Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

HttpTunnel problem

7 views
Skip to first unread message

kor...@gmail.com

unread,
Nov 16, 2006, 11:25:43 AM11/16/06
to
Hi,

Since two of some well-known testing tools can not detect RMI packets
in the network, we have been forced to use HTTP Tunneling for testing
thick application clients. Since we could not find any documentation to
do this for v6.1, we followed the article for v5.1 given below:

http://www-128.ibm.com/developerworks/forums/dw_thread.jsp?forum=266&...

Environment:
Server:
WAS 6.1, Network Deployment
Redhat Enterprise 4
Hibernate 3.0 with JPA and session beans are used.

Client:
Windows 2001,
IBM JDK,
Some jar files in client container to reach server
Thick Application client

When we run the application without tunneling, it works. And when we
use tunneling by specifying following properties on thick clients (also
setting FragmentSize as a custom property on server), user logins,
retrieves data. views data, but can not perform modifications to data. Our
method to verify the data before insert/update fails giving a
CORBA.COMM_FAILURE error.

com.ibm.CORBA.ForceTunnel=ALWAYS
com.ibm.CORBA.TunnelAgentURL=http://server:9080/HTTPTunnelWeb/IIOPTunnelServlet
com.ibm.CORBA.FragmentSize=0

Since it works without tunneling, I think that the problem is with
Websphere and/or tunneling. Has anyone come across a similar problem or
have an idea to resolve it.

Thanks for any help

Error Messages on Server:

4.Kas.2006 17:35:32 com.milsoft.logger.LoggerImpl error

SEVERE: Uncaught exception:

java.lang.ThreadDeath

at java.lang.Thread.stop(Thread.java:991)

at
com.ibm.ws.orbimpl.transport.WSHTTPTransportConnection.removeConnection(WSHTTPTransportConnection.java:140)

at
com.ibm.rmi.iiop.Connection.purge_calls(Connection.java:1883)

at
com.ibm.rmi.iiop.Connection.doReaderWorkOnce(Connection.java:3036)

at
com.ibm.ws.orbimpl.transport.HTTPReaderThread.run(WSHTTPTransportConnection.java:448)

14.Kas.2006 17:35:32 com.milsoft.logger.LoggerImpl error

SEVERE: Uncaught exception:

com.milsoft.uigen.exception.ApplicationTemplateException

at
com.milsoft.uigen.service.RemoteSessionHandler.callEjbMethod(RemoteSessionHandler.java:111)

at
com.milsoft.uigen.service.RemoteSessionHandler.invoke(RemoteSessionHandler.java:141)

at $Proxy1.validateUniqueData(Unknown Source)

at
com.milsoft.gui.creator.editor.swing.GenericEditorGenerator.validateSwingData(GenericEditorGenerator.java:463)

at
com.milsoft.uigen.template.window.MainInternalEditorWindow.onSaveClick(MainInternalEditorWindow.java:277)

at
com.milsoft.app.common.gui.BaseInternalEditorWindow.actionPerformed(BaseInternalEditorWindow.java:402)

at
com.milsoft.app.common.gui.BaseActionListener.actionPerformed(BaseActionListener.java:48)

at
javax.swing.AbstractButton.fireActionPerformed(AbstractButton.java:1879)

at
javax.swing.AbstractButton$Handler.actionPerformed(AbstractButton.java:2199)

at
javax.swing.DefaultButtonModel.fireActionPerformed(DefaultButtonModel.java:450)

at
javax.swing.DefaultButtonModel.setPressed(DefaultButtonModel.java:288)

at
javax.swing.plaf.basic.BasicButtonListener.mouseReleased(BasicButtonListener.java:264)

at
java.awt.AWTEventMulticaster.mouseReleased(AWTEventMulticaster.java:263)

at java.awt.Component.processMouseEvent(Component.java:5529)

at
javax.swing.JComponent.processMouseEvent(JComponent.java:3148)

at java.awt.Component.processEvent(Component.java:5294)

at java.awt.Container.processEvent(Container.java:2001)

