[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]
Subject: Re: [ws-rx] PR022: updated proposal
What about saying it MUST appear in CS and TS, and the number in TS MUST be the same as was in CS? Paul Gilbert Pilz wrote: > Logically if LastMsgNumber appears only in TS then sending a TS is a > MUST. This isn't to say that an RMD that, for whatever reason, doesn't > care about LastMsgNumber can't clean up a sequence after CS. > > - gp > > ------------------------------------------------------------------------ > *From:* Doug Davis [mailto:dug@us.ibm.com] > *Sent:* Monday, November 06, 2006 2:57 PM > *To:* Gilbert Pilz > *Cc:* ws-rx@lists.oasis-open.org > *Subject:* Re: [ws-rx] PR022: updated proposal > > > Gil, > I understand your concerns with adding it to the Close but I > still think it would be nice if we could find a way to do it - > forcing a sequence to wait for a Terminate after a Close before it > can complete it works (in some configurations). Also, I may have > missed the "MUST" but I don't think the Terminate is required to > be sent - the RMS may choose to just let the Sequence timeout. > Wouldn't this cause an interop issue - because the RMD may not get > this new info that we've now added to the TS message and also > because the IncompletSeqBahvior can't do its job (when set to > DiscardEntireSeq) w/o it? > -Doug > > > > *"Gilbert Pilz" <Gilbert.Pilz@bea.com>* > > 11/03/2006 12:59 PM > > > To > <ws-rx@lists.oasis-open.org> > cc > > Subject > [ws-rx] PR022: updated proposal > > > > > > > > > > The attached update incorporates the Chris and PaulF's feedback. > > A number of people have said that they would like to see LastMsgNumber > in CloseSequence as well. While I'm somewhat sympathetic to this idea, > from my point of view it adds a lot of complication to the spec for > little benefit. We have to say that LastMsgNumber must appear in > either > CS or TS and possibly both. If it appears in both CS and TS then the > values have to agree (we might need to define a new fault to cover the > case where they don't agree). We'd have to say that if you include > LastMsgNumber in CS then you didn't necessarily have to send the > TS, but > that if you didn't include LastMsgNumber in CS (or didn't send a CS) > then you *do* have to send a TS with LastMsgNumber, etc. > > - gp > > <<wsrm-1.1-spec-pr-i022.pdf>> > [attachment "wsrm-1.1-spec-pr-i022.pdf" deleted by Doug > Davis/Raleigh/IBM] > -- Paul Fremantle VP/Technology and Partnerships, WSO2 OASIS WS-RX TC Co-chair http://bloglines.com/blog/paulfremantle paul@wso2.com (646) 290 8050 "Oxygenating the Web Service Platform", www.wso2.com
[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]