[jboss-jira] [JBoss JIRA] Closed: (JBMESSAGING-1623) Client failover causes thread leak

[Howard Gao (JIRA)]

     [ https://jira.jboss.org/jira/browse/JBMESSAGING-1623?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel ]

Howard Gao closed JBMESSAGING-1623.
-----------------------------------

    Resolution: Rejected


The leak is gone, as the JBM now using a thread pool to do the job.

So this bug is not valid.

> Client failover causes thread leak
> ----------------------------------
>
>                 Key: JBMESSAGING-1623
>                 URL: https://jira.jboss.org/jira/browse/JBMESSAGING-1623
>             Project: JBoss Messaging
>          Issue Type: Bug
>          Components: JMS Clustering
>    Affects Versions: 1.4.0.SP3.CP07, 1.4.2.GA.SP1, 1.4.3.GA
>            Reporter: Howard Gao
>            Assignee: Howard Gao
>            Priority: Critical
>             Fix For: 1.4.0.SP3.CP08, 1.4.4.GA
>
>
> In a clustered environment, a client failover will cause one thread leak per session, which will eventually used up system thread resources and cause outofmemory exception.
> The leak happens when client failover to a new connection, the old session states will be updated to new connection, where there will be a new SessionState created for each old one. During the creation of a new SessionState, a new JBExecutor will be created in which a new thread will be created and started. But it will never use it later as the existing executor in the old state will be used. So the newly created thread has nothing to do but stay there and never exits. As more client failover happens, more such threads will be created.
> The solution is simple. When the executor is created, son't start the worker thread untill it has tasks to do.

-- 
This message is automatically generated by JIRA.
-
If you think it was sent incorrectly contact one of the administrators: https://jira.jboss.org/jira/secure/Administrators.jspa
-
For more information on JIRA, see: http://www.atlassian.com/software/jira