[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Subject: RE: [ebxml-msg] Can an MSH send a sync ack and then an async resp onse?
David, The words you quoted make the coupling as loose as can be. I really don't know what your concern is. In addition, the CPPA and MSG teams have been working very hard since the work resumed under OASIS to ensure that the two specifications are consistent on all details. 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 ************************************************************************************* "Burdett, David" <david.burdett@commerceone.com> on 01/20/2002 07:05:07 PM To: Martin W Sachs/Watson/IBM@IBMUS cc: ebxml-msg@lists.oasis-open.org Subject: RE: [ebxml-msg] Can an MSH send a sync ack and then an async resp onse? Marti I was responding to the comment made originally made by Cliff Collins at [1] and confirmed by David Fischer at [2] that changes to the CPP/A spec were implicity changing the behavior of the MSH spec. I have gone on record many times before that this tight coupling between the specs is wrong and should have been avoided but accept that it is too late to do anything about it for this version. However if there is a tight coupling between the two specs we should make it explicit and update the introduction to reflect this in the way that David Fischer suggests at [3]. Finally Marty, please remember that I an NOT anti-CPA I just want the relationship between the CPA and the MSH specs to be "right" with less coupling than currently exists. No doubt there will be debate around this when we plan what goes into the next version. David ================ [1] http://lists.oasis-open.org/archives/ebxml-msg/200201/msg00154.html [2] http://lists.oasis-open.org/archives/ebxml-msg/200201/msg00158.html [3] http://lists.oasis-open.org/archives/ebxml-msg/200201/msg00161.html -----Original Message----- From: Martin W Sachs [mailto:mwsachs@us.ibm.com] Sent: Sunday, January 20, 2002 1:15 PM To: David Fischer Cc: Burdett, David; Arvola Chan; Cliff Collins; Cherian Sanjay; ebxml-msg@lists.oasis-open.org Subject: RE: [ebxml-msg] Can an MSH send a sync ack and then an async resp onse? David, I cannot imagine what changes could be made to the CPP-CPA specification that could possibly affect the words that you are quoting. What do you have in mind? 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 **************************************************************************** ********* David Fischer <david@drummondgroup.com> on 01/20/2002 12:02:40 PM To: "Burdett, David" <david.burdett@commerceone.com>, Arvola Chan <arvola@tibco.com>, Cliff Collins <collinsc@sybase.com>, Cherian Sanjay <Sanjay_Cherian@stercomm.com>, ebxml-msg@lists.oasis-open.org cc: Subject: RE: [ebxml-msg] Can an MSH send a sync ack and then an async resp onse? Yes David, you are probably right, but. . . Point of order, we have a vote in progress so all motions for modification are inappropriate. If the vote succeeds, these proposed changes can be part of the OASIS membership critique. If the vote fails, then we can proceed as appropriate. We need to finish the vote before making any motions for modifications. Regards, David Fischer Drummond Group ebXML-MS Editor. -----Original Message----- From: Burdett, David [mailto:david.burdett@commerceone.com] Sent: Sunday, January 20, 2002 10:12 AM To: 'David Fischer'; Arvola Chan; Cliff Collins; Cherian Sanjay; ebxml-msg@lists.oasis-open.org Subject: RE: [ebxml-msg] Can an MSH send a sync ack and then an async resp onse? Folks If changes to the CPA spec implicity change the behavior of the MSH then do we need to rework the introduction (section 1.2.2, lines 350-356) which says "An ebMS-compliant MSH shall respect the in-force agreements between itself and any other ebMS-compliant MSH with which it communicates. In broad terms, these agreements are expressed as Collaboration Protocol Agreements (CPA). This specification identifies the information that must be agreed. It does not specify the method or form used to create and maintain these agreements. It is assumed, in practice, the actual content of the contracts may be contained in initialization/configuration files, databases, or XML documents complying with the ebXML Collaboration Protocol Profile and Agreement Specification [ebCPP]." David -----Original Message----- From: David Fischer [mailto:david@drummondgroup.com] Sent: Friday, January 18, 2002 8:01 PM To: Arvola Chan; Cliff Collins; Cherian Sanjay; ebxml-msg@lists.oasis-open.org Subject: RE: [ebxml-msg] Can an MSH send a sync ack and then an async response? Arvola, This would fix the problem. Would there then be 4 new enumerated values? None SignalsOnly ResponseOnly SignalsAndResponse mshSignalsOnly mshSignalsAndSignals