internet-dra...@ietf.org wrote:
> A diff from the previous version is available at:
>
https://www.ietf.org/rfcdiff?url2=draft-ietf-homenet-front-end-naming-delegation-24
Version 24 is a response to Eric's unicast review comments, posted with
permission.
Eric Vyncke (evyncke) <evyn...@cisco.com> wrote:
> Some comments:
> - figure 1, on the right hand side there is the reverse zone but not on
> the left hand side ? (I would have added .local on the LHS as well)
a) I've added x.y.z.ip6.arpa to the left-hand side. I'm concerned that this
may make some people think that only the delegating ISP can offer this
service.
b) .local isn't really a zone. It's a convention that if you see it, then
you use mDNS. So there never really is a zone by that name.
> - section 6.1, I wonder how plain DNS AXFR can be used to retrieve all
> the information required by HNA
Most of the detail is in section 6.5, and I've punted more clearly to that
section.
I've edited section 6.5 to clarify the templateness of things.
> - section 6.3 is a little unclear about "with its new IPv4 and IPv6
> addresses" (plural) and the rest of the section "the IP address"
> (singular)
I've removed the IPv4 update, as I don't think that we support zone transfers
over IPv4.
> - section 6.5 " encouraged." Should this be a normative "RECOMMENDED" ?
I guess we SHOULD go that way.
> - section 7 " DNS updates are to sent over the Control" should this be
> "DNS Notifies" ? (my spell checker does not like "are to sent")
Thank you, are you are correct.
> - section 9 rather than using "WAN interface" should it be "public
> interface" ? Plural form is used but HNA listens on only one interface
fixed.... however... "public" interface is actually less meaningful in IPv6
world. So I changed it to "internet facing"
In theory, the HNA could be on a device with multiple uplinks.
I'll remove (s) for now.
> - section 10, should the obvious be stated ? I.e., only applicable to
> IPv6 ?
I've thrown in "IPv6" in twice.
> - section 11, " It is RECOMMENDED the HNA sign the Public Homenet
> Zone." But early in the I-D, it was asserted that HNA signs the zones,
> i.e., use MUST and plural form (or is the reverse zone not signed) ?
If there is signing, then the HNA does it, not the DM.
I've removed "also" from the early section.
We have RECOMMENDED here, yes.
Let's change that. RECOMMEND is a SHOULD, and the exceptions are unclear to me.
In particular, I want the zone signed even if the delegation is not secured.
This forces implementations to be ready for a secured delegation.
I've posted -24.
--
Michael Richardson <mcr+i...@sandelman.ca> . o O ( IPv6 IøT consulting )
Sandelman Software Works Inc, Ottawa and Worldwide
signature.asc
Description: PGP signature
_______________________________________________ homenet mailing list homenet@ietf.org https://www.ietf.org/mailman/listinfo/homenet
