Re: [ws-rx] Issue 23: CreateSequenceRefused fault should apply tosender and receiver

[Duane Nickull <dnickull@adobe.com>]

Title: Re: [ws-rx] Issue 23: CreateSequenceRefused fault should apply to sender and receiver

Correct this is it is errant, but wouldn’t a the CSR be indicative of the status of the CS. Regardless of whether or not the CS is treated as idempotent, the SequenceID must be unique.  If the duplicate CS message arrives with the same SequenceID, the RMD must throw a fault.  If a duplicate message arrives and results in a CSR, there is no logical way the CSR would be duplicate as it would indicate the first CS was not received.

D


On 11/15/06 11:18 AM, "Doug Davis" <dug@us.ibm.com> wrote:

Well, during the interop I noticed that some implementations would resend the same CSR (same SeqID) for each resent CS - and that just seemed odd to me - but I could have been remembering wrong.

thanks
-Doug
__________________________________________________
STSM | Web Services Architect | IBM Software Group
(919) 254-6905 | IBM T/L 444-6906 | dug@us.ibm.com



"Gilbert Pilz" <Gilbert.Pilz@bea.com> 11/15/2006 02:10 PM

To

Doug Davis/Raleigh/IBM@IBMUS, <ws-rx@lists.oasis-open.org>

cc
Subject

RE: [ws-rx] Issue 23: CreateSequenceRefused fault should apply to sender and receiver




It isn't clear to me why you think the spec needs clarification on this point. Your are worried that people may think that "resending" a CreateSequence (for now we'll assume this means that all elements and values, including wsa:MessageID, are identical) is different from sending two unique CreateSequence messages (different MessageIDs, different ??) in what way?

-gp

---

From: Doug Davis [mailto:dug@us.ibm.com]
Sent: Tuesday, November 14, 2006 5:30 PM
To: ws-rx@lists.oasis-open.org
Subject: Re: [ws-rx] Issue 23: CreateSequenceRefused fault should apply to sender and receiver


This reminded me of a possible new PR issue but I wanted to ask about it first... the spec is silent (or at least I can't find it) on the topic of resending the CreateSequence.  There really isn't a mechanism for the RMD to detect a resend.  Seems like a resend of a CS should be treated like a totally new CS - and the spec should probably make this clear.  Thoughts?

As for Marc's comment - which I think it related - I don't think the RMD should accept the resend CS in the scenario you describe since it can't detect that its a resend.

thanks
-Doug
__________________________________________________
STSM | Web Services Architect | IBM Software Group
(919) 254-6905 | IBM T/L 444-6906 | dug@us.ibm.com


Marc Goodner <mgoodner@microsoft.com> 11/14/2006 07:57 PM

To

"ws-rx@lists.oasis-open.org" <ws-rx@lists.oasis-open.org>

cc
Subject

[ws-rx] Issue 23: CreateSequenceRefused fault should apply to sender and receiver






I believe we discussed this on a call and I had an outstanding action to come back to the TC on why I thought this was so.

Currently the spec defines this as a Sender fault, so I won’t talk to that. The reason it should also be a receiver fault is that there may be times when the RMD that receives the CS message may simply be incapable of initiating a new sequence. For example the RMD may have no more capacity to process a request for an additional sequence. Therefore if the same CS message is resent later it may be accepted making Sender the wrong code in that instance.

From: Marc Goodner [mailto:mgoodner@microsoft.com]
Sent: Saturday, October 21, 2006 12:12 PM
To: ws-rx@lists.oasis-open.org
Subject: [ws-rx] New Issue: CreateSequenceRefused fault should apply to sender and receiver

In section 4.6 the CreaetSequenceRefused fault is defined as a sender fault. This fault should apply to sender or receiver.

-- 
******************************************************
Sr. Technical Evangelist - Adobe Systems, Inc.       *
Chair - OASIS SOA Reference Model Technical Committee*
Blog: http://technoracle.blogspot.com                *
******************************************************