Document: draft-ietf-l1vpn-ospf-auto-discovery-05.txt
Reviewer: Avri Doria
Review Date: 27 March, 2008
IETF LC End Date: 26 March, 2008
IESG Telechat date: (if known)

Summary:

I think this draft is nearly ready for publication. It is clear and the explanations are complete and understandable.

I have two questions below as well as several suggestions and bit of proof reading trivia.  The suggestions are minor but offered for completeness sake.

Comments:

a. Questions

On Page 8 - there are 2 instances of SHOULD

1.  re:

 > If more than one L1VPN Info TLVs and/or TE Link TLVs are found in  > the LSA, the subsequent TLVs SHOULD be ignored by the receiving  > PEs.

Why not MUST?

2. You indicate that in the case that a new/remove/modified L1VPN LSA is received and the appropriate PIT can be found the PE SHOULD add remove or modify.  In which case would it not need to.  I.e. why is this a SHOULD instead of a MUST?

b. Suggestions:

1. On page 5 LS type 11 - might be good to give its name as well, i.e. that it is the AS-scope Opaque type, even though the next block makes that obvious.  While it is explained in the text below, it would be useful to someone reading the document who did not know/remember what type 11 was.

2. I think it is very useful that you included a list of common L1VPN terminology.  It might also be helpful to include in the terminology section the OSPF is used, e.g.

The reader of this document should also be familiar with [rfc for ospf].  In particularly the following term:

LSA ...

c. proof reading trivia:

Page 5

 > An L1VPN specific TLV is defined in below to propagate VPN  >  membership and port information.

drop 'in'

 > it does participate in the flooding of all of the L1VPN LSA

should be LSAs