Folks,
I heard a number of comments from people expressing their concerns over this matter. I believe that most people just want to have a resolution, rather than debate the technical merits of one solution or another.
I know that a few people will not like this, but my recommendation is that we correct H.225.0v3. I believe that this would be the fairest solution, given that there are H.323v2 implementations in the field which expect the ASN.1 encoding as detailed in the H.323v2 implementers guide and which is inconsistent with H.323v3. While H.323v3 has been decided for a year now, it seems that products are just now emerging, so the impact is less significant.
Bear in mind that my recommendation is merely that: a recommendation. I will present my recommendation at the Rapportuer's meeting and/or the ITU-T SG16 meeting in November. The final decision on this matter will not be official until that time.
However, I would rather have the support of the H.323 community now, rather than a month from now, so that those interested parties trying to bring their H.323v3 products to market can rest assured that their ASN.1 syntax is final.
I have already updated my copy of the Implementers Guide and the ASN.1 file for H.225.0v3 on Packetizer to reflect my proposed change (see http://www.packetizer.com/iptel/h323/h2250v3.asn).
Again, this change is intended to be "fair", giving consideration to equipment that is older and already in the field, which is almost always more difficult to change than recently released or emerging products. In addition, the error here truly is in H.225.0v3, not H.225.0v2, so it is more logical to make the correction in H.225.0v3.
Best Regards, Paul
----- Original Message ----- From: "Paul E. Jones" paulej@packetizer.com To: "XuPeili" xupeili@huawei.com Cc: h323implementors@imtc.org; ITU-SG16@mailbag.cps.intel.com; h323implementors@pulver.com Sent: Wednesday, October 04, 2000 1:35 AM Subject: Error in H.323v3 ASN.1
XuPeili and other H.323 Developers,
This does appear to be in error. Unfortunately, this was published this
way
in the H.225.0v3 Recommendation. However, you are correct that a previous H.323v2 Implementers Guide was published which contained the aliasesInconsistent field.
So, we have an issue to contend with. I must ask the developer
community--
I do not want to make this change without wide support for making such a change.
Since the Version 2 field was added via the Implementers Guide, it is entirely possible that H.323v2 vendors did not even include it in their ASN.1
So, I need to hear from all of the developers on this issue. This may
have
been addressed, but I don't have notes on this matter. A similar issue
with
the LocationRejectReason was found and I corrected that back in June when discussing with everybody that "I would never change the ASN.1 and more". Well, unfortunately, it appears that we have one last error-- honestly,
I'm
quite shocked this one slipped through.
So... should the "aliasesInconsistent" field be moved above the "routeCallToSCN" element in the AdmissionRejectReason sequence as shown in an earlier H.323v2 Implementers Guide?
Please post and debate this publicly. We need to resolve this matter quickly so as to minimize impact on everybody.
Best Regards, Paul
----- Original Message ----- From: "XuPeili" xupeili@huawei.com To: "Paul E. Jones" paulej@packetizer.com Cc: h323implementors@imtc.org Sent: Tuesday, October 03, 2000 9:42 PM Subject: ;X84: Multiple Call Proceedings in a H.323 call
hi Paul,
In the H225.0v3.asn downloaded form www.packetizer.com the ARJ is specified like this
AdmissionRejectReason ::= CHOICE { calledPartyNotRegistered NULL, -- cannot translate address invalidPermission NULL, -- permission has expired requestDenied NULL, -- no bandwidth available undefinedReason NULL, callerNotRegistered NULL, routeCallToGatekeeper NULL, invalidEndpointIdentifier NULL, resourceUnavailable NULL, ..., securityDenial NULL, qosControlNotSupported NULL, incompleteAddress NULL, routeCallToSCN SEQUENCE OF PartyNumber, aliasesInconsistent NULL -- multiple aliases in request identify
distinct
people }
Since the routeCallToSCN is a new choice in H.225.0v3, I think it should
be
placed after the aliasesInconsistent choice which is already exist in v2.
-
Please send E-mail to contact@imtc.org mailto:contact@imtc.org to subscribe or unsubscribe from this list
-
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ For help on this mail list, send "HELP ITU-SG16" in a message to listserv@mailbag.intel.com