Re: Consensus Call: draft-weil-shared-transition-space-request

CM

> One question, though, that I wish to address to the authors of
> draft-weil-shared-transition-space-request and perhaps others: why
> would not an allocation from 240/4 (the former Class E address
> space) work for CGN space?  I'm well aware that it would be very
> difficult to use this as ordinary IPv4 address space because of the
> long history of treating it as a "Martian" address range.  It seems
> to me, however, that this would NOT be an issue for CGN boxes --
> those being new equipment, the software can be upgraded to treat
> this address range differently than it traditionally has been.  It
> would be more difficult for CPE equipment to use -- especially stuff
> that's already deployed -- but that's actually an ADVANTAGE since
> such devices are not supposed to use CGN space.  And an allocation
> from 240/4 would not use up scarce global IPv4 space, which is one
> of the main objections I've been hearing to this allocation.

It may be possible to use 240/4 on the CGN box (code upgrade), but the
CPEs would have a problem.  These CPEs are a key part of the problem. The
space would need to work with what's in the field.  The CGN space would be
assigned the the CGN zone which may be configured on the internal
interface of the CGN box, but would absolutely be configured on the CPE
WAN interface. Numbering tens of CGN boxes is not the issue, numbering
thousands to millions of CPEs is.

240/4 would require firmware upgrades to many (if not all) CPEs, of which
many are not under the control of the operator (home user bought and
controlled).  It is not feasible to upgrade the entire CPE base.

This is one of the key issues with 240/4.

Regards,

Victor K


> So ... to the authors of draft-weil-shared-transition-space-request
> and other advocates of this allocation : please tell us whether an
> allocation from 240/4 would work for CGN space, and if not, why not.
>
> To the IESG: please require the authors and/or other advocates of
> the propsed allocation to answer the above question before approving
> the allocation request.  If they agree that it will do, approve an
> allocation from that space.  If they provide a cogent argument as to
> why 240/4 won't work, then I advise (reluctantly) approving the
> allocation from the remaining IPv4 global space.
>
> Thanks for listening.
>
> //cmh
>
> On Sat, 3 Dec 2011, Bernard Aboba wrote:
> > The same thought occurred to me.  A very large enterprise will not
> > utilize this /10 on a whim; they'd talk to their ISP first.  A
> > consumer is unlikely to modify the settings of their home router,
> > except if they download malware that does it for them :) and a
> > consumer router vendor has such a low margin that the last thing
> > they want is to utilize this forbidden /10, generating thousands
> > of tech support calls they can't afford to answer.
> >
> >
> > On Dec 3, 2011, at 20:54, "Henning Schulzrinne" 
> > <hgs(_at_)cs(_dot_)columbia(_dot_)edu>
> > wrote:
> > > Almost all residential customers will use a standard home
> > > router; as long as that home router does not make the new space
> > > available to customers, it will not be used. Almost all
> > > residential users get their home NAT box either from the ISP
> > > (who obviously won't ship such a box) or from one of a handful
> > > of retail consumer equipment vendors, who won't suddenly switch
> > > from RFC 1918 addresses, either (because they don't want to get
> > > the support calls).
> > >
> > > I don't think your consumer ISP will have much sympathy if you
> > > call them up and tell them that you decided to use 128.59.x.x
> > > internally, reconfigured the gateway and can no longer get to
> > > Columbia University.
> > >
> > > This is an economics issue: If one big corporate customer with a
> > > too-creative sysadmin calls up after "finding" this new address
> > > space, this can be dealt with.  (Indeed, that large corporate
> > > customer probably has non-1918 outward-facing addresses to begin
> > > with and will keep them, so they are the least likely target of
> > > CGNs.) If 10,000 consumer customers call up because their
> > > Intertubes aren't working, the ISP has a problem.
> > >
> > > Thus, I'm having a hard time believing in the theory that the
> > > new space will be immediately appropriated for consumer ISPs. By
> > > whom, exactly, and on what scale and with what motivation?
> > >
> > > Henning
> > >
> > > On Dec 3, 2011, at 8:36 PM, Noel Chiappa wrote:
> > >
> > > > > From: Doug Barton <dougb(_at_)dougbarton(_dot_)us>
> > > > > This argument has been raised before, but IMO the value is exactly
> > > > > zero. The fact that you have a finger to wag at someone doesn't make
> > > > > the costs of dealing with the conflict any smaller.
> > > > Perhaps. But I don't know the ISPs' business as well as they do. So
> > I'd like
> > > > to hear their views on this point. (They may well have considered
> > this point
> > > > before deciding to ask for CGN space, and decided the space was
> > still enough
> > > > use to be worth it.)
> > > >
> > > >    Noel
> > > > _______________________________________________
> > > > Ietf mailing list
> > > > Ietf(_at_)ietf(_dot_)org
> > > > https://www.ietf.org/mailman/listinfo/ietf
> _______________________________________________
> Ietf mailing list
> Ietf(_at_)ietf(_dot_)org
> https://www.ietf.org/mailman/listinfo/ietf

_______________________________________________
Ietf mailing list
Ietf(_at_)ietf(_dot_)org
https://www.ietf.org/mailman/listinfo/ietf