Dan,
You make some excellent points.
I suspect that MSH builders will be much more concerned about the
complexity of implementing a large repertoire of error messages than about
the cost of sending one occasionally. On a project earlier in my career,
we had defined a large repertoire of error messages. Then the implementers
told us that the were going to process most of them in exactly the same
way. They saw no value in doing fine-grained analysis of the error
conditions. We quickly agreed on a much smaller number of more generic
error messages. This could be the same situation even if there were not
the ambiguity of the receiving the duplicate message that exceeded time to
live.
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
*************************************************************************************
Dan Weinreb <dlw@exceloncorp.com> on 12/12/2001 10:16:35 AM
Please respond to Dan Weinreb <dlw@exceloncorp.com>
To: david@drummondgroup.com
cc: david.burdett@commerceone.com, ebxml-msg@lists.oasis-open.org
Subject: Re: [ebxml-msg] Section 3.1.6.4 TimeToLiveElement
Date: Wed, 12 Dec 2001 08:39:29 -0600
From: David Fischer <david@drummondgroup.com>
+1
Except a receiving MSH does not just ignore a message passed TTL, it
sends back
an error -- TimeToLiveExpired.
As I think I've said before, it's not clear to me that there's much
value in sending back an error message in this case. What's the From
MSH going to do when it gets such an error message? I guess it could
use it to record metering statistics that somebody might look at.
(Note that the From MSH cannot assume, from receiving such an error
message, that the original message did not get through. Maybe the
original message got through very quickly just fine, but also a
duplicate was generated, and the duplicate arrived much later.)
I'm not sure the cost of the error message is worth the benefit.
But presumably this won't happen often, so the performance cost
is low, so it's harmless.
----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.oasis-open.org/ob/adm.pl>