RE: [PWE3] Proposed change to 5..4.1 of PWE3 architecture

["Yaakov Stein" <yaakov_s at rad.com>]

> 1. Never suggested to remove support for IP/UDP 
> encapsulation. Where did this come from?

This is what Stewart suggested if the changes were too radical.

 
> 2. It seems more consistent to have the order of RTP and 
> control word be dependent on the demultiplexing layer (MPLS 
> vs. L2TP or UDP) and not PSN layer (IP or MPLS), as 
> potentially the PSN layer tunneling technique may change 
> along the path without a need to change the upper payload 
> convergence layer. As a minimal change I therefore suggest to 
> change the 'can' to a 'must' in the following text:
> 
>     As a special case, if the PW Demultiplexer is an MPLS label, the
>     protocol architecture of section 5.4.2 can be used instead of the
>     protocol architecture of this section.

The reason the differentiation was based on IP was simple.
In the IP world the RTP header has a well-known location,
and it is not for use to change this, although it is inconvenient
for the MPLS case. Hence we based the decision on IP vs MPLS.

However, I agree that your comment is valid and it is more consistent to

base the positioning on the multiplexing layer.  I am just wondering
if anyone else has a problem with this (AVT or L2TPEXT WGs ...)

Y(J)S

_______________________________________________
pwe3 mailing list
pwe3@ietf.org
https://www1.ietf.org/mailman/listinfo/pwe3