RE: [ws-rx] PR022: strawman proposal

["Durand, Jacques R." <JDurand@us.fujitsu.com>]

For the sake of catching inconsistencies, the spec should require an Error be generated in case a CloseSequence message (or a TerminateSequence) is received that has no LastMsgNum while the CSR had specified IncompletSeqBeh==DiscardAll.

I think that just discarding the entire sequence in that case is rather harsh - need to give the RMS a chance to get it right !

 

So yes, inclusion of LastMsgNum is technically optional, but there are combinations that do not make sense regardless of the out-of-band agreement, and should be flagged.

 

 

-Jacques

---

From: Gilbert Pilz [mailto:gpilz@bea.com]
Sent: Thursday, December 14, 2006 3:49 PM
To: Doug Davis; Bob Freund-Hitachi
Cc: Hal Lockhart; ws-rx@lists.oasis-open.org
Subject: RE: [ws-rx] PR022: strawman proposal

I'm ok with this. For the use case I am concerned with the RMS and RMD will have to have some out of band (and unspecified) agreement in which the RMS promises to:

 

1.) send either CloseSequence or TerminateSequence (or both)

 

2.) agrees to include LastMsgNum in any CSs or TSs it sends

 

3.) agrees to retry CS and/or TS until such time as it becomes clear that the RMD has received the message (CSR, TSR, UnknownSequence fault, etc.) or it becomes clear that all hope is lost

 

This is all optional and none of it will appear in the spec, but I just wanted to see if people agree that it was a valid use of the WS-RM protocol.

 

- gp

---

From: Doug Davis [mailto:dug@us.ibm.com]
Sent: Thursday, December 14, 2006 3:29 PM
To: Bob Freund-Hitachi
Cc: Gilbert Pilz; Hal Lockhart; ws-rx@lists.oasis-open.org
Subject: RE: [ws-rx] PR022: strawman proposal

I'm not thrilled with making Close a required operation since I think it really should remain necessary only when things go wrong.  So how about:

LastMsgNum SHOULD be on Close and Terminate - but it is optional.
If not there then the RMD MUST assume there are gaps at the end and for IncompletSeqBeh==DiscardAll, it would discard the entire sequence.  For other values it has no effect.

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

"Bob Freund-Hitachi" <bob.freund@hitachisoftware.com>

12/14/2006 09:43 AM

To	Doug Davis/Raleigh/IBM@IBMUS, "Gilbert Pilz" <gpilz@bea.com>
cc	"Hal Lockhart" <hlockhar@bea.com>, <ws-rx@lists.oasis-open.org>
Subject	RE: [ws-rx] PR022: strawman proposal

I would seriously prefer mandating close with re-tries as required or indication of a final (empty) sequence message.
-bob
 

---

From: Doug Davis [mailto:dug@us.ibm.com]
Sent: Thursday, December 14, 2006 7:17 AM
To: Gilbert Pilz
Cc: Hal Lockhart; ws-rx@lists.oasis-open.org
Subject: Re: [ws-rx] PR022: strawman proposal
 

This would imply that if a TerminateSequence is sent (w/o a Close) and there is a gap at the end of the Sequence that the RMD will think the Sequence is complete and when IncompleteSeqBehavior is set to 'discardEntireSequence' will not discard anything.  This could leave the RMS and RMD in different states - is this ok to the originators of the IncompleteSeqBehavior feature?

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

"Gilbert Pilz" <gpilz@bea.com>

12/14/2006 03:49 AM

To	<ws-rx@lists.oasis-open.org>
cc	"Hal Lockhart" <hlockhar@bea.com>
Subject	[ws-rx] PR022: strawman proposal

 

From previous discussions on this issue it is evident that the TC doesn't think that the motivating requirement is pressing enough to justify any large changes to the spec (i.e. making TerminateSequence or CloseSequence required). With that in mind I propose that we resolve this issue by doing the following:

1.) Adding the mandatory LastMsgNumber element to CloseSequence. As previously discussed, the description of LastMsgNumber should be something along the lines of:

The LastMsgNumber element specifies the highest assigned message number of all the Sequence Traffic Messages for the Sequence being closed. The RM Destination can use this information, for example, to implement the behavior indicated by wsrm:CreateSequenceResponse/wsrm:IncompleteSequenceBehavior.

2.) The use of CloseSequence remains optional. Any agreement between the RMS and the RMD about the use of CloseSequence to allow the RMD to determine if it has succesfully received all the messages sent by the RMS is out of scope.

3.) An RMS that sends a CloseSequence but does not receive a CloseSequenceResponse is free to retry the CloseSequence message or not depending upon local policy etc.

- gp