I am just guessing here, since I don't use weblogic. It may be possible that
web logic forces only RMI-IIOP arguements. In which case any thing passed on
the wire must be a primitive, Serializable or Remote type. If they are not,
they are invalid ant may fail on deployment when weblogic verifies your beans
and such. I don't believe that orion forces RMI-IIOP. See javasoft for more
info about RMI-IIOP.
M
Ismael Blesa Part wrote:
> What does (Is your agruement RMI-IIOP valid) mean?, this class is used from a
> javabean used from a JSP. Its parameters are two classes from my own and
> java.lang.String from Sun. We are not using neither clustering neither
> persistant sessions (at least I think that this is the default
> configuration)
>
> Something that shock me is that it runs fine under Orion, then what could be
> happening?
>
> Matt Bauer wrote:
>
> > Is your agruement RMI-IIOP valid?
> >
> > M
> >
> > Ismael Blesa Part wrote:
> >
> > > We have developed a web application that works fine under OrionServer.
> > > It has JSP, JavaBeans and Taglibs.
> > > Now we are testing with Bea Weblogic and there are some strange errors
> > > thrown by the server.
> > >
> > > It complains about (java.lang.VerifyError: (class:
> > > com/test/logic/integration/connectors/BeanConnector, method: connect
> > > signature:
> > > com/test/logic/MyData;Lcom/test/logic/Environment;Ljava/lang/String;)V)
> > > Incompatible object argument for function call)
> > >
> > > It is a very strange error, I have compiled all my code with javac
> > > (1.3.0_01) and also with jikes last version. I have changed also bea
> > > weblogic to point jdk to my installed jdk.
> > >
> > > The best of all this strange error is that this method is not called.
> > > ie: looking on the print messages I have put on the code, execution
> > > stops before the invocation of this method. I think that this error
> > > comes when the java virtual machine tries to load this class.
> > >
> > > Anybody has any idea about what is causing this error to appear?
> > >
> > > Regards,
> > > Ismael ([EMAIL PROTECTED])
