Re: FW: T2, Proposed solution for ... Re: SyncReplyandReliableMessagingMethod in QualityOfServiceInfo

[christopher ferris <chris.ferris@Sun.COM>]

Dan Weinreb wrote:
> 
>    Date: Thu, 20 Sep 2001 16:45:07 -0500
>    From: David Fischer <david@drummondgroup.com>
> 
>    From: Chris.Ferris@Sun.COM [mailto:Chris.Ferris@Sun.COM]
>    Sent: Thursday, September 20, 2001 11:50 AM
> 
>    How does this work if the sending MSH knows nothing of intermediaries?
>    Are you suggesting that the Via element is now REQUIRED on all messages?
> 
>    <df>Yes, this is a problem, not only for retryCount but for all parameters in
>    Via.  I have asked this same question before, is Via always required?  If we are
>    doing RM at all, then Via is required in order to pass the ackRequested and
>    reliableMessagingMethod parameters.  I guess Via is REQUIRED for all RM
>    messages -- regardless of retryCount.
>    </df>

Here, I disagree. The parties to a single-hop exchange can via the
CPA "know" whether the message is being sent via ebXML RM and hence there
is no need whatsoever for ackRequested. Again, I want to make it clear
that the "CPA" is a "virtual CPA" in the sense that there is an understanding
between the parties that is shared whether by means of a physical CPA
or otherwise. I think you know my preference;-)

> 
> Right, the question is whether it's an RM transmission or not.
> 
> Every hop in a communication is either using ebXML RM or not: ebXML RM
> is being used if (a) this is supposed to be a reliable communication
> and (b) reliableMessageMethod says that the reliability is done with
> ebXML RM rather than by the transport layer.
> 
> Even if there is no IM, there is still one hop, and the hop is either
> using ebXML RM or not.
> 
>    How does an intermediary distinguish itself as such?
> 
>    <df>The IM must know that it is an IM in order to forward rather than process
>    the payload.  If the IM is not the From Party or the To Party then it must be an
>    IM.  This is not quite true for a mailroom situation (need to work on this --
>    something like MTA routing tables?  or MX records?).
>    </df>

Again, I disagree here. I think of an MSH as being unaware of the fact that
it is anything but a MSH. The IM-ness of a node is captured in application
logic as we have discussed, not as an aspect of the MSH itself. IMO,
the fact that intermediary functionality is being applied by the "routing app"
is no different than if the "application" were a responding application
of a true endpoint.

I suppose that the MSH node could detect that the ToParty is not itself
and apply a different set of logic, but again, this seems to be a bit more
complexity than might be needed otherwise.

	   RoutingApp
	   ^	    |
	   |	    V
	MSH-in   MSH-out

> 
> And whenever a hop happens, the MSH's at each end of the hop have to
> "know" somehow what the reliableMessageMethod parameter is, so that
> they know whether to talk ebXML RM or not.

Agreed.

> 
>    retryInterval applies to the RM retry. You are now stating that the DR is the RM
>    artifact, not the Acknowledgment. What happens when no DR is requested/required?
>    Does the sending MSH wait for the response message?
> 
>    <df>Chris, you are right, almost.  There is a CPA to the first IM and a CPA with
>    the end.
> 
> If we're going to be using CPA's this way, I think the CPA is going to
> need some work.

Agreed. This is a known issue.

> 
> The overall idea of the CPA is that you examine what business process
> you're doing, what role you're in, and what service and action you're
> performing; you submit this to the CPA, and it gives you back a
> DeliveryChannel and a Packaging.  DeliveryChannel, in turn, give you
> a Transport and a DocExchange, plus some parameter values.
> 
> The information in Transport is only relevant to hop-to-hop
> communications; it's meaningless to ask for, say, the TCP address of
> the To Party when there are IM's, since the From Party doesn't talk
> directly to the To Party.
> 
> The information in Packaging is all about the payload, and so it's
> only relevant to end-to-end communications; IM's don't pay attention
> to the contents.
> 
> How about the parameters in DeliveryChannel, and DocExchange?  It
> looks to me like they are about the communication as a whole, i.e. all
> the hops, but this perhaps should be clarified.

Supposed to be.

> 
> Should the IM's really be looking at the Service and Action fields of
> the message, and looking through the Override elements of a CPA, in
> order to decide what parameters to use?