Ballot for draft-ietf-bmwg-ipv6-tran-tech-benchmarking
Yes
No Objection
Note: This ballot was opened for revision 07 and is now closed.
I was surprised not to find any mention of RFC 4380 (Teredo) in this document. If its omission was intentional, a statement to that effect (say, in the introduction) is probably warranted.
I note that this document discusses using jumbo Ethernet frames up. The IEEE has a standard now for baby giant Ethernet frames - though there is still concern about the EtherType value used. An informative reference might be helpful.
Section 4. (Test Setup): In terms of route setup, the recommendations of [RFC2544] Section 13 are valid for this document assuming that an IPv6 version of the routing packets shown in appendix C.2.6.2 is used. However, rfc2544 says in several places that the packets in the appendix are just examples. The frame in C.2.6.2 is a RIP update -- but Section 11.3 references the rate at which "frames SHOULD be sent" (also in the appendix) which include OSPF and IGRP, so I'm assuming that any routing protocol used should work (if the recommendations are followed in terms of frequency, etc.). I note that rfc5180 doesn't really say anything about routing setup for IPv6 either. :-( I know this is not the document to define a complete set of (or even update) recommendations for routing setup, so my suggestion is to simply take off the reference to the appendix: In terms of route setup, the recommendations of [RFC2544] Section 13 are valid for this document assuming that IPv6 capable routing protocols are used.
-7.2, Reporting Format: Is it conventional to use 2119 keywords to describe report formatting? (Or is this paragraph really about content, rather than format?) (Comment repeats for other "format" related paragraphs.)
* I am surprised that this document does not use (and does not even mention) the Well Known Prefix (64:ff9b::/96) for the algorithmic mapping between IPv4 and IPv6 on translators as specified by RFC6052. Is there a reason why this is omitted? * It is not clear from the document whether the time taken by the DNS64 resolution procedure is included in the latency measurements. It might be useful to note this.