RE: Re: MessageContext->Stub and Stub->MessageContext?

From: Nan Xiong <>
Date: Tue, 28 Oct 2003 15:04:39 -0800

Thanks Bobby for the information. I did a quick try with session_maintain_property on, and using
stash token in the mimeheader in the response handler, but the mime header is null in the request handler,
please do let me know if it works for you.

> -----Original Message-----
> From: Bobby Bissett - Javasoft []
> Sent: Tuesday, October 28, 2003 1:00 PM
> To:
> Subject: Re: MessageContext->Stub and Stub->MessageContext?
> Hi Nan et al,
> I'm away from the office for a couple days, but wanted to
> give an idea
> here. I definitely understand the problem better -- you need
> a way for
> the handler to know which client the request has come from,
> yes? I think
> with the SI there is a 1 to 1 mapping of client to handler chain, but
> that doesn't mean the same mapping applies to handlers
> themselves, and
> other implementations could do it differently...
> The simple fix is to use http header session id across calls
> and set the
> session maintain property on the client. Of course, if you don't have
> access to the service, you can't make that change! Here's an
> idea that
> might help, but I can't test it now and don't know if it will work or
> not. You can give it a shot or I may try it later in the week
> -- anyway,
> I wanted to put it out there for you.
> Set the session maintain property on the stub as if the
> session id were
> coming back in the http header (not soap header). In your client
> handler, you get your token id out of the soap header in
> handleResponse(). From the SOAPMessage object, get the mime
> headers and
> add a session id header. I don't have the javadocs handy for
> the method
> names, but if you check the jwsdp javadocs for SOAPMessage you should
> see the calls. That might work in that the client will now
> remember that
> http "cookie" and set it in the mime headers again when it makes the
> next request. Then your handler can read it back out of the
> MimeHeaders
> object during the request and set it in the SOAPHeader.
> Again, until I
> can test it or look at the code in detail I'm not sure if that will
> work, but I wanted to get the idea to you sooner than later.
> I will also check with the spec lead on the idea to see if it
> has been
> considered.
> Cheers,
> Bobby
> ---------------------------------------------------------------------
> To unsubscribe, e-mail:
> For additional commands, e-mail:

To unsubscribe, e-mail:
For additional commands, e-mail: