Why "reaper-1" Thread keeps occupying highly CPU ?

[programTang]

My zmq application's "reaper-1" thread cost most cpu, so it makes the work thread send messages very slowly. jeromq version: 0.43 java version: 1.8.0_112

[fbacchella]

java version: 1.8.0_112.

That's Old. Do you have failling network connections ? With a lot of new TCP connexions created and closed ? That might happens when using Curve security

[scm-git]

Hi ZMQ team we encounter the same issuse, and use "jstack" command repeatly, can get the following two stack info :

"reaper-1" #12741 daemon prio=5 os_prio=0 tid=0x00007f38580c0000 nid=0x5fb5 runnable [0x00007f37a90e3000]
   java.lang.Thread.State: RUNNABLE
        at sun.nio.ch.FileDispatcherImpl.read0(Native Method)
        at sun.nio.ch.FileDispatcherImpl.read(FileDispatcherImpl.java:46)
        at sun.nio.ch.IOUtil.readIntoNativeBuffer(IOUtil.java:223)
        at sun.nio.ch.IOUtil.read(IOUtil.java:197)
        at sun.nio.ch.SourceChannelImpl.read(SourceChannelImpl.java:167)
        - locked <0x00000000d305be20> (a java.lang.Object)
        at zmq.Signaler.recv(Signaler.java:165)
        at zmq.Mailbox.recv(Mailbox.java:97)
        at zmq.SocketBase.processCommands(SocketBase.java:937)
        at zmq.SocketBase.inEvent(SocketBase.java:1065)
        at zmq.poll.Poller.run(Poller.java:273)
        at java.lang.Thread.run(Thread.java:748)

and

"reaper-1" #12741 daemon prio=5 os_prio=0 tid=0x00007f38580c0000 nid=0x5fb5 runnable [0x00007f37a90e3000]
   java.lang.Thread.State: RUNNABLE
        at sun.nio.ch.NativeThread.current(Native Method)
        at sun.nio.ch.SourceChannelImpl.read(SourceChannelImpl.java:165)
        - locked <0x00000000d305be20> (a java.lang.Object)
        at zmq.Signaler.recv(Signaler.java:165)
        at zmq.Mailbox.recv(Mailbox.java:97)
        at zmq.SocketBase.processCommands(SocketBase.java:937)
        at zmq.SocketBase.inEvent(SocketBase.java:1065)
        at zmq.poll.Poller.run(Poller.java:273)
        at java.lang.Thread.run(Thread.java:748)

and I do find the “while(true)” in Reaper.inEvent() method, but I don't know why it happened? and I found the similar issue: https://github.com/zeromq/jeromq/issues/172; but no resolve

[gaofei1982]

我也遇到这个情况，并且解决了，原因是ZMQ.socket被多线程操作了，你检查一下是否有类似使用场景。 I also encountered this situation and solved it because ZMQ.socket is operated by multiple threads. Check whether there are similar scenarios.

[nethibernate]

我也遇到这个情况，并且解决了，原因是ZMQ.socket被多线程操作了，你检查一下是否有类似使用场景。 I also encountered this situation and solved it because ZMQ.socket is operated by multiple threads. Check whether there are similar scenarios.

内部hashmap多线程死循环了？？ HashMap in multithread access causes infinite loop？

[fbacchella]

What do you espect us to do with comments not written in english ?