[jboss-jira] [JBoss JIRA] (REMJMX-166) IllegalThreadStateException after idle jmx connection

[Märt Bakhoff (Jira)]

Märt Bakhoff created REMJMX-166:
-----------------------------------

             Summary: IllegalThreadStateException after idle jmx connection
                 Key: REMJMX-166
                 URL: https://issues.jboss.org/browse/REMJMX-166
             Project: Remoting JMX
          Issue Type: Bug
          Components: Connection
    Affects Versions: 3.0.3.Final, 3.0.2.Final
         Environment: org.jboss.remotingjmx:remoting-jmx:3.0.3.Final + java8
            Reporter: Märt Bakhoff
            Assignee: Darran Lofthouse


Start wildfly-17.0.1/bin/standalone.sh, then run this code snippet: 

{noformat}
JMXServiceURL url = new JMXServiceURL("service:jmx:remote+http://127.0.0.1:9990");
try (JMXConnector connector = new RemotingConnectorProvider().newJMXConnector(url, Collections.emptyMap())) {
  connector.connect();
  MBeanServerConnection beanServer = connector.getMBeanServerConnection();
  RuntimeMXBean bean = ManagementFactory.newPlatformMXBeanProxy(beanServer, ManagementFactory.RUNTIME_MXBEAN_NAME, RuntimeMXBean.class);
  Thread.sleep(70_000);
  System.out.println("uptime: " + bean.getUptime());
}
{noformat}

The following exception is always thrown:
{noformat}
Exception in thread "XNIO-1 task-12" java.lang.IllegalThreadStateException
	at java.lang.ThreadGroup.addUnstarted(ThreadGroup.java:867)
	at java.lang.Thread.init(Thread.java:405)
	at java.lang.Thread.init(Thread.java:349)
	at java.lang.Thread.<init>(Thread.java:599)
	at org.jboss.remotingjmx.protocol.v2.ClientExecutorManager$1.newThread(ClientExecutorManager.java:56)
	at java.util.concurrent.ThreadPoolExecutor$Worker.<init>(ThreadPoolExecutor.java:619)
	at java.util.concurrent.ThreadPoolExecutor.addWorker(ThreadPoolExecutor.java:932)
	at java.util.concurrent.ThreadPoolExecutor.execute(ThreadPoolExecutor.java:1378)
	at org.jboss.remotingjmx.protocol.v2.ClientExecutorManager.execute(ClientExecutorManager.java:64)
	at org.jboss.remotingjmx.protocol.v2.ClientCommon$MessageReceiver.handleMessage(ClientCommon.java:118)
	at org.jboss.remoting3.remote.RemoteConnectionChannel.lambda$handleMessageData$3(RemoteConnectionChannel.java:430)
	at org.jboss.remoting3.EndpointImpl$TrackingExecutor.lambda$execute$0(EndpointImpl.java:926)
	at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1149)
	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:624)
	at java.lang.Thread.run(Thread.java:748)
{noformat}

The cause is in org.jboss.remotingjmx.protocol.v2.ClientExecutorManager.<init>. It creates a thread pool with Executors.newCachedThreadPool that has the default keepAliveTime of 60s.
The thread factory is using a daemon thread group REMOTING_JMX that will self-destruct when the cached thread is terminated.

The same code works when using older org.jboss.remotingjmx:remoting-jmx:3.0.1.Final. The regression is likely caused by commit https://github.com/jbossas/remoting-jmx/commit/2d6ae6c26da43304b752fc48f15bdefe445466e4 



--
This message was sent by Atlassian Jira
(v7.12.1#712002)