Re: Support for 240/4 and 0/8 addresses in NetBSD

[Mouse <mouse%Rodents-Montreal.ORG@localhost>]

>   A) Are any of the addresses that are currently not allowed by
>      standards actually in use?

I would say "probably".  Not transiting the DFZ, presumably, and
obviously not with a proper assignment, but there is too much potential
for people to grab a piece of, say, 240/4 and start experimenting with
it internally for me to believe that it hasn't happened.  Especially in
view of the known incidents of people using properly assigned address
space that just happens to be assigned to someone else.

No, I'm not aware of any specific examples.

>   C) We hear that the documents are not yet standards-track work
>      items, and people offlist have told me (paraphrasing vastly)
>      that this is because "IETF complicated" and "IETF more political
>      than technical these days".  If that's true, then it seems that
>      for this to become reality so that A is true, documents will
>      need to progress through standardization, to the point where
>      registries are willing to issues addresses.

Not really.  That's one way to get them in use.  But (A) could also be
made true by people doing as I sketched above: grab a piece of address
space and experiment with it.

I'm now running with lo0 configured to 127.0.0.1/16 instead of /8 on
almost all of my machines, with the last one to follow soon, probably
today sometime.  Once that's been running for a while without problems,
I plan to start using a few other addresses from 127/8 but not 127.0/16
(internally to my house network, of course) to see what breaks.
Similarly, I expect to experiment with pieces of 240/4 and 0/8.  I do
not plan, nor expect, to wait for a proper assignment (if indeed I
could get one, which I doubt I could).  (I will, of course, make a
point of not using them in the "as if" sense, that is, keeping the
experiments confined to my house network.)

>      Assuming that's true, why is it useful for NetBSD to do
>      anythning now, aside from stepping into IETF politics via B)2) ?

The most obvious answer, it seems to me, is "so as to know NetBSD is
ready to go if/when the address space starts getting assigned and
used".

Next most obvious, to me, is "to be a platform for testing".  Once
NetBSD interoperates with itself with these pieces of address space, it
can be used as a test peer for other systems.

> All that said, dyoung@'s idea of replacing the kernel checks with
> REJECT routes has merit.

I concur.  When I saw it I was a little bit surprised it hadn't
occurred to me to suggest that.

/~\ The ASCII				  Mouse
\ / Ribbon Campaign
 X  Against HTML		mouse%rodents-montreal.org@localhost
/ \ Email!	     7D C8 61 52 5D E7 2D 39  4E F1 31 3E E8 B3 27 4B