Re: DoD IP Space

[Tom Beecher <beecher () beecher cc>]

Using the generally accepted definition of a bogon ( RFC 1918 / 5735 /
6598 + netblock not allocated by an RiR ), 22/8 is not a bogon and
shouldn't be treated as one.

The DoD does not announce it to the DFZ, as is their choice, but nothing
says they may not change that position tomorrow. There are plenty of
subnets out there that are properly allocated by an RiR, but the assignees
do not send them to the DFZ because of $reasons.

In my opinion, creating bogon lists that include allocated but not
advertised prefixes is poor practice that is likely to end up biting an
operator at one point or another.

On Tue, Nov 5, 2019 at 9:45 AM Töma Gavrichenkov <ximaera () gmail com> wrote:

> Peace,
>
> On Tue, Nov 5, 2019, 4:55 PM David Conrad <drc () virtualized org> wrote:
> > On Nov 4, 2019, at 10:56 PM, Grant Taylor via NANOG <nanog () nanog org>
> wrote:
> > > This thread got me to wondering, is there any
> > > legitimate reason to see 22/8 on the public
> > > Internet?  Or would it be okay to treat 22/8
> > > like a Bogon and drop it at the network edge?
> >
> > Given the transfer market for IPv4 addresses,
> > the spot price for IPv4 addresses, and the need
> > of even governments to find “free” (as in
> > unconstrained) money, I’d think treating any
> > legacy /8 as a bogon would not be prudent.
>
> It has been said before in this thread that the DoD actively uses this
> network internally.  I believe if the DoD were to cut costs, they
> would be able to do it much more effectively in many other areas, and
> their IPv4 networks would be about the last thing they would think of
> (along with switching off ACs Bernard Ebbers-style).  With that in
> mind, treating the DoD networks as bogons now makes total sense to me.
>
> --
> Töma