Akka round-robin: Sending response from remote routees to sender - scala

I am using Akka Cluster (version 2.4.10) with few nodes designated for "front-end" role and few others as "workers". The workers are on remote machines. The incoming work is distributed by the front-end actor to workers by round-robin routing. The issue is sending back the response from the "workers" back to the front-end actor. I can see that the work is getting completed by the workers. But the message sent by the workers to front-end does not reach and ends up as dead-letters. I see the below error in the log.
[Cluster-akka.actor.default-dispatcher-21] [akka://Cluster/deadLetters] Message [scala.collection.immutable.$colon$colon] from Actor[akka://Cluster/user] to Actor[akka://Cluster/deadLetters] was not delivered. [6] dead letters encountered.
I have seen this and I am following the same in my code. I have also seen this, but the solution suggested does not apply in this case, because I do not know the routees up-front. It comes through the configuration and it can change. The round-robin router configuration is as below.
akka.actor.deployment {
/frontEnd/hm = {
router = round-robin-group
nr-of-instances = 5
routees.paths = ["/user/hmWorker"]
cluster {
enabled = on
use-role = backend
allow-local-routees = on
}
}
}
The router is instantiated in front-end actor like below.
val router = context.actorOf(FromConfig.props(), name = "hm")
val controller = context.actorOf(Props(classOf[Controller], router))
The controller and the worker codes are below.
// Node 1 : Controller routes requests using round-robin
class Controller(router: ActorRef) extends Actor {
val list = List("a", "b") // Assume this is a big list
val groups = list.grouped(500)
override def receive: Actor.Receive = {
val futures = groups.map(grp => (router ? Message(grp)).mapTo[List[String]]))
val future = Future.sequence(futures).map(_.flatten)
val result = Await.result(future, 50 seconds)
println(s"Result is $result")
}
}
// Node 2
class Worker extends Actor {
override def receive: Actor.Receive = {
case Message(lst) =>
val future: Future[List[String]] = // Do Something asynchronous
future onComplete {
case Success(r) => sender.!(r)(context.parent) // This message is not delivered to Controller actor.
case Failure(th) => // Error handling
}
}
}
Please let me know what I am doing wrong here. Appreciate your help.

You shouldn't use sender() in the callback on a Future. By the time the callback is processed, the sender() is likely referring to something different than it was when you received the message.
Consider either saving the reference outside of the callback first like:
override def receive: Actor.Receive = {
case Message(lst) =>
val future: Future[List[String]] = // Do Something asynchronous
val replyTo: ActorRef = sender()
future onComplete {
case Success(r) => replyTo.!(r)(context.parent) // This message is not delivered to Controller actor.
case Failure(th) => // Error handling
}
}
Or even better, use the pipe pattern:
import akka.pattern.pipe
override def receive: Actor.Receive = {
case Message(lst) =>
val future: Future[List[String]] = // Do Something asynchronous
future.pipeTo(sender())
}

