Title: RE: [ebxml-msg] reliable messaging
A more detailed definition of Reliability might answer this,
so that expectations remain reasonable...
The spec may have to distinguish two aspects, somewhat orthogonal:
- the quality of the delivery, e.g "at-least-once" and "at-most-once", which
we know cannot be always enforced, for the same reasons Chris mentioned.
The number of retries states the limits of the effort. Probably the scope of
the duplicate check should also be specified somehow - e.g. in CPA .
(e.g. while duplicate elimination may not be guaranteed over a long time,
it may be critical for a business to know that an MSH guarantees the elimination
of duplicates over a 1 week period.)
- the quality of the sender-receiver synchronization, where several [if not all]
out-of-sync scenarios can be addressed.
The message status service gives primarily a way to resynchronize
at application level. This service could be more explicitly integrated
in a reliability policy, at MSH level, where status values such as "received",
"processed" make more sense.
(For example, a refined policy could be that an MSH not receiving any Ack after
all retries timeout, will send a status request... and notify a delivery failure to
the app only if the status response is "NotRecognized", and will not if
status is "Received" or "Processed". )
Regards,
Jacques
-----Original Message-----
From: Martin W Sachs [mailto:]
Sent: Monday, October 14, 2002 10:32 AM
To: Christopher B Ferris
Cc:
Subject: Re: [ebxml-msg] reliable messaging
Chris,
Yes, as I told Komal, the message status services seems to be just what is
needed. However, its presence is not enough. Reliable Messaging has its
own interoperability matters and prescribing the use of the message status
service along with prescribing minimal state to be persisted is still
necessary.
Of course, it's impossible to close the loop completely. However, people
who view that the important reason for reliable messaging is to cover
(most) cases of system failure and recovery view absence of system failure
coverage as a major omission.
Identifying and stating the known limitations is also part of a
specification writer's responsibilities.
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
*************************************************************************************
Christopher B
Ferris To: Martin W Sachs/Watson/IBM@IBMUS
cc:
10/14/2002 12:27 Subject: Re: [ebxml-msg] reliable messaging(Document link: Martin W. Sachs)
PM
Marty,
ebMS *does* indeed provide such a status query. Granted that its required
use in the
failure mode you articulate is not specified (it could easily be). I do not
believe that
the protocol is necessarily broken in this regard, however it could
certainly be reinforced
and made more clear.
I should also point out that no matter how hard one tries, it is impossible
to close the
loop entirely. If B never recovers, then A and B are permanently and
unreconcilably
out of synch w/r/t their shared understanding of the state of the exchange.
Further comments below.
Cheers,
Christopher Ferris
Architect, Emerging e-business Industry Architecture
email:
phone: +1 508 234 3624
Martin W Sachs/Watson/IBM@IBMUS wrote on 10/14/2002 11:46:01 AM:
>
>
>
>
> It has been pointed out to me that ebXML reliable messaging is not
reliable
> under system failure. At least one person who mentioned it considers
ebXML
> messaging to be broken as a result. Here is a scenario:
>
> Party A send a message reliably to Party B.
>
> Party B's MSH receives and persists the message.
>
> Party B's MSH attempts to send the reliable-messaging acknowledgment but
> Party B's system goes down before the acknowledgment gets on the wire.
>
> Party A exhausts its retries and concludes that the message was not
> delivered.
>
> Party B eventually comes up and the destination application processes the
> persisted message as prescribed in the MSG specification.
>
> Parties A and B are now out of sync with respect to that transaction and
do
> not know they are out of sync. Party A believes that the transaction
> failed. Party B has in fact processed the message that it received from
> Party A. Reliable messaging has failed to deliver on its promise.
>
> The solution to this problem is not trivial and the MSG team needs to
give
> it a lot of thought. At a minimum, the following are needed in the spec:
>
> 1. Both parties to the message exchange MUST persist enough state to
allow
> recovery and getting back in sync. Specific state variables must be
This is already prescribed in the spec.
> prescribed. They are at least those variables needed to restore the
state
> of the transaction and conversation after system recovery, such as the
> conversation ID, CPA Id, service, action, and perhaps other parts of the
> message header.
>
> 2. Timeouts and retries, as prescribed in the MSG spec, are not
sufficient
> to cover system failures since the failure could last a very long time.
> Instead, if the party that sent the message doesn't receive a reply in a
> reasonable time, it must be able to send a status query to the other
party
> and keep requesting status periodically until it receives a response.
The
> status query protocol must be defined in the MSG specification. If the
The protocol is defined, see section 7.
> appropriate state information is persisted at both ends, when party B
comes
> up, it will receive and respond properly to the status query. The
timeouts
> could be retained in the spec but their main use would be to signal the
> "attached human" to make a phone call.
That is always an option:)
>
> The MSG team should consider this a work item for version 3. Should the
> team not wish to solve this problem, at the very least, a caveat should
be
> added to the MSG specification that messaging reliability under
conditions
> of system failure is outside the scope of the MSG team.
Again, I believe that much of your concerns are already addressed. There is
no
doubt in my mind that they could be reinforced, making it abundantly clear
to the reader.
>
> 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
>
*************************************************************************************
>
>
> ----------------------------------------------------------------
> To subscribe or unsubscribe from this elist use the subscription
> manager: <http://lists.oasis-open.org/ob/adm.pl>
----------------------------------------------------------------
To subscribe or unsubscribe from this elist use the subscription
manager: <http://lists.oasis-open.org/ob/adm.pl>
