Hi, We have an application which uses jacorb 2.3.1 for java to C++ communication. What we see on AIX-WAS environment is randomly the jacorb calls hang and bring down the container with below stack all the time (last 4-5 lines). Is there any fix for this? 3XMTHREADINFO "WebContainer : 191" J9VMThread:0x0000000038A75C00, j9thread_t:0x0000010047E4C280, java/lang/Thread:0x000000078442F498, state:CW, prio=5 3XMJAVALTHREAD (java/lang/Thread getId:0x30525, isDaemon:true) 3XMTHREADINFO1 (native thread ID:0x1E502BF, native priority:0x5, native policy:UNKNOWN) 3XMTHREADINFO3 Java callstack: 4XESTACKTRACE at java/lang/Object.wait(Native Method) 4XESTACKTRACE at java/lang/Object.wait(Object.java:196(Compiled Code)) 4XESTACKTRACE at org/jacorb/orb/giop/ReplyPlaceholder.getInputStream(ReplyPlaceholder.java:113(Compiled Code)) 4XESTACKTRACE at org/jacorb/orb/ReplyReceiver.getReply(ReplyReceiver.java:275(Compiled Code)) 4XESTACKTRACE at org/jacorb/orb/Delegate.invoke_internal(Delegate.java:1074(Compiled Code)) 4XESTACKTRACE at org/jacorb/orb/Delegate.invoke(Delegate.java:939(Compiled Code)) 4XESTACKTRACE at org/omg/CORBA/portable/ObjectImpl._invoke(ObjectImpl.java:484(Compiled Code)) 4XESTACKTRACE at com/manh/wmos/services/corba/_IDLBaseFactory_IFStub.is_alive(Bytecode PC:19(Compiled Code)) 4XESTACKTRACE at com/manh/ils/wmservices/corba/CorbaInitialContextFactory.createFactory(CorbaInitialContextFactory.java:550(Compiled Code)) 4XESTACKTRACE at com/manh/ils/wmservices/corba/CorbaInitialContextFactory.getFactoryObject(CorbaInitialContextFactory.java:457(Compiled Code)) 4XESTACKTRACE at com/manh/ils/wmservices/corba/CorbaInitialContextFactory.createLoginObject(CorbaInitialContextFactory.java:325(Compiled Code)) 4XESTACKTRACE at com/manh/ils/wmservices/corba/CorbaInitialContextFactory.initializeLoginObject(CorbaInitialContextFactory.java:275(Compiled Code)) 4XESTACKTRACE at com/manh/ils/wmservices/corba/CorbaInitialContextFactory.getLoginObject(CorbaInitialContextFactory.java:192(Compiled Code)) 4XESTACKTRACE at com/manh/ils/wmservices/corba/CorbaFactory.getLoginObject(CorbaFactory.java:473(Compiled Code)) 4XESTACKTRACE at com/manh/ils/wmservices/corba/CorbaHelper.getInitializedLoginObject(CorbaHelper.java:1716(Compiled Code)) 4XESTACKTRACE at com/manh/wmos/services/wmcommon/service/RFLoginUserServiceImpl.getLoginObject(RFLoginUserServiceImpl.java:517(Compiled Code)) 4XESTACKTRACE at com/manh/wmos/services/wmcommon/ejb/WMCommonServicesSessionBean.getLoginObject(WMCommonServicesSessionBean.java:1198(Compiled Code)) 4XESTACKTRACE at com/manh/wmos/services/wmcommon/ejb/EJSRemoteStatelessWMCommonServices_5eb30bf6.getLoginObject(Bytecode PC:83(Compiled Code)) 4XESTACKTRACE at com/manh/wmos/services/wmcommon/ejb/_WMCommonServices_Stub.getLoginObject(_WMCommonServices_Stub.java:1915(Compiled Code)) 4XESTACKTRACE at com/manh/wmos/services/wmcommon/bd/RFLoginUserBDImpl.getLoginObject(RFLoginUserBDImpl.java:367(Compiled Code)) 4XESTACKTRACE at com/manh/wmos/services/inventorymgmt/backingbean/RFMenuBackingBean.getClearTaskMgmtHandle(RFMenuBackingBean.java:141(Compiled Code)) 4XESTACKTRACE at com/manh/wmos/services/inventorymgmt/backingbean/RFMenuBackingBean.loginAction(RFMenuBackingBean.java:1112) Thank you, Jacob
Please retest with the current version (3.4) and ideally submit a test case.
Hi, Moving from 2.3.1 to a higher version may not be possible for us because of the build issues. We can not reproduce this at will this happens on production server randomly and brings the container down. Some times one day some times a week. Thank you, jacob
There were a considerable amount of fixes and improvements between 2.3.1 and 3.4. One of those may well resolve your problem. You could try utilising Byteman or attaching a debugger to understand what is happening.
What "build issues" are there involved in migrating?
The build issue I was thinking of was about the major version change from 2.3.1 to 3.*. I am not sure how many API's changed. I will give it a try and see.
one question since we are moving from 2.3 to 3.4 should we build and regenerate anything or just replacing the jar file will do?
Although I think dropping the new jars should work (note there are different jars now) I would advise regenerating your IDL.
I am closing this for now; if you can reproduce it on the current 3.5(SNAPSHOT) codebase then please reopen (and ideally supply a test case).