Related

Akka Supervisor Strategy - Correct Use Case

I have been using Akka Supervisor Strategy to handle business logic exceptions.
Reading one of the most famous Scala blog series Neophyte, I found him giving a different purpose for what I have always been doing.
Example:
Let's say I have an HttpActor that should contact an external resource and in case it's down, I will throw an Exception, for now a ResourceUnavailableException.
In case my Supervisor catches that, I will call a Restart on my HttpActor, and in my HttpActor preRestart method, I will call do a schedulerOnce to retry that.
The actor:
class HttpActor extends Actor with ActorLogging {
implicit val system = context.system
override def preRestart(reason: Throwable, message: Option[Any]): Unit = {
log.info(s"Restarting Actor due: ${reason.getCause}")
message foreach { msg =>
context.system.scheduler.scheduleOnce(10.seconds, self, msg)
}
}
def receive = LoggingReceive {
case g: GetRequest =>
doRequest(http.doGet(g), g.httpManager.url, sender())
}
A Supervisor:
class HttpSupervisor extends Actor with ActorLogging with RouterHelper {
override val supervisorStrategy =
OneForOneStrategy(maxNrOfRetries = 5) {
case _: ResourceUnavailableException => Restart
case _: Exception => Escalate
}
var router = makeRouter[HttpActor](5)
def receive = LoggingReceive {
case g: GetRequest =>
router.route(g, sender())
case Terminated(a) =>
router = router.removeRoutee(a)
val r = context.actorOf(Props[HttpActor])
context watch r
router = router.addRoutee(r)
}
}
What's the point here?
In case my doRequest method throws the ResourceUnavailableException, the supervisor will get that and restart the actor, forcing it to resend the message after some time, according to the scheduler. The advantages I see is the fact I get for free the number of retries and a nice way to handle the exception itself.
Now looking at the blog, he shows a different approach in case you need a retry stuff, just sending messages like this:
def receive = {
case EspressoRequest =>
val receipt = register ? Transaction(Espresso)
receipt.map((EspressoCup(Filled), _)).recover {
case _: AskTimeoutException => ComebackLater
} pipeTo(sender)
case ClosingTime => context.system.shutdown()
}
Here in case of AskTimeoutException of the Future, he pipes the result as a ComebackLater object, which he will handle doing this:
case ComebackLater =>
log.info("grumble, grumble")
context.system.scheduler.scheduleOnce(300.millis) {
coffeeSource ! EspressoRequest
}
For me this is pretty much what you can do with the strategy supervisor, but in a manually way, with no built in number of retries logic.
So what is the best approach here and why? Is my concept of using akka supervisor strategy completely wrong?
You can use BackoffSupervisor:
Provided as a built-in pattern the akka.pattern.BackoffSupervisor implements the so-called exponential backoff supervision strategy, starting a child actor again when it fails, each time with a growing time delay between restarts.
val supervisor = BackoffSupervisor.props(
Backoff.onFailure(
childProps,
childName = "myEcho",
minBackoff = 3.seconds,
maxBackoff = 30.seconds,
randomFactor = 0.2 // adds 20% "noise" to vary the intervals slightly
).withAutoReset(10.seconds) // the child must send BackoffSupervisor.Reset to its parent
.withSupervisorStrategy(
OneForOneStrategy() {
case _: MyException => SupervisorStrategy.Restart
case _ => SupervisorStrategy.Escalate
}))

Kill actor if it times out in Spray app

In my Spray app, I delegate requests to actors. I want to be able to kill a actor that takes too long. I'm not sure whether I should be using Spray timeouts, Akka ask pattern or something else.
I have implemented:
def processRouteRequest(system: ActorSystem) = {
respondWithMediaType(`text/json`) {
params { p => ctx =>
val builder = newBuilderActor
builder ! Request(p) // the builder calls `ctx.complete`
builder ! PoisonPill
system.scheduler.scheduleOnce(routeRequestMaxLife, builder, Kill)
}
}
}
The idea being that the actor lives only for the duration of a single request and if it doesn't complete within routeRequestMaxLife it gets forcibly killed. This approach seems over-the-top (and spews a lot of info about undelivered messages). I'm not even certain it works correctly.
It seems like what I'm trying to achieve should be a common use-case. How should I approach it?
I would tend to using the ask pattern and handling the requests as follows:
class RequestHandler extends Actor {
def receive = {
case "quick" =>
sender() ! "Quick Reply"
self ! PoisonPill
case "slow" =>
val replyTo = sender()
context.system.scheduler.scheduleOnce(5 seconds, self, replyTo)
case a:ActorRef =>
a ! "Slow Reply"
self ! PoisonPill
}
}
class ExampleService extends HttpService with Actor {
implicit def actorRefFactory = context
import context.dispatcher
def handleRequest(mode: String):Future[String] = {
implicit val timeout = Timeout(1 second)
val requestHandler = context.actorOf(Props[RequestHandler])
(requestHandler ? mode).mapTo[String]
}
val route: Route =
path("endpoint" / Segment) { str =>
get {
onComplete(handleRequest(str)) {
case Success(str) => complete(str)
case Failure(ex) => complete(ex)
}
}
}
def receive = runRoute(route)
}
This way the actor takes care of stopping itself, and the semantics of Ask give you the information about whether or not the request timed out.

