Returns a scala.concurrent.Future that will be completed with the success or failure of the provided value after the specified duration.

Returns a scala.concurrent.Future that will be completed with the success or failure of the provided value after the specified duration.

Returns a java.util.concurrent.CompletionStage that will be completed with the success or failure of the provided value after the specified duration.

Returns a java.util.concurrent.CompletionStage that will be completed with the success or failure of the provided value after the specified duration.

Sends a message asynchronously and returns a scala.concurrent.Future holding the eventual reply message; this means that the target actor needs to send the result to the sender reference provided.

The Future will be completed with an pekko.pattern.AskTimeoutException after the given timeout has expired; this is independent from any timeout applied while awaiting a result for this future (i.e. in Await.result(..., timeout)). A typical reason for AskTimeoutException is that the recipient actor didn't send a reply.

Warning: When using future callbacks, inside actors you need to carefully avoid closing over the containing actor’s object, i.e. do not call methods or access mutable state on the enclosing actor from within the callback. This would break the actor encapsulation and may introduce synchronization bugs and race conditions because the callback will be scheduled concurrently to the enclosing actor. Unfortunately there is not yet a way to detect these illegal accesses at compile time.

Recommended usage:

val f = ask(worker, request)(timeout)
f.map { response =>
  EnrichedMessage(response)
} pipeTo nextActor

Import this implicit conversion to gain ? and ask methods on pekko.actor.ActorSelection, which will defer to the ask(actorSelection, message)(timeout) method defined here.

import org.apache.pekko.pattern.ask

val future = selection ? message             // => ask(selection, message)
val future = selection ask message           // => ask(selection, message)
val future = selection.ask(message)(timeout) // => ask(selection, message)(timeout)

All of the above use an implicit pekko.util.Timeout.

Sends a message asynchronously and returns a scala.concurrent.Future holding the eventual reply message; this means that the target actor needs to send the result to the sender reference provided.

The Future will be completed with an pekko.pattern.AskTimeoutException after the given timeout has expired; this is independent from any timeout applied while awaiting a result for this future (i.e. in Await.result(..., timeout)). A typical reason for AskTimeoutException is that the recipient actor didn't send a reply.

Warning: When using future callbacks, inside actors you need to carefully avoid closing over the containing actor’s object, i.e. do not call methods or access mutable state on the enclosing actor from within the callback. This would break the actor encapsulation and may introduce synchronization bugs and race conditions because the callback will be scheduled concurrently to the enclosing actor. Unfortunately there is not yet a way to detect these illegal accesses at compile time.

Recommended usage:

val f = ask(worker, request)(timeout)
f.map { response =>
  EnrichedMessage(response)
} pipeTo nextActor

Import this implicit conversion to gain ? and ask methods on pekko.actor.ActorRef, which will defer to the ask(actorRef, message)(timeout) method defined here.

import org.apache.pekko.pattern.ask

val future = actor ? message             // => ask(actor, message)
val future = actor ask message           // => ask(actor, message)
val future = actor.ask(message)(timeout) // => ask(actor, message)(timeout)

All of the above use an implicit pekko.util.Timeout.

Use for messages whose response is known to be a pekko.pattern.StatusReply. When a pekko.pattern.StatusReply.Success response arrives the future is completed with the wrapped value, if a pekko.pattern.StatusReply.Error arrives the future is instead failed.

Use for messages whose response is known to be a pekko.pattern.StatusReply. When a pekko.pattern.StatusReply.Success response arrives the future is completed with the wrapped value, if a pekko.pattern.StatusReply.Error arrives the future is instead failed.

Returns a scala.concurrent.Future that will be completed with success (value true) when existing messages of the target actor has been processed and the actor has been terminated.

Useful when you need to wait for termination or compose ordered termination of several actors, which should only be done outside of the ActorSystem as blocking inside Actors is discouraged.

IMPORTANT NOTICE: the actor being terminated and its supervisor being informed of the availability of the deceased actor’s name are two distinct operations, which do not obey any reliable ordering. Especially the following will NOT work:

def receive = {
  case msg =>
    Await.result(gracefulStop(someChild, timeout), timeout)
    context.actorOf(Props(...), "someChild") // assuming that that was someChild’s name, this will NOT work
}

If the target actor isn't terminated within the timeout the scala.concurrent.Future is completed with failure pekko.pattern.AskTimeoutException.

