Re: [dhcwg] WGLC for draft-ietf-dhc-topo-conf-05.txt - Respond by July 29, 2015

[神明達哉 <[e-mail]>]

I've read draft-ietf-dhc-topo-conf-05.  It's generally well-written
and I see it useful for some class of readers.

I have one comment/question that might be better to be addressed
before sending it to the IESG (although I don't necessarily think it a
blocking issue): In Section 3.2 it sates:

   For completeness, we also mention an uncommon, but valid case, where
   relay agents set link-local address in the link-address field in
   relayed Relay-forward messages.  This may happen if the relay agent
   doesn't have any address with a larger scope.  Even though link local
   addresses cannot be automatically used to associate relay agent with
   a given link, with sufficient information provided the server is
   still able to correctly select the proper link.  That requires the
   DHCP server software to be able to specify relay agent link-address
   or a feature similar to 'shared subnets' (see Section 9).

First, I'm not sure how I should interpret this sentence: "This may
happen if the relay agent doesn't have any address with a larger
scope."  In its literal sense it seems to say the relay agent only has
link-local addresses.  Is that the intent?  If so, how can it even
forward messages to a server?  In theory, it's still possible if the
server (or the "next relay agent") and the relay agent shares the same
link (which might be a different link from which the relay receives
client messages).  But it's not clear if this sentence talks about
such a special case, and it won't work in the general case where a
server is located in a remote link anyway.  Or, should this sentence
actually read " This may happen if the relay agent doesn't have any
address with a larger scope on the interface to that specific link."?

Secondly, I'm not sure specifically what this means: "the DHCP server
software to be able to specify relay agent link-address".  Is there
anything specific intended here?  If so, I think it's better to
provide at least one example; if not, I'd rather remove this vague
statement.

I also have some minor (more or less) editorial comments:

- Section 3:

   Figure 2 illustrates a more complex case.  [...
   ...]  Note that some nodes act as routers (which forward all
   IPv6 traffic) and some are relay agents (i.e. run DHCPv6 specific
   software that forwards only DHCPv6 traffic).

Until we see discussions in Section 7 it's not obvious just from
Figure 2 that this is an example of IPv6 network using DHCPv6.  It
would be more reader-friendly to explain this is an example case using
IPv6 (and DHCPv6) explicitly here.

- Section 3.1: s/iserted/inserted/

   that is iserted by a relay agent.  This option is particularly useful

- Section 3.2:

   If for whatever reason that is not feasible (e.g. because the relay
   agent does not have a global address or ULA [RFC4193] configured on
   the client-facing interface), the relay agent includes an Interface-

To be pedantic "a global address or ULA" sounds a bit awkward since
ULA is also "global" in terms of address scope.  Perhaps "a globally
unique address or ULA"?

- Section 3.2: s/link local/link-local/

   doesn't have any address with a larger scope.  Even though link local
   addresses cannot be automatically used to associate relay agent with

- Section 7, Figure 4: Unless there's a specific reason, I'd suggest
  using "2001:db8::/40" etc, instead of "2001:db8:0:0::/40" based on
  the recommendation of RFC5952.  For the same reason, I'd replace
  "2001:DB8:0:0::/40" with "2001:db8::/40", etc, in the text that
  follows.

--
JINMEI, Tatuya