This is documentation for the next version of Agent. For the latest stable release, go to the latest version.

Grafana Agent Flow mode Reference Components reference otelcol.processor.probabilistic_sampler

otelcol.processor.probabilistic_sampler

EXPERIMENTAL: This is an experimental component. Experimental components are subject to frequent breaking changes, and may be removed with no equivalent replacement.

otelcol.processor.probabilistic_sampler accepts logs and traces data from other otelcol components and applies probabilistic sampling based on configuration options.

You can specify multiple otelcol.processor.probabilistic_sampler components by giving them different labels.

Usage

river Copy

otelcol.processor.probabilistic_sampler "LABEL" {
  output {
    logs    = [...]
    traces  = [...]
  }
}

Arguments

otelcol.processor.probabilistic_sampler supports the following arguments:

hash_seed determines an integer to compute the hash algorithm. This argument could be used for both traces and logs. When used for logs, it computes the hash of a log record. For hashing to work, all collectors for a given tier, for example, behind the same load balancer, must have the same hash_seed. It is also possible to leverage a different hash_seed at different collector tiers to support additional sampling requirements.

sampling_percentage determines the percentage at which traces or logs are sampled. All traces or logs are sampled if you set this argument to a value greater than or equal to 100.

attribute_source (logs only) determines where to look for the attribute in from_attribute. The allowed values are traceID or record.

from_attribute (logs only) determines the name of a log record attribute used for sampling purposes, such as a unique log record ID. The value of the attribute is only used if the trace ID is absent or if attribute_source is set to record.

sampling_priority (logs only) determines the name of a log record attribute used to set a different sampling priority from the sampling_percentage setting. 0 means to never sample the log record, and greater than or equal to 100 means to always sample the log record.

The probabilistic_sampler supports two types of sampling for traces:

1. sampling.priority semantic convention as defined by OpenTracing.
2. Trace ID hashing.

The sampling.priority semantic convention takes priority over trace ID hashing. Trace ID hashing samples based on hash values determined by trace IDs.

The probabilistic_sampler supports sampling logs according to their trace ID, or by a specific log record attribute.

Exported fields

The following fields are exported and can be referenced by other components:

input accepts otelcol.Consumer OTLP-formatted data for any telemetry signal of these types:

• logs
• traces

Component health

otelcol.processor.probabilistic_sampler is only reported as unhealthy if given an invalid configuration.

Debug information

otelcol.processor.probabilistic_sampler does not expose any component-specific debug information.

Examples

Basic usage

river Copy

otelcol.processor.probabilistic_sampler "default" {
  hash_seed           = 123
  sampling_percentage = 15.3

  output {
    logs = [otelcol.exporter.otlp.default.input]
  }
}

Sample 15% of the logs

river Copy

otelcol.processor.probabilistic_sampler "default" {
  sampling_percentage = 15

  output {
    logs = [otelcol.exporter.otlp.default.input]
  }
}

Sample logs according to their “logID” attribute

river Copy

otelcol.processor.probabilistic_sampler "default" {
  sampling_percentage = 15
  attribute_source    = "record"
  from_attribute      = "logID"

  output {
    logs = [otelcol.exporter.otlp.default.input]
  }
}

Sample logs according to a “priority” attribute

river Copy

otelcol.processor.probabilistic_sampler "default" {
  sampling_percentage = 15
  sampling_priority   = "priority"

  output {
    logs = [otelcol.exporter.otlp.default.input]
  }
}