Re: Interface Identifier in ASPTM message in M2UA

My thoughts on this are that there is really no point in trying to put
requests
for multiple interfaces interfaces into a single message.
 
All it does is make the code more complex and might lead to interworking
issues - at best it will be exercising little-used code paths.
 
It might appear to be useful in order to reduce the number of message
sent,
but the number is small compared to what is likely to happen when
the links are in use.
In theory the multiple requests can be transferred in a single ethernet
packet - which will get most of the benefits.
 
And, if any of the Interface identifiers are invalid, any error
reporting
is somewhat easier...
 
    David


________________________________

	From: [email protected] [mailto:[email protected]]
On Behalf Of Sandeep Singh
	Sent: 02 September 2011 13:25
	To: [email protected]; Rajesh Makhija; [email protected];
[email protected]; [email protected]
	Subject: Re: [Sigtran] Interface Identifier in ASPTM message in
M2UA
	
	
	Hi everybody,   
	
	Thanks for your valuable inputs. 
	
	It is very clear from Page 36 of RFC 3331 that the length of
Interface Identifier Set having Tag (0x1) can have Length more than 8. 
	
	According to the following figure, the length should be (n*4 +
4) if there are n Interface Identifiers:
	
	    0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5  6 7 8 9 0 1 2 3 4 5 6 7 8 9
0 1
	   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	   |           Tag (0xb)                    |            Length
= 8               |
	   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	   |                               Traffic Mode Type
|
	   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	   |     Tag (0x1=integer)               |            Length
|
	   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	   /
\
	   \                                   Interface Identifiers*
/
	   /
\
	   +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
	
	Please let me know if there is anything wrong in my
understanding. 
	
	
	Best Regards,
	
	Sandeep Singh

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