Spray route get response from child actor

I am trying to figure out how I can setup a Master Actor that calls the appropriate children, in support of some spray routes where I am trying to emulate db calls. I am new to akka / spray, so just trying to gain a better understanding of how you would properly setup spray -> actors -> db calls (etc.). I can get the response back from the top level actor, but when I try to get it back from one actor level below the parent I can't seem to get anything to work.
When looking at the paths of the actors, it appears that from the way I am making the call from my spray route that I am passing from a temp actor. Below is what I have so far for stubbing this out. This has to be just user error / ignorance, just not sure how to proceed. Any suggestions would be appreciated.
The Demo Spray Service and Redis Actor code snippets below show where I am calling the actor from my route and the multiple actors where I am having the issue (want my route to get response from SummaryActor). Thanks!
Boot:
object Boot extends App {
// we need an ActorSystem to host our application in
implicit val system = ActorSystem("on-spray-can")
// create and start our service actor
val service = system.actorOf(Props[DemoServiceActor], "demo-service")
implicit val timeout = Timeout(5.seconds)
// start a new HTTP server on port 8080 with our service actor as the handler
IO(Http) ? Http.Bind(service, interface = "localhost", port = 8080)
}
Demo Service Actor (For Spray)
class DemoServiceActor extends Actor with Api {
// the HttpService trait defines only one abstract member, which
// connects the services environment to the enclosing actor or test
def actorRefFactory = context
// this actor only runs our route, but you could add
// other things here, like request stream processing
// or timeout handling
def receive = handleTimeouts orElse runRoute(route)
//Used to watch for request timeouts
//http://spray.io/documentation/1.1.2/spray-routing/key-concepts/timeout-handling/
def handleTimeouts: Receive = {
case Timedout(x: HttpRequest) =>
sender ! HttpResponse(StatusCodes.InternalServerError, "Too late")
}
}
//Master trait for handling large APIs
//http://stackoverflow.com/questions/14653526/can-spray-io-routes-be-split-into-multiple-controllers
trait Api extends DemoService {
val route = {
messageApiRouting
}
}
Demo Spray Service (Route):
trait DemoService extends HttpService with Actor {
implicit val timeout = Timeout(5 seconds) // needed for `?` below
val redisActor = context.actorOf(Props[RedisActor], "redisactor")
val messageApiRouting =
path("summary" / Segment / Segment) { (dataset, timeslice) =>
onComplete(getSummary(redisActor, dataset, timeslice)) {
case Success(value) => complete(s"The result was $value")
case Failure(ex) => complete(s"An error occurred: ${ex.getMessage}")
}
}
def getSummary(redisActor: ActorRef, dataset: String, timeslice: String): Future[String] = Future {
val dbMessage = DbMessage("summary", dataset + timeslice)
val future = redisActor ? dbMessage
val result = Await.result(future, timeout.duration).asInstanceOf[String]
result
}
}
Redis Actor (Mock no actual redis client yet)
class RedisActor extends Actor with ActorLogging {
// val pool = REDIS
implicit val timeout = Timeout(5 seconds) // needed for `?` below
val summaryActor = context.actorOf(Props[SummaryActor], "summaryactor")
def receive = {
case msg: DbMessage => {
msg.query match {
case "summary" => {
log.debug("Summary Query Request")
log.debug(sender.path.toString)
summaryActor ! msg
}
}
}
//If not match log an error
case _ => log.error("Received unknown message: {} ")
}
}
class SummaryActor extends Actor with ActorLogging{
def receive = {
case msg: DbMessage =>{
log.debug("Summary Actor Received Message")
//Send back to Spray Route
}
}
}
The first problem with your code is that you need to forward from the master actor to the child so that the sender is properly propagated and available for the child to respond to. So change this (in RedisActor):
summaryActor ! msg
To:
summaryActor forward msg
That's the primary issue. Fix that and your code should start working. There is something else that needs attention though. Your getSummary method is currently defined as:
def getSummary(redisActor: ActorRef, dataset: String, timeslice: String): Future[String] =
Future {
val dbMessage = DbMessage("summary", dataset + timeslice)
val future = redisActor ? dbMessage
val result = Await.result(future, timeout.duration).asInstanceOf[String]
result
}
The issue here is that the ask operation (?) already returns a Future, so there and you are blocking on it to get the result, wrapping that in another Future so that you can return a Future for onComplete to work with. You should be able to simplify things by using the Future returned from ask directly like so:
def getSummary(redisActor: ActorRef, dataset: String, timeslice: String): Future[String] = {
val dbMessage = DbMessage("summary", dataset + timeslice)
(redisActor ? dbMessage).mapTo[String]
}
Just an important comment on the above approaches.
Since the getSummary(...) function returns a Future[String] object and you call it in onComplete(...) function you need to import:
import ExecutionContext.Implicits.global
That way you will have ExecutionContext in scope by letting Future
declare an implicit ExecutionContext parameter.
** If you don't, you will end up getting a mismatching error
since onComplete(...) expects an onComplete Future
magnet Object but you gave a Future[String] Object.

