Re: [Tsvwg] Review of draft-ietf-tsvwg-rsvp-security-groupkeying-04

["Carlberg, Kenneth G." <[e-mail]>]

Francois & Michael,

Here is another review of the draft.  Its mostly comprised of nits and
requests for additional clarification.

-ken


1) page 4,  "Implicitly each node trusts each other node with which
   it has a trust relationship established via the mechanisms here to
   adhere to the protocol specifications laid out by the various
   standards."

The sentence is a bit long and confusing.  You need to break up the
sentence and clarify the content of the word "here".  Also,
s/other/other's
 

2) page 5, 3'rd para.  s/R2 and will ignore/R2 will ignore

3) page 5, 5'th para.  
   "With neighbor based keys, an RSVP key is bound to an interface plus
a
   neighbor on that interface.  It allows the distinction of different
   trust groups on a single subnet.  (Assuming that layer-2 security is
   correctly implemented to prevent layer-2 attacks.)"

Does the above text concerning neighbor based keys allow for different
trust groups per interface?  If so, this should be stated to the reader.

4) Page 6, 5'th para.  s/domains however/domains. However

5) page 14, 2'nd para.
   "In
   particular, such keying can be used for RSVP authentication (e.g.,
   using RSVP authentication or IPsec AH) and/ or for RSVP encryption
   (e.g., using IPsec ESP in tunnel mode)."

The sentence is a bit unclear with respect to using the term RSVP
authentication as an example of keying "used for RSVP authentication".
You'll either need to clarify the text, or just rely on IPsec AH as a
single example.

6) Page 15, late sentence of 4'th para.  
   "Therefore the impact of subverted nodes on the path is comparable,
   independently whether per-interface, per-neighbor or group keys are
   used."

The reader would probably ask, comparable to what?  It would be helpful
if you could complete the object of the sentence.