[英]Akka - Fixed thread pool for Blocking Operations - Blocking Dispatcher
我已经为我的应用程序将要执行的阻塞操作配置了一个调度程序,如下所示:
engine {
blocking-io-dispatcher {
type = Dispatcher
executor = "thread-pool-executor"
thread-pool-executor {
fixed-pool-size = 3
}
throughput = 1
}
}
但是,我可以在日志中看到每 5 秒有 3 个以上的线程正在处理请求,(线程 6、16、5、7、15)
控制台输出:
system-engine.blocking-io-dispatcher-6
system-engine.blocking-io-dispatcher-16
system-engine.blocking-io-dispatcher-5
system-engine.blocking-io-dispatcher-7
system-engine.blocking-io-dispatcher-15
请在下面找到代码:
import akka.actor.Actor
import akka.actor.ActorSystem
import akka.actor.Props
import java.util.Date
import akka.routing.RoundRobinRouter
import akka.routing.DefaultResizer
class SendPushNotificationActor extends Actor {
def receive = {
case SendPushNotificationMessage(message) => {
println(Thread.currentThread().getName)
// Simulate blocking operation which takes 5 seconds to send the push notification message
Thread.sleep(5000)
}
}
}
object SendPushNotificationTest {
def main(args: Array[String]): Unit = {
val system = ActorSystem("system")
val sendPushNotificationActor = system.actorOf(Props[SendPushNotificationActor].withDispatcher("engine.blocking-io-dispatcher")
.withRouter(RoundRobinRouter(nrOfInstances = 5)))
for (i <- 1.to(100)) {
sendPushNotificationActor ! SendPushNotificationMessage("Push Notification Message")
}
}
}
case class SendPushNotificationMessage(message: String)
任何有关如何修复池的帮助将不胜感激。
池中的线程可能由于某种原因死亡,在这种情况下,可以立即产生新线程以替换死线程并使池保持固定大小。 新线程被赋予递增的 ID,这是您在日志文件中看到的。 线程池的大小肯定是固定的。
声明:本站的技术帖子网页,遵循CC BY-SA 4.0协议,如果您需要转载,请注明本站网址或者原文地址。任何问题请咨询:yoyou2525@163.com.