Struct SamplerConfig 

pub struct SamplerConfig { /* private fields */ }

Settings supplied by one APM_SAMPLING configuration.

The payload and each settings value in it must be a JSON object or null, as the Datadog Agent’s JSON decoder requires; serde’s derived visitors would also accept an array of field values, such as [].

Call resolve to combine the configuration with local settings for one environment.

{
  "all_envs": { "priority_sampler_target_TPS": 10, "errors_sampler_target_TPS": 5, "rare_sampler_enabled": true },
  "by_env": [{ "env": "prod", "config": { "rare_sampler_enabled": false } }]
}

Missing and null fields fall back through all_envs to local settings; 0 and false are explicit values. A payload that sets nothing, such as {} or null, is still an assigned configuration (RemoteSampling::Remote) and resolves entirely to local settings, not to previous remote values. This differs from an empty assignment, which leaves the last accepted configuration in place.

The Datadog Agent sends the *_TPS field names. Lowercase *_tps names also work; unknown fields are ignored.

Implementations

impl SamplerConfig

pub fn resolve( &self, env: &str, static_settings: &SamplerSettings, ) -> SamplerSettings

Combines remote overrides with local sampling settings for the Agent’s environment.

The result contains target traces per second for priority and error sampling, plus an on/off switch for rare sampling. Priority sampling honors existing decisions to keep or drop traces and uses its target to compute sampling rates. Error sampling can keep error traces that would otherwise be dropped; rare sampling keeps examples of rarely seen operations.

env identifies the Agent’s deployment environment, such as prod, not an individual trace’s environment. The caller supplies the Agent’s resolved, normalized environment. This method neither derives that environment nor normalizes names: a remote name of Prod does not match prod.

This method selects the last by_env entry whose name exactly matches env. Earlier matching entries contribute nothing, even if the selected entry leaves fields unset. For each setting independently, it uses:

1. The value in the selected entry’s config, if set.
2. Otherwise, the value in all_envs, the remote settings shared by all environments.
3. Otherwise, the value in static_settings, the original local settings, not previous remote values.

Missing or null fields are unset; 0 and false are explicit overrides. If no entry matches env, each setting falls back to all_envs, then static_settings. Numeric targets pass through without validation or clamping.

For example, suppose the selected prod entry sets only the priority target to 20 and all_envs sets only the error target to 5. The result uses 20 for priority sampling, 5 for error sampling, and the local setting for rare sampling.

Trait Implementations

impl Clone for SamplerConfig

fn clone(&self) -> SamplerConfig

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for SamplerConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for SamplerConfig

fn default() -> SamplerConfig

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for SamplerConfig

fn deserialize<D: Deserializer<'de>>(deserializer: D) -> Result<Self, D::Error>

Deserialize this value from the given Serde deserializer.

impl PartialEq for SamplerConfig

fn eq(&self, other: &SamplerConfig) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for SamplerConfig