at java.awt.Component.dispatchEventImpl(Component.java:3996)

at java.awt.Container.dispatchEventImpl(Container.java:2059)

at java.awt.Component.dispatchEvent(Component.java:3844)

at
java.awt.LightweightDispatcher.retargetMouseEvent(Container.java:4249)

at
java.awt.LightweightDispatcher.processMouseEvent(Container.java:3929)

at
java.awt.LightweightDispatcher.dispatchEvent(Container.java:3859)

at java.awt.Container.dispatchEventImpl(Container.java:2045)

at java.awt.Window.dispatchEventImpl(Window.java:1810)

at java.awt.Component.dispatchEvent(Component.java:3844)

at java.awt.EventQueue.dispatchEvent(EventQueue.java:545)

at
java.awt.EventDispatchThread.pumpOneEventForHierarchy(EventDispatchThread.java:268)

at
java.awt.EventDispatchThread.pumpEventsForHierarchy(EventDispatchThread.java:197)

at
java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:191)

at
java.awt.EventDispatchThread.pumpEvents(EventDispatchThread.java:183)

at
java.awt.EventDispatchThread.run(EventDispatchThread.java:144)

Caused by: java.lang.reflect.InvocationTargetException

at sun.reflect.NativeMethodAccessorImpl.invoke0(Native Method)

at
sun.reflect.NativeMethodAccessorImpl.invoke(NativeMethodAccessorImpl.java:64)

at
sun.reflect.DelegatingMethodAccessorImpl.invoke(DelegatingMethodAccessorImpl.java:43)

at java.lang.reflect.Method.invoke(Method.java:615)

at
com.milsoft.uigen.service.RemoteSessionHandler.callEjbMethod(RemoteSessionHandler.java:107)

... 31 more

Caused by: java.rmi.MarshalException: CORBA COMM_FAILURE 0x4942f306
Maybe; nested exception is:

org.omg.CORBA.COMM_FAILURE: purge_calls:1844 Reason: CONN_ABORT
(1), State: ABORT (5) vmcid: IBM minor code: 306 completed: Maybe

at
com.ibm.CORBA.iiop.UtilDelegateImpl.mapSystemException(UtilDelegateImpl.java:248)

at javax.rmi.CORBA.Util.mapSystemException(Util.java:84)

at
com.milsoft.cdms.co.frmwrkapp._FrameworkBeanInterface_Stub.validateUniqueData(_FrameworkBeanInterface_Stub.java:893)

... 36 more

Caused by: org.omg.CORBA.COMM_FAILURE: purge_calls:1844 Reason:
CONN_ABORT (1), State: ABORT (5) vmcid: IBM minor code: 306
completed: Maybe

at
com.ibm.rmi.iiop.Connection.purge_calls(Connection.java:1843)

at
com.ibm.rmi.iiop.Connection.doReaderWorkOnce(Connection.java:3036)

at
com.ibm.ws.orbimpl.transport.HTTPReaderThread.run(WSHTTPTransportConnection.java:448)

Cannot call EJB method.parameters=[Ljava.lang.Object;@32123212

class=class com.milsoft.cdms.co.frmwrkapp._FrameworkBeanInterface_Stub

globalMessage=appTempExc

Paul Ilechko

unread,
Nov 18, 2006, 3:41:48 PM11/18/06
to
kor...@gmail.com wrote:
> Hi,
>
> Since two of some well-known testing tools can not detect RMI packets
> in the network, we have been forced to use HTTP Tunneling for testing
> thick application clients.


I have no idea what you mean by this. Why is detecting packets at the
network level a prerequisite for any kind of useful testing?

toprak

unread,
Nov 19, 2006, 9:08:15 AM11/19/06
to
Hi,
When performing performance tests using these tools (such as Rational
Robot, or Mercury), they record the packets sent by the client to the
server during recording phase. And later, when running the tests, they
re-send the packets to server several times(iterations) for several
users.

Although it is also possible to use API recording (an option to network
recording), it is not useful for applications having J2EE
client/server architecture.

Serkan Toprak


Paul Ilechko yazdi:

0 new messages