You can find it in section D.2 of the Megaco/H.248 specification.
-----Original Message----- From: Liu, Gary [mailto:gLiu@unispherenetworks.com] Sent: Monday, March 05, 2001 8:21 AM To: Taylor, Tom-PT [NORSE:B881:EXCH]; ITU-SG16@MAILBAG.INTEL.COM Subject: RE: New comer questions
Is RFC1006 recommended for MEGACO implementation or vendor specific implementation?
-----Original Message----- From: Tom-PT Taylor [mailto:taylor@nortelnetworks.com] Sent: Sunday, March 04, 2001 3:28 PM To: Liu, Gary; ITU-SG16@MAILBAG.INTEL.COM Subject: RE: New commer questions
We use the RFC 1006 length specifier to indicate total message length in TCP.
The "ALF" is actually anything but message delimitation -- not sure how it got that name. The functions performed by the application are
-- reliable delivery (aided by Megaco's transaction orientation) -- sequencing (not much needed in Megaco except in sorting out event reports) -- throttling in the face of network congestion.
-----Original Message----- From: Liu, Gary [ mailto:gLiu@UNISPHERENETWORKS.COM
mailto:gLiu@UNISPHERENETWORKS.COM ]
Sent: Friday, March 02, 2001 11:16 AM To: ITU-SG16@MAILBAG.INTEL.COM Subject: New commer questions
Hi,
I am a new commer to the MEGACO mailing list and have some dump questions. The text enconding is new to me. If TCP is used as
transport, how is a
transaction message framed out from the data stream. Is "MEGACO/1" the frame mark? The standard requies MGC/MC to support UDP/ALF. UDP is
already message
based. Why is ALF (application level framing) needed? I though the other way around. Framing is needed for TCP. Thanks.
Gary Liu
Unisphere Networks, Inc. < http://www.unispherenetworks.com
http://www.unispherenetworks.com >
7800 Congress Ave., Suit 100 Boca Raton, FL 33487 T:(561)981-7022, F:(561)981-7001
mailto:gliu@unispherenetworks.com mailto:gliu@unispherenetworks.com
For help on this mail list, send "HELP ITU-SG16" in a message to listserv@mailbag.intel.com