Subscribing multiple actors to Dead Letters in Akka

I am trying to create a simple application that has two actors:
Master actor that handles some App actions
DeadLettersListener that is supposed to handle all dead or unhandled messages
Here is the code that works perfectly:
object Hw extends App {
// creating Master actor
val masterActorSystem = ActorSystem("Master")
val master = masterActorSystem.actorOf(Props[Master], "Master")
// creating Dead Letters listener actor
val deadLettersActorSystem = ActorSystem.create("DeadLettersListener")
val listener = deadLettersActorSystem.actorOf(Props[DeadLettersListener])
// subscribe listener to Master's DeadLetters
masterActorSystem.eventStream.subscribe(listener, classOf[DeadLetter])
masterActorSystem.eventStream.subscribe(listener, classOf[UnhandledMessage])
}
According to the akka manual though, ActorSystem is a heavy object and we should create only one per application. But when I replace these lines:
val deadLettersActorSystem = ActorSystem.create("DeadLettersListener")
val listener = deadLettersActorSystem.actorOf(Props[DeadLettersListener])
with this code:
val listener = masterActorSystem.actorOf(Props[DeadLettersListener], "DeadLettersListener")
The subscription does not work any more and DeadLettersListener is not getting any Dead or Unhandled messages.
Can you please explain what am I doing wrong and give an advice how to subscribe to Dead Letters in this case?
I can't really imagine what are you doing wrong, I created a small example, and it seems to work:
object Hw extends App {
class Master extends Actor {
override def receive: Receive = {
case a => println(s"$a received in $self")
}
}
class DeadLettersListener extends Actor {
override def receive: Actor.Receive = {
case a => println(s"$a received in $self")
}
}
// creating Master actor
val masterActorSystem = ActorSystem("Master")
val master = masterActorSystem.actorOf(Props[Master], "Master")
val listener = masterActorSystem.actorOf(Props[DeadLettersListener])
// subscribe listener to Master's DeadLetters
masterActorSystem.eventStream.subscribe(listener, classOf[DeadLetter])
masterActorSystem.eventStream.subscribe(listener, classOf[UnhandledMessage])
masterActorSystem.actorSelection("/unexistingActor") ! "yo"
}
Could you try it?

