Re: [v6ops] M/O flags and PD

[Tore Anderson <[e-mail]>]

Hi Gert,

* Gert Doering <[e-mail]>

> I do not have RFCs to counter that, but my interpretation is a bit
> different.  I see that section of 4861 as traditionally applied to 
> *hosts*, while IA_PD traditionally is a *router* function, so out of 
> scope for the "host part" of 4861.

Section 4.2 of RFC 4861 simply describes the RA message format, it's
not in a "host part" of the RFC as far as I can tell. Keep in mind that
RAs are typically multicast to all IPv6 nodes, so the originating router
does not know whether it is a host or a router (or both, or none) that
will be receiving the RA.

Thus, when determining whether or not to set the M/O flags, the
[operator configuring the] RA originating router will have to ask
himself the following question:

«Is there information available through DHCPv6 to nodes on this link?»

If the answer to that question is «yes», the M or O flag should be set.
If the answer is «no», they should be both unset.

> Now this gets all muddled together when you have a router that also 
> wants an address for itself (= acting as a host for that purpose) or 
> a host that wants a prefix (= serving as a router for that prefix).

I'll note that this discussion started with a comment on
draft-jjmb-v6ops-unique-ipv6-prefix-per-host. This draft talks about
hosts, not routers. It even says so right there in the file name and
the title.

Then you have RFC 7084 which to the best of my knowledge is describes
the only use case for DHCPv6-PD which has seen significant deployment
to date. Its WPD-4 requirement clearly indicates that the availability
of DHCPv6-PD is something that should be communicated through the M/O
options.

The WPD-4 requirement essentially states that if you operate a network
that provides IA_PD and sets M=O=0, it's unspecified ("out of scope")
whether or not any connecting devices conforming to RFC 7084 end up
with a delegated prefix or not.

Tore