You are viewing a plain text version of this content. The canonical link for it is here.
Posted to issues@flex.apache.org by "Douglas Yeager (JIRA)" <ji...@apache.org> on 2015/11/30 16:51:10 UTC

[jira] [Created] (FLEX-34978) Shutdown errors when running Blazed DS inside of Tomcat & Java 8.

Douglas Yeager created FLEX-34978:
-------------------------------------

             Summary: Shutdown errors when running Blazed DS inside of Tomcat & Java 8.
                 Key: FLEX-34978
                 URL: https://issues.apache.org/jira/browse/FLEX-34978
             Project: Apache Flex
          Issue Type: Bug
          Components: .Unspecified - Framework
    Affects Versions: BlazeDS 4.7
         Environment: Tomcat 7.0.58
Java 8
            Reporter: Douglas Yeager


Inside flex.messaging.FlexSessionManager a ConcurrentHashMap member variable is declared and later the keySet() is accessed.  The ConcurrentHashMap interface has changed between Java 7 and Java 8.  So although it will work when running with Java 7; it will break if you run with Java 8.  It should be updated to declare the member as just a Map or ConcurrentMap (which is done elsewhere in the BlazeDS source); however, in the FlexSessionManager it is declared as:
    private final ConcurrentHashMap<Class<? extends FlexSession>, AbstractFlexSessionProvider> providers = new ConcurrentHashMap<Class<? extends FlexSession>, AbstractFlexSessionProvider>();

Then in the FlexSessionManager.stop() implementation it performs the following directly off of the ConcurrentHashMap which results in a failure:
    /**
     * Stops the <tt>FlexSessionManager</tt>.
     * Any registered <tt>FlexSession</tt> providers are stopped and unregistered. 
     */
    @Override
    public void stop()
    {
        if (!isStarted())
            return;
        
        super.stop();
        
        for (Class<? extends FlexSession> sessionClass : providers.keySet())
        {
            unregisterFlexSessionProvider(sessionClass);
        }
        providers.clear();
    }

The following is the stacktrace seen in the error log on shutdown:
2015-11-25 13:47:01,804 [localhost-startStop-2] ERROR org.apache.catalina.core.ContainerBase.[Catalina].[localhost].[/context] - Servlet MessageBrokerServlet threw unload() exception
javax.servlet.ServletException: Servlet.destroy() for servlet MessageBrokerServlet threw exception
    at org.apache.catalina.core.StandardWrapper.unload(StandardWrapper.java:1507)
    at org.apache.catalina.core.StandardWrapper.stopInternal(StandardWrapper.java:1847)
    at org.apache.catalina.util.LifecycleBase.stop(LifecycleBase.java:232)
    at org.apache.catalina.core.StandardContext.stopInternal(StandardContext.java:5707)
    at org.apache.catalina.util.LifecycleBase.stop(LifecycleBase.java:232)
    at org.apache.catalina.core.ContainerBase$StopChild.call(ContainerBase.java:1591)
    at org.apache.catalina.core.ContainerBase$StopChild.call(ContainerBase.java:1580)
    at java.util.concurrent.FutureTask.run(FutureTask.java:262)
    at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1145)
    at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:615)
    at java.lang.Thread.run(Thread.java:745)
Caused by: java.lang.NoSuchMethodError: java.util.concurrent.ConcurrentHashMap.keySet()Ljava/util/concurrent/ConcurrentHashMap$KeySetView;
    at flex.messaging.FlexSessionManager.stop(FlexSessionManager.java:293)
    at flex.messaging.MessageBroker.stop(MessageBroker.java:380)
    at flex.messaging.MessageBrokerServlet.destroy(MessageBrokerServlet.java:263)
    at org.apache.catalina.core.StandardWrapper.unload(StandardWrapper.java:1486)
    ... 10 more

There are two other classes that declare ConcurrentHashMap in the code that should be cleaned up too (although keySet() is not referenced in those classes , it would be better to address them to ensure this issue doesn't re-occur):
flex.messaging.endpoints.BaseStreamingHTTPEndpoint
flex.messaging.MessageBroker



--
This message was sent by Atlassian JIRA
(v6.3.4#6332)