If you want to invoke specialized stopping logic on your target actor instead of PoisonPill, you can pass your stop command as a parameter:

gracefulStop(someChild, timeout, MyStopGracefullyMessage).onComplete {
   // Do something after someChild being stopped
}

Import this implicit conversion to gain the pipeTo method on scala.concurrent.Future:

import org.apache.pekko.pattern.pipe
// requires implicit ExecutionContext, e.g. by importing `context.dispatcher` inside an Actor

Future { doExpensiveCalc() } pipeTo nextActor

or

pipe(someFuture) to nextActor

The successful result of the future is sent as a message to the recipient, or the failure is sent in a pekko.actor.Status.Failure to the recipient.

Import this implicit conversion to gain the pipeTo method on scala.concurrent.Future:

import org.apache.pekko.pattern.pipe
// requires implicit ExecutionContext, e.g. by importing `context.dispatcher` inside an Actor

Future { doExpensiveCalc() } pipeTo nextActor

or

pipe(someFuture) to nextActor

The successful result of the future is sent as a message to the recipient, or the failure is sent in a pekko.actor.Status.Failure to the recipient.

Given a function from Unit to Future, returns an internally retrying Future.

When the future is completed, the shouldRetry predicate is always been invoked with the result (or null if none) and the exception (or null if none). If the shouldRetry predicate returns true, then a new attempt is made, each subsequent attempt will be made after the 'delay' return by delayFunction (the input next attempt count start from 1). Returns scala.None for no delay.

A scheduler (eg context.system.scheduler) must be provided to delay each retry. You could provide a function to generate the next delay duration after first attempt, this function should never return null, otherwise an java.lang.IllegalArgumentException will be through.

If attempts are exhausted the returned future is simply the result of invoking attempt. Note that the attempt function will be invoked on the given execution context for subsequent tries and therefore must be thread safe (i.e. not touch unsafe mutable state).

Example usage:

//retry with back off

protected val sendAndReceive: HttpRequest => Future[HttpResponse]
protected val shouldRetry: (HttpResponse, Throwable) => throwable ne null
private val sendReceiveRetry: HttpRequest => Future[HttpResponse] = (req: HttpRequest) => retry[HttpResponse](
  attempt = () => sendAndReceive(req),
  shouldRetry,
  attempts = 10,
  delayFunction = attempted => Option(2.seconds * attempted)
)

Given a function from Unit to Future, returns an internally retrying Future. The first attempt will be made immediately, each subsequent attempt will be made after the 'delay' return by delayFunction (the input next attempt count start from 1). Returns scala.None for no delay.

A scheduler (eg context.system.scheduler) must be provided to delay each retry. You could provide a function to generate the next delay duration after first attempt, this function should never return null, otherwise an java.lang.IllegalArgumentException will be through.

If attempts are exhausted the returned future is simply the result of invoking attempt. Note that the attempt function will be invoked on the given execution context for subsequent tries and therefore must be thread safe (i.e. not touch unsafe mutable state).

Example usage:

//retry with back off

protected val sendAndReceive: HttpRequest => Future[HttpResponse]
private val sendReceiveRetry: HttpRequest => Future[HttpResponse] = (req: HttpRequest) => retry[HttpResponse](
  attempt = () => sendAndReceive(req),
  attempts = 10,
  delayFunction = attempted => Option(2.seconds * attempted)
)

Given a function from Unit to Future, returns an internally retrying Future.

When the future is completed, the shouldRetry predicate is always been invoked with the result (or null if none) and the exception (or null if none). If the shouldRetry predicate returns true, then a new attempt is made, each subsequent attempt will be made after the 'delay' return by delayFunction (the input next attempt count start from 1). Returns scala.None for no delay.

If attempts are exhausted the returned future is simply the result of invoking attempt. Note that the attempt function will be invoked on the given execution context for subsequent tries and therefore must be thread safe (i.e. not touch unsafe mutable state).

