Re: [ebxml-msg] Re: Comments on the 1.09 about ConversationId

[Martin W Sachs <mwsachs@us.ibm.com>]

But if you want to make messages saved for ordering purposes, and the
conversationId itself, subject to persistDuration, you need to make sure
that persistDuration doesn't kill a conversation which legitimately has a
long wait.  See Chris' and my prior postings on this.  It appears that
persistDuration is a minimum, so nothing is changed by holding onto certain
things  until the conversation ends.

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 12/04/2001 06:10:08 PM

Please respond to "Dan Weinreb" <dlw@exceloncorp.com>

To:    david@drummondgroup.com
cc:    Martin W Sachs/Watson/IBM@IBMUS, ebxml-msg@lists.oasis-open.org
Subject:    Re: [ebxml-msg] Re: Comments on the 1.09 about ConversationId



   Date: Tue, 04 Dec 2001 13:39:43 -0600
   From: David Fischer <david@drummondgroup.com>

   Yes Marty, absolutely.  I think there are potential problems (discussed
   below)
   which make MessageOrder quite fragile.

But being "fragile" is the inherent flip side of "providing stronger
guarantees".  That is, if you ask for a higher degree of service,
naturally there are more things that can disrupt that service.  For
example, if you ask for authentication, then there's a whole new kind
of error that could not happen before, namely "the authentication
check failed".  If you ask for encryption, then you are subject to
failures like having the wrong key, having an expired certificate,
etc.  So if you ask for message ordering, then there are also new
errors if the messages come out of order and if some don't come at
all.

        I am shuddering over the thought of
   changing the persistDuration rules depending upon whether or not
   MessageOrder is
   present (wouldn't this constitute a CPA override!).

I agree that it would not be good to change the persistDuration rules
depending on whether or not MessageOrder is present.