Scala Akka Consumer/Producer: Return Value

Problem Statement
Assume I have a file with sentences that is processed line by line. In my case, I need to extract Named Entities (Persons, Organizations, ...) from these lines. Unfortunately, the tagger is quite slow. Therefore, I decided to parallelize the computation, such that lines could be processed independent from each other and the result is collected in a central location.
Current Approach
My current approach comprises the usage of a single producer multiple consumer concept. However, I'm relative new to Akka, but I think my problem description fits well into its capabilities. Let me show you some code:
Producer
The Producer reads the file line by line and sends it to the Consumer. If it reaches the total line limit, it propagates the result back to WordCount.
class Producer(consumers: ActorRef) extends Actor with ActorLogging {
var master: Option[ActorRef] = None
var result = immutable.List[String]()
var totalLines = 0
var linesProcessed = 0
override def receive = {
case StartProcessing() => {
master = Some(sender)
Source.fromFile("sent.txt", "utf-8").getLines.foreach { line =>
consumers ! Sentence(line)
totalLines += 1
}
context.stop(self)
}
case SentenceProcessed(list) => {
linesProcessed += 1
result :::= list
//If we are done, we can propagate the result to the creator
if (linesProcessed == totalLines) {
master.map(_ ! result)
}
}
case _ => log.error("message not recognized")
}
}
Consumer
class Consumer extends Actor with ActorLogging {
def tokenize(line: String): Seq[String] = {
line.split(" ").map(_.toLowerCase)
}
override def receive = {
case Sentence(sent) => {
//Assume: This is representative for the extensive computation method
val tokens = tokenize(sent)
sender() ! SentenceProcessed(tokens.toList)
}
case _ => log.error("message not recognized")
}
}
WordCount (Master)
class WordCount extends Actor {
val consumers = context.actorOf(Props[Consumer].
withRouter(FromConfig()).
withDispatcher("consumer-dispatcher"), "consumers")
val producer = context.actorOf(Props(new Producer(consumers)), "producer")
context.watch(consumers)
context.watch(producer)
def receive = {
case Terminated(`producer`) => consumers ! Broadcast(PoisonPill)
case Terminated(`consumers`) => context.system.shutdown
}
}
object WordCount {
def getActor() = new WordCount
def getConfig(routerType: String, dispatcherType: String)(numConsumers: Int) = s"""
akka.actor.deployment {
/WordCount/consumers {
router = $routerType
nr-of-instances = $numConsumers
dispatcher = consumer-dispatcher
}
}
consumer-dispatcher {
type = $dispatcherType
executor = "fork-join-executor"
}"""
}
The WordCount actor is responsible for creating the other actors. When the Consumer is finished the Producer sends a message with all tokens. But, how to propagate the message again and also accept and wait for it? The architecture with the third WordCount actor might be wrong.
Main Routine
case class Run(name: String, actor: () => Actor, config: (Int) => String)
object Main extends App {
val run = Run("push_implementation", WordCount.getActor _, WordCount.getConfig("balancing-pool", "Dispatcher") _)
def execute(run: Run, numConsumers: Int) = {
val config = ConfigFactory.parseString(run.config(numConsumers))
val system = ActorSystem("Counting", ConfigFactory.load(config))
val startTime = System.currentTimeMillis
system.actorOf(Props(run.actor()), "WordCount")
/*
How to get the result here?!
*/
system.awaitTermination
System.currentTimeMillis - startTime
}
execute(run, 4)
}
Problem
As you see, the actual problem is to propagate the result back to the Main routine. Can you tell me how to do this in a proper way? The question is also how to wait for the result until the consumers are finished? I had a brief look into the Akka Future documentation section, but the whole system is a little bit overwhelming for beginners. Something like var future = message ? actor seems suitable. Not sure, how to do this. Also using the WordCount actor causes additional complexity. Maybe it is possible to come up with a solution that doesn't need this actor?
Consider using the Akka Aggregator Pattern. That takes care of the low-level primitives (watching actors, poison pill, etc). You can focus on managing state.
Your call to system.actorOf() returns an ActorRef, but you're not using it. You should ask that actor for results. Something like this:
implicit val timeout = Timeout(5 seconds)
val wCount = system.actorOf(Props(run.actor()), "WordCount")
val answer = Await.result(wCount ? "sent.txt", timeout.duration)
This means your WordCount class needs a receive method that accepts a String message. That section of code should aggregate the results and tell the sender(), like this:
class WordCount extends Actor {
def receive: Receive = {
case filename: String =>
// do all of your code here, using filename
sender() ! results
}
}
Also, rather than blocking on the results with Await above, you can apply some techniques for handling Futures.