Re: [AVT] RE: I-D ACTION:draft-ash-avt-hc-over-mpls-protocol-00.txt

[Colin Perkins <csp at csperkins.org>]

Jerry,

On 21 Apr 2005, at 23:01, Ash, Gerald R ((Jerry)), ALABS wrote:
> Please review and comment on the (significantly) updated draft  
> "Protocol
> Extensions for Header Compression over MPLS"
> (http://www.ietf.org/internet-drafts/draft-ash-avt-hc-over-mpls- 
> protocol
> -00.txt).

The new approach (using MPLS PW) looks to be a big improvement, and  
the draft seems to be taking shape. I do have a number of comments,  
however:

The definition of terms and reference to RFC 2119 needs to be moved  
within the body of the text, rather than being a separate unnumbered  
section. I'd suggest placing the paragraph at the start of Section 2.

Section 1, 2nd paragraph: "the compressed packet routed on an MPLS  
LSP using MPLS labels" --> "the compressed packet _is_then_ routed..."

Section 2, 1st paragraph: I found the phrasing about pseudowires  
transporting the "HC context and other control messages" a little  
confusing, implying that the context and control messages are  
transported separately to the compressed data packets. It might be  
clearer to say "MPLS pseudowires (PWs) are used to transport the  
header compressed packets between the ingress and egress...".

Section 2, 3rd paragraph: The sentence "Compressed packets and HC  
control packets are routed on a separate MPLS LSP/PW..." could be  
interpreted as a requirement to setup two PWs, one for the HC  
compressed data packets and one for the HC control packets. It might  
be clearer to state "Compressed data is routed on a separated MPLS  
LSP/PW from compressor to decompressor"?

Section 2.1, top of page 7: I agree that there is no problem with  
duplicate CIDs since they're received on separate sessions. However,  
the text can be read to imply that something other than normal CID  
processing is required. This confusion comes from the statement  
following ("...the HD can distinguish these this it knows the sour by  
means of the PW label"), which implies that there is a single HD for  
all PWs. It would be clearer if the draft made it explicit that each  
PW had a logically separate HD instance, independent of all other PWs.

Section 2.1, last paragraph: in order to assign a new PW type, the  
IANA must be given explicit guidance on what to assign. This requires  
a statement much like those in section 2.1 of the pwe3-iana- 
allocation draft, listing a numeric type to assign along with a  
description for that type.

Section 2.2: how are the reserved values to be allocated? The draft  
needs to either define an allocation policy (and IANA registry) or  
mandate that they MUST NOT be assigned.

Colin

_______________________________________________
Audio/Video Transport Working Group
avt at ietf.org
https://www1.ietf.org/mailman/listinfo/avt