Re: T.38 session name in examples

I agree with Albrecht.  MGs must craft compliant SDP.  It is not the
MGC's responsibility to modify non-compliant SDP from an MG into
compliant SDP.
 
MGs that omit required lines are non-compliant to both RFC 2327 and
H.248.1.
 
Kevin

________________________________

From: [email protected] [mailto:[email protected]] On Behalf
Of Schwarz Albrecht
Sent: Tuesday, May 13, 2008 3:36 AM
To: RUIKAR RAHUL A; [email protected]
Subject: Re: [Megaco] T.38 session name in examples


Agree to Kevin, this was also made more explicit in the past in an IG
("In session descriptions sent from the MG to the MGC, the SDP must
comply with RFC 2327. "). See also H.248.1 V3 Amendment 1.
 
I may note that there are in the meanwhile many H.248 profile
specification which defining default values for the SDP o=, s= and t=
lines in MG-to-MGC directions.
 
Thus, the T.38 H.248 examples are incorrect in my understanding.
But implementers should implement the protocol, and not examples.
(T.38 was never officially reviewed by Q.3/16 wrt H.248 examples (as far
as I know)).
 


________________________________

	From: RUIKAR RAHUL A 
	Sent: Dienstag, 13. Mai 2008 05:10
	To: Kevin Boyle; [email protected]
	Cc: Schwarz Albrecht
	Subject: RE: [Megaco] T.38 session name in examples
	
	
	Thanks for explanation.if "s" line is optional only in MGC->MG
direction and not in MG->MGC
	but then I see that those attributes ( o,s,t ) are missing in
reply from gateway. 
	Now I have following doubt regarding this,
	1) Should MG send RFC 2327 conformant SDP to MGC in reply ( eg.
adding missing o,s,t lines )
	2) or is it MGC which has to manage these missing attribute and
add it to SDP which is to be sent out to remote side.
	 
	 
	Example From Page No. 78 and 79
	 
	MGC to MG1:
	====================
	MEGACO/1.0 [123.123.123.4]:55555
	Transaction = 11 {
	Context = $ {
	Add = DS0/1/1 {
	Events = 1 {al/on, ctyp/dtone }
	}, ; SCN termination prepared to listen for tones
	Add = $ {
	Media {
	Stream = 1 {
	LocalControl { Mode = ReceiveOnly, ReserveGroup = True,
ReserveValue =
	True},
	Local {
	v=0
	c=IN IP4 $
	m=audio $ RTP/AVP 18 0
	v=0
	c=IN IP4 $
	m=image $ udptl t38
	}; IP termination for audio } }}}}
	
	Reply from MG1 -> MGC
	=====================
	 
	MEGACO/1.0 [124.124.124.222]:55555
	Reply = 11 {
	Context = 2000 {
	Add = DS0/1/1, ; SCN termination added
	Add = RTP/1 {
	Media {
	Stream = 1 {
	Local {
	v=0
	c=IN IP4 124.124.124.222
	m=audio 2222 RTP/AVP 18 0 ;MG1 supports both offered codecs
	a=ptime:20
	} ; IP termination added }}}}}
	
	 
	Regards,
	Rahul Ruikar
	Alcatel-Lucent
	Email: [email protected]
	Alcanet: 2721 4732
	 

________________________________

	From: Kevin Boyle [mailto:[email protected]] 
	Sent: Tuesday, May 13, 2008 2:38 AM
	To: RUIKAR RAHUL A; [email protected]
	Cc: Schwarz Albrecht
	Subject: RE: [Megaco] T.38 session name in examples
	
	
	This misconception continues to pervade, and it is worrisome.
	 
	The s= line (and other mandatory portions of the SDP
specification) are *NOT* optional in H.248.  They are only optional in
the MGC->MG direction.  In communications from the MG to the MGC, the
SDP must be conformant to RFC 2327.  
	 
	The text in Clause 7.1.8 is explicit that the exceptions to SDP
are valid only in the MGC->MG direction.  This is to allow the MGC to
provide over- or under-specified SDP.
	 
	Kevin

________________________________

	From: [email protected] [mailto:[email protected]]
On Behalf Of RUIKAR RAHUL A
	Sent: Sunday, May 11, 2008 6:07 AM
	To: [email protected]
	Cc: Schwarz Albrecht
	Subject: [Megaco] T.38 session name in examples
	
	
	Just an observation about ITU-T Rec. T.38 (09/2005)
	according to SIP RFC "s" line in SIP messages is mandatory, it
is missing in examples 
	given in document. 
	Though for H248 this attribute is optional but for messages
sent/received from/to SIP agent 
	shall have session name attribute present.
	 
	 
	Regards,
	Rahul Ruikar
	Alcatel-Lucent
	Email: [email protected]
	Alcanet: 2721 4732

_______________________________________________
Megaco mailing list
[email protected]
https://www.ietf.org/mailman/listinfo/megaco