RE: [ws-rx] issue 84 proposal for rms colums e-h

[Matthew Lovett <MLOVETT@uk.ibm.com>]

Hi all,

As i084 is back up for debate this week,
I thought that I'd try and restart this email thread. I agree with Doug,
I think that the text in Section 4.7 is clear (line 752 - 754 in CD2).
To repeat:

This fault is sent by an RM Destination
to indicate that the specified sequence has been closed. This fault
MUST be generated when an RM Destination
is asked to receive a message for a sequence that is
closed.

This also couples with the text in lines
364 - 368:

Upon receipt of this message, or
subsequent to the RM Destination closing
the Sequence of its own volition, the RM Destination MUST
include a final SequenceAcknowledgement
(that MUST include the <wsrm:Final> element) header block
on each message destined to the RM Source,
including the CloseSequenceResponse message and on
any Sequence Fault transmitted to the
RMS.


So what happens if the RMS receives
a SequenceClosedFault?

I'd say that the RMS now knows that
the RMD view of the Sequence is that it is closed, and don't think that
there can be any other interpretation. Thanks to the second quotation,
we also know that the RMS has 'final' ack state for the sequence.  The
RMS is now free to do whatever it thinks it should with that information,
and I don't think the spec needs to say any more.

So, I'm sticking to my proposal, e.g.
close with no action. However, I'm happy to hear counter proposals, and
I'd be interested to see concrete proposals for replacement / additional
text to help clarify.

Bob, do you have any ideas you'd like
to circulate before the call? I don't see any reason to object, if you
come up with some clearer text.

Thanks

Matt

Doug Davis <dug@us.ibm.com>

19/01/2006 22:34

To	ws-rx@lists.oasis-open.org
cc
Subject	RE: [ws-rx] issue 84 proposal for rms colums e-h

But your proposal doesn't say what the RMS should do with it - it just
repeats what was said 
before - that the sequence is closed.  :-) 
-Doug 


"Bob Freund-Hitachi" <bob.freund@hitachisoftware.com> wrote
on 01/19/2006 05:01:27 PM:

> No, but the spec should say what an RMS to do if the fault is 
> received at any time. 
> We could say 
> ignore it, 
> Lock up, 
> Close the sequence 
> Right now, it is left to the imagination
>   
> 
> From: Doug Davis [mailto:dug@us.ibm.com] 
> Sent: Thursday, January 19, 2006 1:49 PM
> To: ws-rx@lists.oasis-open.org
> Subject: RE: [ws-rx] issue 84 proposal for rms colums e-h
>   
> 
> hmmm, does the spec say that the fault can be sent w/o some RMS 
> initiated action? 
> -Doug 
> 
> 
> "Bob Freund-Hitachi" <bob.freund@hitachisoftware.com>
wrote on 
> 01/19/2006 04:10:04 PM:
> 
> > One reason is that it is written that the fault is in response
to 
> > the receipt of a message.  If the fault is NOT as a response
to an 
> > RMS initiated message and it is simply received by the RMS, what
> is it to do? 
> > Our readers would prefer that an explicit action (rather than
an 
> > implicit understanding) would clarify the situation. 
> > -bob 
> >   
> > 
> > From: Doug Davis [mailto:dug@us.ibm.com] 
> > Sent: Thursday, January 19, 2006 12:56 PM
> > To: ws-rx@lists.oasis-open.org
> > Subject: Re: [ws-rx] issue 84 proposal for rms colums e-h 
> >   
> > 
> > Bob, 
> >   the fault says: 
> > This fault is sent by an RM Destination to indicate that the
> > specified sequence has been closed. This fault 
> > MUST be generated when an RM Destination is asked to receive
a 
> > message for a sequence that is 
> > closed. 
> > 
> > repeating... "to indicate that the specified sequence has
been 
> > closed".  Why would 
> > adding "This fault when received by the RMS indicates that
the 
> > sequence has been closed by the RMD" 
> > make this more clear?  What assumption could the RMS make
aside from
> > "the sequence has been closed"? 
> > I'm not necessarily against adding your text I just don't see
how it
> > helps to basically 
> > repeat the same thing twice. 
> > -Doug 
> > 
> > 
> > "Bob Freund-Hitachi" <bob.freund@hitachisoftware.com>
wrote on 
> > 01/19/2006 03:41:01 PM:
> > 
> > > I feel that WD07 lines 759-761 describes under what conditions
> > > Sequence Closed is sent by the RMD, not what actions must
be taken 
> > > by RMS if it is received when RMS believes that the sequence
may 
> > notbe closed. 
> > > Proposal: 
> > > Add following text after line 761: (ref WD07) 
> > > This fault when received by the RMS indicates that the sequence
has 
> > > been closed by the RMD.