Example usage:

protected val sendAndReceive: HttpRequest => Future[HttpResponse]
protected val shouldRetry: (HttpResponse, Throwable) => throwable ne null
private val sendReceiveRetry: HttpRequest => Future[HttpResponse] = (req: HttpRequest) => retry[HttpResponse](
  attempt = () => sendAndReceive(req),
  shouldRetry,
  attempts = 10,
  delay = 2.seconds
)

Given a function from Unit to Future, returns an internally retrying Future. The first attempt will be made immediately, each subsequent attempt will be made after 'delay'. A scheduler (eg context.system.scheduler) must be provided to delay each retry.

If attempts are exhausted the returned future is simply the result of invoking attempt. Note that the attempt function will be invoked on the given execution context for subsequent tries and therefore must be thread safe (i.e. not touch unsafe mutable state).

Example usage:

protected val sendAndReceive: HttpRequest => Future[HttpResponse]
private val sendReceiveRetry: HttpRequest => Future[HttpResponse] = (req: HttpRequest) => retry[HttpResponse](
  attempt = () => sendAndReceive(req),
  attempts = 10,
  delay = 2.seconds
)

Given a function from Unit to Future, returns an internally retrying Future.

When the future is completed, the shouldRetry predicate is always been invoked with the result (or null if none) and the exception (or null if none). If the shouldRetry predicate returns true, then a new attempt is made, each subsequent attempt will be made after the 'delay' return by delayFunction (the input next attempt count start from 1). Returns scala.None for no delay.

If attempts are exhausted the returned future is simply the result of invoking attempt. Note that the attempt function will be invoked on the given execution context for subsequent tries and therefore must be thread safe (i.e. not touch unsafe mutable state).

Example usage:

protected val sendAndReceive: HttpRequest => Future[HttpResponse]
protected val shouldRetry: (HttpResponse, Throwable) => throwable ne null
private val sendReceiveRetry: HttpRequest => Future[HttpResponse] = (req: HttpRequest) => retry[HttpResponse](
  attempt = () => sendAndReceive(req),
  shouldRetry,
  attempts = 10,
  minBackoff = 1.seconds,
  maxBackoff = 2.seconds,
  randomFactor = 0.5
)

Given a function from Unit to Future, returns an internally retrying Future. The first attempt will be made immediately, each subsequent attempt will be made with a backoff time, if the previous attempt failed.

If attempts are exhausted the returned future is simply the result of invoking attempt. Note that the attempt function will be invoked on the given execution context for subsequent tries and therefore must be thread safe (i.e. not touch unsafe mutable state).

Example usage:

protected val sendAndReceive: HttpRequest => Future[HttpResponse]
private val sendReceiveRetry: HttpRequest => Future[HttpResponse] = (req: HttpRequest) => retry[HttpResponse](
  attempt = () => sendAndReceive(req),
  attempts = 10,
  minBackoff = 1.seconds,
  maxBackoff = 2.seconds,
  randomFactor = 0.5
)

Given a function from Unit to Future, returns an internally retrying Future.

When the future is completed, the shouldRetry predicate is always been invoked with the result (or null if none) and the exception (or null if none). If the shouldRetry predicate returns true, then a new attempt is made, each subsequent attempt will be made after the 'delay' return by delayFunction (the input next attempt count start from 1). Returns scala.None for no delay.

If attempts are exhausted the returned future is simply the result of invoking attempt. Note that the attempt function will be invoked on the given execution context for subsequent tries and therefore must be thread safe (i.e. not touch unsafe mutable state).

Example usage:

def possiblyFailing(): Future[Something] = ???
val shouldRetry: (Something, Throwable) => throwable ne null
val withRetry: Future[Something] = retry(attempt = possiblyFailing, shouldRetry, attempts = 10)

Given a function from Unit to Future, returns an internally retrying Future. The first attempt will be made immediately, each subsequent attempt will be made immediately if the previous attempt failed.

If attempts are exhausted the returned future is simply the result of invoking attempt. Note that the attempt function will be invoked on the given execution context for subsequent tries and therefore must be thread safe (i.e. not touch unsafe mutable state).

Example usage:

def possiblyFailing(): Future[Something] = ???
val withRetry: Future[Something] = retry(attempt = possiblyFailing, attempts = 10)

Java 8+ API for Pekko patterns such as ask, pipe and others which work with java.util.concurrent.CompletionStage.

For working with Scala scala.concurrent.Future from Java you may want to use pekko.pattern.Patterns instead.