TransportFailurePolicy

class TransportFailurePolicy(val receiveRetry: Retry = receiveRetry())

How a Transport reacts when its receive stream fails, the same for every backend.

The receive stream (TransportReceiver.records) is subscribed again after each failure, with receiveRetry's backoff. Only once the retries are exhausted does the failure reach the dispatcher, which reports it to the runtime. Retries count consecutive failures: a record received in between starts the count again (transient errors).

The default retries 3 times from a 10 s backoff, which is what the Kafka receiver did before 9.3.0; Redis Streams retried nothing before 9.3.0.

Constructors

Link copied to clipboard
constructor(receiveRetry: Retry = receiveRetry())

Types

Link copied to clipboard
object Companion

Properties

Link copied to clipboard
val receiveRetry: Retry

Functions

Link copied to clipboard
fun <T : Any> retryReceive(records: Flux<T>): Flux<T>

records, subscribed again on failure according to receiveRetry.