mshSignalsAndResponse mshSignalsAndSignalsAndResponse Cliff is right that this implicitly changes the behavior of the MSH (at least from a programmer's view) but can we avoid a change in the MS spec? David. -----Original Message----- From: Arvola Chan [mailto:arvola@tibco.com] Sent: Friday, January 18, 2002 7:26 PM To: Cliff Collins; Cherian Sanjay; ebxml-msg@lists.oasis-open.org Subject: Re: [ebxml-msg] Can an MSH send a sync ack and then an async response? Cliff: The syncReplyMode value of "mshSignalsOnly" was agreed by the team during the conference call of 10/22/2001. See http://lists.oasis-open.org/archives/ebxml-msg/200110/msg00174.html. The sentence on lines 1632-1634 may cause some confusion: "If the syncReplyMode is not none, a SyncReply element MUST be present and the MSH must return any response from the application or business process in the payload of the synchronous reply message, as appropriate." If syncReplyMode is mshSignalsOnly, response from the application or business process would not be contained in the synchronous reply message, only the MSH level Acknowledgment would be returned synchronously; business signals and response have to be returned asynchronously. I agree with Ralph that this is a reasonable behaviour that should be configurable. We may need to rephrase the sentence on lines 1632-1634 to allow this to happen. -Arvola ----- Original Message ----- From: "Cliff Collins" <collinsc@sybase.com> To: "Arvola Chan" <arvola@tibco.com>; "Cherian Sanjay" <Sanjay_Cherian@stercomm.com>; <ebxml-msg@lists.oasis-open.org> Sent: Friday, January 18, 2002 1:11 PM Subject: RE: [ebxml-msg] Can an MSH send a sync ack and then an async response? > Doesn't this conflict with the MS 2.0 spec and indirectly change the > functionality of a MSH without the MS team voting on it? > > MS Lines 1632-1643 (section 7.4.7) would be in conflict with the CPA 1.1/2.0 > and would need to change. > > Cliff > > > > -----Original Message----- > > From: Arvola Chan [mailto:arvola@tibco.com] > > Sent: Friday, January 18, 2002 1:03 PM > > To: Cherian Sanjay; ebxml-msg@lists.oasis-open.org; > > ebxml-cppa@lists.oasis-open.org > > Subject: Re: [ebxml-msg] Can an MSH send a sync ack and then an async > > response? > > > > > > Sanjay: > > > > The next version of the CPP/A spec (don't yet know whether it is > > going to be > > called 1.1 or 2.0, but due end of 2/2002) that the ebxml-cppa TC > > is working > > on will have an additional synchronous reply mode defined, > > "mshSignalsOnly". > > In that mode, the receiver is expected to return the Acknowledgment > > synchronously, and then any business level signal or response > > asynchronously. > > > > -Arvola > > > > ----- Original Message ----- > > From: "Cherian, Sanjay" <Sanjay_Cherian@stercomm.com> > > To: <ebxml-msg@lists.oasis-open.org> > > Sent: Friday, January 18, 2002 12:08 PM > > Subject: [ebxml-msg] Can an MSH send a sync ack and then an async > > response? > > > > > > > Hi, > > > > > > The question is 'Can an MSH send a sync ack and then an async response?' > > > > > > Lines 1632-1643 (section 7.4.7) in ebMS v2.0 states: > > > > > > If the syncReplyMode parameter is not none, a SyncReply element MUST > > > be present and the MSH must return any response from the application or > > > business process in the payload of the synchronous reply message, as > > > appropriate. > > > > > > It would seem useful if the MSH could return the ack message > > synchronously, > > > close the HTTP connection and > > > then send the business signal and/or response asynchronously. This is > > > necessary if the business process is > > > long lived. It also seems necessary for reliable messaging to work with > > > SyncReply set. > > > > > > I would greatly appreciate an answer. > > > > > > Thanks a lot in advance. > > > > > > Regards, > > > Sanjay J. Cherian > > > Sterling Commerce, > > > Irving, TX > > > > > > > > > ---------------------------------------------------------------- > > > To subscribe or unsubscribe from this elist use the subscription > > > manager: <http://lists.oasis-open.org/ob/adm.pl> > > > > > > ---------------------------------------------------------------- > > To subscribe or unsubscribe from this elist use the subscription > > manager: <http://lists.oasis-open.org/ob/adm.pl> > > > ---------------------------------------------------------------- To subscribe or unsubscribe from this elist use the subscription manager: <http://lists.oasis-open.org/ob/adm.pl> ---------------------------------------------------------------- To subscribe or unsubscribe from this elist use the subscription manager: <http://lists.oasis-open.org/ob/adm.pl>
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [Elist Home]
Powered by eList eXpress LLC