadog with the tag tag_key_2:value_2.

Further Reading

Additional helpful documentation, links, and articles:

Introduction to DogStatsDDOCUMENTATION Official and Community created API and DogStatsD client librariesDOCUMENTATION