H.323 Annex E: sequence numbering?

Chip Sharp chsharp at CISCO.COM
Mon Oct 4 13:27:25 EDT 1999


Sorry, it's too late to comment on this.  Annex E was approved in
Santiago.  I assume it will be available for ordering from ITU at some
point.  The final draft may be available on the picturetel avc site.

Chip

P.S.  I noticed that I did not select Reply all when replying to the mail,
but it is sending it to the list anyway.

Chip

At 03:19 PM 9/30/99 -0700, you wrote:
>  Hi All,
>
>  I have a few concerns about H.323 Annex E proposed document.
>  My reference is the "Proposed changes to H.323 Annex E"
>  document which was presented at Santiago meeting of SG13/16.
>
>  Section E.1.3.2 states Serial Model. It says that Annex E
>  stack will wait until a positive reply is returned * for the
>  same Session-Identifier *. This means that an outgoing Annex
>  E PDU can only contain multiple payloads of the same session
>  only. Second, it implies that the sequence number (which is
>  present in Annex E PDU header) must be
>  unique now across < sender host+port, rcv host+port, session-id>.
>
>  We can not make sequence number unique across <sender host+port,
>  rcv host+port> only, because the message sequences should be blocking
>  only for same session-identifier, as stated by the NOTE appended
>  below section E.1.3.2.
>
>  The Annex E Ack message, however, only contains the sequence
>  number of Annex E PDU. How can the sender of an Annex E PDU
>  will distinguish for what session's PDU this ack has come for?
>  Because sender might have sent Annex E PDUs with the same
>  sequence number for different sessions over a given < sender host+port,
>  rcv host+port>.
>
>  Am I missing something here. I will appreciate any clarifications
>  on the above.
>
>  regards,
>  Manoj Paul.

--------------------------------------------------
Chip Sharp                 Consulting Engineering
Cisco Systems              Telco Bio-region
Reality - Love it or Leave it.
--------------------------------------------------



More information about the sg16-avd mailing list