Draft:  draft-ietf-rddp-ddp-06.txt
Reviewer: Francis Dupont [Francis.Dupont@point6.net]
Review Date: Monday 7/31/2006 11:44 AM CST
IETF LC Date: 8/02/2006

Summary: not Ready

I have the same concern than for the companion I-D about DDP: the IPsec
part (8.2.2) refers to an obsolete version of IPsec/IKE.

Detailed comments:
 - i.e. -> i.e., and e.g. -> e.g.,

 - 1.2 page 7: With -> with

 - 1.3 figure 2 page 11: I suggest to add // and lines to the payload.

 - 3.2 page 26: what is the "yesSTag" (IMHO there is a typo)?

 - 4.1 page 28: what are IETF RNICs and RDMA RNICs (and for the second
   the "R" in RNIC stands for RDMA, doesn't it)?

 - 4.4 pages 32/33: the title "RDMA Read Message Size" is very ambiguous
   (i.e., the common meaning is not the intented one).

 - 4.8 figure 9 page 37: the Nones for Local Catastrophic Error doesn't
   specify what to put in the (BTW not optional) fields.

 - 5.4 page 46: the DDP layer mark -> marks?

 - 5.5 20. page 50: more than one ... is -> are?
   (same 7.1 24. page 55)

 - 7.1 21. page 55: Errors -> errors.

 - 7.1 24. page 55: the rules 2 and 3 above are really 2 and 3 of 5.5?
   Or are they 22 and 23?

 - 8.1.1 8. page 58: range available -> available range.

 - 8.2.1 page 60: RFC2401 -> RFC4301 and IPsec an guarantee anti-replay,
   not sequencing.

 - 8.2.2 page 61...: look at my comments about DDP I-D.

 - 8.2.2 7. page 62: I disagree with the recommendation. I suggest
   to cite directly the draft-ietf-pki4ipsec-ikecert-profile-10.txt
   document than to overload the CERTREQ payload which as its name
   suggest requests that the peer sends a CERT payload through IKE...

 - 10.1/2 page 65: RFC240x -> RFC430y.

 - 10.2 page 66: RFC2246 -> RFC4346 (and TLS 1.0 -> 1.1).