[teiid-issues] [JBoss JIRA] Commented: (TEIID-601) Unable to connect to server from clients: session service is unresponsive (NPEs)

[Ramesh Reddy (JIRA)]

    [ https://jira.jboss.org/jira/browse/TEIID-601?page=com.atlassian.jira.plugin.system.issuetabpanels:comment-tabpanel&focusedCommentId=12467865#action_12467865 ] 

Ramesh Reddy commented on TEIID-601:
------------------------------------

Based on the conversation with shawkins, this is timing issue where the original process is not completely killed but a new process is invoked immediately. However, Teiid does not have any pid checking logic that checks for the duplicate process thus new duplicate process with same configuration is created, thus making the service remote where the NPE is thrown.

The fix is here is to create a pid file based on the process name when the process starts and obtain a lock on that file for the life time of the process. When duplicate process starts, it checks for existence of the pid file and tries to modify it, if successful, it deletes that one and creates a new one and proceeds. If can not, that means original process is still alive, then it will not start. 

As work around until then, make sure that all the processes are dead before starting a new process.

> Unable to connect to server from clients: session service is unresponsive (NPEs)
> --------------------------------------------------------------------------------
>
>                 Key: TEIID-601
>                 URL: https://jira.jboss.org/jira/browse/TEIID-601
>             Project: Teiid
>          Issue Type: Bug
>          Components: Server
>    Affects Versions: 6.1.0
>         Environment: Fedora 10, JBEDSP server snapshot from 13 May
>            Reporter: Paul Nittel
>            Assignee: Steven Hawkins
>            Priority: Critical
>             Fix For: 6.1.1
>
>         Attachments: console.log, pnittel_localdomain_MMProcess.log, SQuirreLExceptions.txt
>
>
> I've set up a JBEDSP server on my laptop. I have been able to connect from the Console, but not really able to do much querying. This morning, I am no longer able to connect to the server (I tried SQuirreL and Console). 
> Steve H. reviewed the server log and found the following to be central to the issue at hand. NOTE: I am NOT running this in  a cluster.
> Exception in thread "ServiceMonitoringThread" No response for remote method call
> 	at com.metamatrix.common.messaging.jgroups.JGroupsMessageBus$2.invoke(JGroupsMessageBus.java:145)
> 	at $Proxy27.checkState(Unknown Source)
> 	at com.metamatrix.platform.registry.ProcessMonitor$2.run(ProcessMonitor.java:125)
> 	at java.util.TimerThread.mainLoop(Timer.java:512)
> 	at java.util.TimerThread.run(Timer.java:462)
> May 14, 2009 16:15:48.437 [Worker3_SocketWorker1|0] ERROR <ROUTER|0> Unexpected exception for session 1900
> java.lang.NullPointerException
> 	at com.metamatrix.platform.registry.ServiceRegistryBinding.setService(ServiceRegistryBinding.java:298)
> 	at com.metamatrix.platform.registry.ServiceRegistryBinding.markServiceAsBad(ServiceRegistryBinding.java:314)
> 	at com.metamatrix.platform.service.proxy.SingleInvokationProxy.invoke(SingleInvokationProxy.java:89)
> 	at $Proxy24.validateSession(Unknown Source)
> 	at com.metamatrix.common.comm.platform.socket.server.ServerWorkItem.run(ServerWorkItem.java:93)
> 	at com.metamatrix.common.queue.WorkerPoolFactory$StatsCapturingSharedThreadPoolExecutor$1.run(WorkerPoolFactory.java:211)
> 	at java.util.concurrent.ThreadPoolExecutor$Worker.runTask(ThreadPoolExecutor.java:886)
> 	at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:908)
> 	at java.lang.Thread.run(Thread.java:619)

-- 
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