Re: Idempotency attribute (Section 7.6.4.2, line 1569)

[Martin W Sachs <mwsachs@us.ibm.com>]

Given that we already have the idempotency attribute on ReliableMessaging,
Chris' suggestion is the best we can do.  Another way of implementing the
consistency constraint is to build it into CPA composition tools.

Really, it's gotten a bit confusing this way.  The child elements of
ReliableMessaging only apply when deliverySemantics="OnceAndOnlyOnce" since
their counterparts in the message header relate only to ReliableMessaging.
A consistency check really should check everything under ReliableMessaging.
In fact there is an error in the description of the child elements in the
CPP-CPA spec.  They MUST be all present if deliverySemnatics
= "OnceAndOnlyOnce" and MUST all be absent (or ignored?) if
deliverySemantics = "BestEffort".

A better idea, if we could go that far without a migration problem ,is to
change the cardinality of ReliableMessaging, to 0 or 1 (omitting the
deliverySemantics attribute) and express idempotency and anything that
applies to both values of deliverySemantics outside the ReliableMessaging
element.  Some consistency checks might still be needed but it would be
easier to understand.

Regards,
Marty

*************************************************************************************

Martin W. Sachs
IBM T. J. Watson Research Center
P. O. B. 704
Yorktown Hts, NY 10598
914-784-7287;  IBM tie line 863-7287
Notes address:  Martin W Sachs/Watson/IBM
Internet address:  mwsachs @ us.ibm.com
*************************************************************************************



christopher ferris <chris.ferris@east.sun.com>@Sun.COM on 07/31/2001
10:24:07 AM

Sent by:  Chris.Ferris@Sun.COM


To:   Martin W Sachs/Watson/IBM@IBMUS
cc:   Arvola Chan <arvola@tibco.com>, ebxml-cppa@lists.oasis-open.org
Subject:  Re: Idempotency attribute (Section 7.6.4.2, line 1569)



Arvola,

As Marty has already responded, the document exchange is an abstract
notion of the CPP/A that relates to information that is useful to the MSH.
It doesn't define the MSH, nor is it intended to do so.

As to idempotency, even in the absence of reliable messaging, providing
for some manner of idempotency check is useful. As to whether it is
sensible
for one to define:

> What is the meaning of  deliverySemantics=OnceAndOnlyOnce and
> idempotency=false?
>

it would be meaningless as you cannot achieve onceAndOnlyOnce without
idempotency. In the CPP/A spec, the example at line 1538 uses this
combination, and IMHO that should be fixed in the errata to be:

<ReliableMessaging deliverySemantics="OnceAndOnlyOnce"
     idempotency="true"
     messageOrderSemantics="Guaranteed">
...
</ReliableMessaging>

Also, a constraint should be added to the CPP/A spec in section 7.6.4.2
that precludes the combination you cited. However, this constraint cannot
be
modelled in the XML Schema definition itself and would need to be
externally
enforced through some mechanism other than schema validation. It is likely
that
RELAX-NG or possibly schematron could be used to express this manner of
constraint.

 Martin W Sachs wrote:
>
> Arvola,
>
> To answer the simple question first:  The document exchange layer of the
> CPP-CPA contains the variables related to the Message Service.
>
> As to idempotency, I hope that Chris Ferris will reply.  I believe that
you
> have turned up another loose end from the TP-TRP interlocks during the
> first life of ebXML.  It certainly seems that idempotency checking cannot
> be optional with onceAndOnlyOnce semantics.
>
> Regards,
> Marty
>
>
*************************************************************************************

>
> Martin W. Sachs
> IBM T. J. Watson Research Center
> P. O. B. 704
> Yorktown Hts, NY 10598
> 914-784-7287;  IBM tie line 863-7287
> Notes address:  Martin W Sachs/Watson/IBM
> Internet address:  mwsachs @ us.ibm.com
>
*************************************************************************************

>
> Arvola Chan <arvola@tibco.com> on 07/26/2001 05:27:49 PM
>
> To:   ebxml-cppa@lists.oasis-open.org
> cc:
> Subject:  Idempotency attribute (Section 7.6.4.2, line 1569)
>
> This concept is not mentioned any where in the  Messaging Service spec.
In
> order to implement the OnceAndOnlyOnce delivery  semantics, the MSH
already
> performs duplicate detection and filtering. It is not  clear how the
> idempotency test applied at the document exchange layer (described  on
line
> 1575) differs from the duplicate detection and handling described in
> Section 10.3 in the Messaging Service spec.
>
> Is the document exchange layer part of the  MSH?
>
> What is the meaning of  deliverySemantics=OnceAndOnlyOnce and
> idempotency=false?
>
> -Arvola
>
> Arvola Chan (arvola@tibco.com)
> TIBCO Software  (on loan to RosettaNet)
> +1-650-846-5046 (US-Pacific)
>
> ------------------------------------------------------------------
> To unsubscribe from this elist send a message with the single word
> "unsubscribe" in the body to: ebxml-cppa-request@lists.oasis-open.org