Struct RetryConfig 

pub struct RetryConfig {
    pub attempts: u32,
    pub delay: HumanDuration,
    pub max_delay: HumanDuration,
    pub max_elapsed: Option<HumanDuration>,
}

User-facing retry configuration block (retry: at config root).

All fields are optional in YAML; missing fields fall back to the defaults (10 attempts, 10s base delay, 5m cap).

Fields

attempts: u32

Total attempts (including the first). Default 10. Values < 1 are clamped up to 1 by the policy layer.

delay: HumanDuration

Initial delay before the second attempt. Default 10s. Subsequent delays grow exponentially (delay × 2^(n-2)) up to Self::max_delay.

max_delay: HumanDuration

Upper bound on any individual sleep between attempts. Default 5m. Without this cap, an exponential backoff with delay=10s would stretch attempt 9 to ~42 minutes.

max_elapsed: Option<HumanDuration>

Cap on TOTAL retry wall-time across all attempts of a single operation. Retrying stops before a backoff sleep would push elapsed time past this budget, so a long transient storm fails cleanly (with the last error, resumable on an idempotent re-run) instead of running the full attempt ladder. Unset (field default None) resolves to a 15-minute budget (crate::retry::DEFAULT_MAX_ELAPSED); set it to raise or lower that ceiling. A publisher honors this budget by threading crate::context::Context::retry_deadline into its retry ladder; those whose surrounding CI job has a hard timeout should keep it below that timeout.

Implementations

impl RetryConfig

pub const DEFAULT_ATTEMPTS: u32 = 10

Default attempt count (10).

pub const DEFAULT_DELAY: Duration

Default initial delay (10s).

pub const DEFAULT_MAX_DELAY: Duration

Default delay cap (5m).

pub fn to_policy(&self) -> RetryPolicy

Bridge to the internal RetryPolicy consumed by crate::retry::retry_sync / crate::retry::retry_async.

If max_delay < delay, every backoff is immediately capped to max_delay. This is parity-correct passthrough but almost certainly a config mistake, so a tracing::warn! fires once at conversion time to surface the issue in logs.

pub fn max_elapsed_duration(&self) -> Option<Duration>

The configured total-retry-wall-time budget, if any.

Trait Implementations

impl Clone for RetryConfig

fn clone(&self) -> RetryConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for RetryConfig

impl Debug for RetryConfig

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

Formats the value using the given formatter.

impl Default for RetryConfig

fn default() -> Self

Returns the “default value” for a type.

impl<'de> Deserialize<'de> for RetryConfig

where RetryConfig: Default,

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>

where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl JsonSchema for RetryConfig

fn schema_name() -> Cow<'static, str>

The name of the generated JSON Schema.

fn schema_id() -> Cow<'static, str>

Returns a string that uniquely identifies the schema produced by this type.

fn json_schema(generator: &mut SchemaGenerator) -> Schema

Generates a JSON Schema for this type.

fn inline_schema() -> bool

Whether JSON Schemas generated for this type should be included directly in parent schemas, rather than being re-used where possible using the $ref keyword.

impl Serialize for RetryConfig

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>

where __S: Serializer,

Serialize this value into the given Serde serializer.