[CCAMP] comments to draft-gandhi-ccamp-gmpls-restoration-lsp-01

["Zhangxian (Xian)" <[e-mail]>]

Hi, All, 

    Due to short of time, I post my questions to this draft here (based on the slides).  

1) what is the scope of this document? From the examples given, it seems to me only to end to end case. If it is not the intention, it would be good to clarify. 

2) Page 3 of the slides, last sentence. 
I do not think it is accurate. A restoration LSP can be established before failure as specified in RFC4872, unless you are targeting fully dynamic rerouting. So i think the following text in the draft need modification:

"Restoration LSP differs from a secondary LSP
   in the way that secondary LSP does not reserve resources in the data
   plane and is not able to carry any traffic until it is refreshed
   whereas restoration LSP does reserve resources and is able to carry
   traffic."

3) Page 8, the note. 
Could you point the exact piece of text saying this in RFC6689? At the moment, i cannot find it, thus cannot judge the need of this draft. 

4) Page 9: 
4.1: It seems to me the draft are using the bit P different from what it is defined in  RFC4872. Since it is a BCP draft, i do not expect this. Please clarify. 

4.2: not sure why assign a restoration LSP for a protection LSP? If the intention is to protection two failures, assign one protection and one restoration LSP for the working LSP is probably a better way to do. Since the failure of protection LSP does not won't trigger any update in RSVP-TE. 


Regards,
Xian