[ http://issues.apache.org/jira/browse/GERONIMO-2079?page=all ]

David Jencks updated GERONIMO-2079:
-----------------------------------

    Attachment: GERONIMO-2079.failed1.patch

My attempts to fix this so far haven't worked.  I've attached a patch of my 
latest failing effort  and rolled the code back to its state before my first 
failed fix.

I believe the problem is that the actual race condition occurs in 
EJBProxyFactory rather than EJBMethodInterceptor, so a similar solution inside 
EJBProxyFactory might work.

IIUC this change resulted in a ~20% performance decrease from 1500 page 
views/sec to 1200.  If we can eliminate the volatile in even some circumstances 
it would seem worthwhile to do so.

> No methodProxy in EJBInvocation at startup under heavy load
> -----------------------------------------------------------
>
>          Key: GERONIMO-2079
>          URL: http://issues.apache.org/jira/browse/GERONIMO-2079
>      Project: Geronimo
>         Type: Bug
>     Security: public(Regular issues) 
>   Components: OpenEJB
>     Versions: 1.1
>     Reporter: David Jencks
>     Assignee: David Jencks
>      Fix For: 1.2
>  Attachments: GERONIMO-2079.failed1.patch
>
> After deploying daytrader, if you apply heavy load (50 threads) you get about 
> 60 exceptions like this:
> java.lang.NullPointerException
>         at 
> org.openejb.proxy.EJBMethodInterceptor.createEJBInvocation(EJBMethodInterceptor.java:203)
>         at 
> org.openejb.proxy.EJBMethodInterceptor.intercept(EJBMethodInterceptor.java:131)
>         at 
> org.openejb.proxy.EntityEJBLocalObject$$EnhancerByCGLIB$$84c9ae5f.getDataBean(<generated>)
>         at 
> org.apache.geronimo.samples.daytrader.web.prims.PingServlet2EntityLocal.doGet(PingServlet2EntityLocal.java:126)
>         at 
> org.apache.geronimo.samples.daytrader.web.prims.PingServlet2EntityLocal.doPost(PingServlet2EntityLocal.java:54)
>         at javax.servlet.http.HttpServlet.service(HttpServlet.java:615)
>         at javax.servlet.http.HttpServlet.service(HttpServlet.java:688)
>         at 
> org.apache.catalina.core.ApplicationFilterChain.internalDoFilter(ApplicationFilterChain.java:252)
>         at 
> org.apache.catalina.core.ApplicationFilterChain.doFilter(ApplicationFilterChain.java:173)
>         at 
> org.apache.catalina.core.StandardWrapperValve.invoke(StandardWrapperValve.java:213)
>         at 
> org.apache.catalina.core.StandardContextValve.invoke(StandardContextValve.java:178)
>         at 
> org.apache.geronimo.tomcat.valve.DefaultSubjectValve.invoke(DefaultSubjectValve.java:46)
>         at 
> org.apache.geronimo.tomcat.GeronimoStandardContext$SystemMethodValve.invoke(GeronimoStandardContext.java:342)
>         at 
> org.apache.geronimo.tomcat.valve.GeronimoBeforeAfterValve.invoke(GeronimoBeforeAfterValve.java:31)
>         at 
> org.apache.catalina.core.StandardHostValve.invoke(StandardHostValve.java:126)
>         at 
> org.apache.catalina.valves.ErrorReportValve.invoke(ErrorReportValve.java:105)
>         at 
> org.apache.catalina.core.StandardEngineValve.invoke(StandardEngineValve.java:107)
>         at 
> org.apache.catalina.connector.CoyoteAdapter.service(CoyoteAdapter.java:148)
>         at 
> org.apache.coyote.http11.Http11Processor.process(Http11Processor.java:869)
>         at 
> org.apache.coyote.http11.Http11BaseProtocol$Http11ConnectionHandler.processConnection(Http11BaseProtocol.java:667)
>         at 
> org.apache.tomcat.util.net.PoolTcpEndpoint.processSocket(PoolTcpEndpoint.java:527)
>         at 
> org.apache.tomcat.util.net.LeaderFollowerWorkerThread.runIt(LeaderFollowerWorkerThread.java:80)
>         at 
> org.apache.tomcat.util.threads.ThreadPool$ControlRunnable.run(ThreadPool.java:684)
>         at java.lang.Thread.run(Thread.java:534)                              
>                                                                               
>   
> Looking into the code, it appears that cglib is calling EJBMethodInterceptor 
> with a null methodProxy.  After these startup exceptions, which appear to 
> occur within the first 2 seconds of load, there are no further exceptions: i 
> ran it for 17 minutes at 1200 pages/sec.

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

Reply via email to