Re: [v4tov6transition] draft-arkko-ipv6-transition-guidelines WGLC

[Cameron Byrne <cb.list6@gmail.com>]

On Wed, Aug 18, 2010 at 12:45 PM, Hemant Singh (shemant)
<shemant@cisco.com> wrote:
> None of the items in the grocery list below seem to warrant a BOF or new WG.  V6ops is fully capable and chartered to handle this list including not being that busy to not find time for the new documents or questions.  In fact, just blast new drafts to the v6ops mailing list or ask these questions of v6ops mailer and folks will reply.  If a draft sent to the mailer is not being discussed, let the Chairs know and they will designate reviewers.  In fact some of the questions asked can be answered right now.
>
> How do you assign an address in your network?
>   (recommended prefix length and value of interface ID)
>
> For hosts in your network see if SLAAC can be used in your network and SLAAC defaults to a /64. If not SLAAC, use DHCPv6 and see what sense it makes in your network to use for prefix length.  If your network is a SP network with edge and core routers then addressing such routers is also a well-know operation besides the controversial /127. You have to be more specific than the question above.
>
>  How do you use link-local?
>
>   For ND operations and any other IPv6 control that travels over the link-local like MLDv2, and DHCPv6.
>
>  Is there RFC1918 space in IPv6?
>
>   The ULA.
>
>  Is there such a thing as secondary address with IPv6?
>
>   No.

These sound more like training issues that are better handled by ops
folks, and ops folks generally participate more in groups like NANOG
and other *NOGs.  There are also some very relevant IPv6 operations
mailing lists that handle tactical issues.

As Hermant points out above, most of these questions have straight
forward immediate answers.  Generally, addressing a "HOW" is an
operator specific issue that depends largely on the context and
problem to be solve.  Budget, existing systems, timeline ... they
determine the "HOW", not the IETF.  Nobody told me "HOW" to build my
IPv4 networks, and if somebody tried, i am unsure the "HOW" would be
universally.

Also, to be frank, i do not believe the majority of the participants
in the IETF even work or have current experience at network operators.
 The IETF folks are great and very helpful and Jari's draft is a good
example of how the IETF can be helpful.  Jari has helped paint the
picture and catalog the options, and he has not gone further than that
because there would not be value in going further.

Cameron