[ppml] Policy Proposal: Decreasing Exponential Rationing of IPv4 IP Addresses

[Dean Anderson]

On Sat, 25 Aug 2007, Iljitsch van Beijnum wrote:

> On 24-aug-2007, at 20:29, Dean Anderson wrote:
> 
> >> Stretching IPv4 only means that people will delay the move to IPv6,
> 
> > This seems to be the unproven assumption
> 
> Hm, IPv4 addresses are still available and few people are using IPv6.  
> Maybe not "proof" 

Air is still available, and few people are using IPv6.  Perhaps its not
the air, or the IPv4, but rather IPv6 that is the problem.

> but so far the facts are on my side.

How's that??? I think perhaps you've miscounted the facts. Lets review
that facts:

Fact 1: The Available IP Pool (AIP) will be exhausted on or about March,
2010 if no changes are made.

Fact 2: If the AIP is exhausted, it is unknown when address space will 
be returned that can be delegated. 

Fact 3: Disruptions of unknown duration are more harmful to 
business planning than disruptions of known duration.

Corollary 3: Disruption of known duration can be managed by changes
business plans.

Fact 4: Hoarding and other anti-social behavior will increase as a
limited resource nears exhaustion. The AIP has no special immunity from
this feature of human behavior.

Fact 5: Rationing any kind of limited resource inhibits hoarding, and
mitigates the effect of exhaustion by creating gradually smaller and
more predictable reductions in resource consumption.

Fact 6: IPv4 usage will not end because of AIP exhaustion.

Fact 7: IPv4 allocation will eventually resume after exhaustion, no 
matter what happens. However, after exhaustion, we won't be able to 
predict when allocation might resume.


Your reason for objection is to promote IPv6.  IPv6 should have nothing
to do with IPv4 resource management.

ARIN has an obligation to responsibly and prudently manage IPv4
resources.  Avoidable resource exhaustion, and the consequently
avoidable problems with exhaustion is not a responsible act.

Perhaps we need to separate IPv6 address management from IPv4
management, especially if we can no longer trust ARIN to manage IPv4
responsibly and without sabotaging IPv4 to promote IPv6.

> > IPv4 will continue after exhaustion of the Available pool.  Address
> > space will be returned or reclaimed, and allocations will eventually
> > resume.  Exhaustion in this case just creates a bump in IPv4
> > operations.  That bump, and consequences of the bump, can be
> > avoided.
> 
> Yes, and after all the cars have been turned into big piles of rust,
> you'll be able to get oil again, too. But that doesn't mean things
> revert back to the way they were before.

We won't have to wait thousands of years for IPv4 address to become
available again. No doubt MERIT will be returning a great deal of its
dialup pool soon, since they no longer operate a large dialup network.  
Etc. Others may also be able to return some space.  Indeed, people have
proposed all sorts of (unnecessary) things to get back space.

> > However, there is no case where that bump greatly improves the
> > transistion to IPv6.
> 
> There are enough extremely pigheaded people in the world, it's
> possible that they'll forego IPv6 even when IPv4 moves past its
> expiration date.

There is no expiration date for IPv4.  That's your fallacy (or fantasy).

> > Your scheme seems to be desperate and makes me more suspicious of
> > IPv6.  If IPv6 could stand on its own merits, you wouldn't need to
> > try to wreck IPv4 to promote it.
> 
> I'm not trying to wreck IPv4.

I think you are. You are attempting to stop prudent action to avert a
foreseeable and avoidable problem with a shortage in IPv4 IP Address
Availability.  That doesn't help keep it in a "usable state".

Your justification for your view is that you want to get IPv4 'over and
done with' to move onto IPv6.  That's just sabotage of IPv4.

If you need to sabotage IPv4 to get people to move to IPv6, there must
obviously be something wrong with IPv6.

> Right, and so many people are now running AppleTalk, IPX, DECNET and
> so on.

There are indeed. IPX is still a commercial product.

> China has gotten 30 million addresses so far this year. That's more
> than all year last year. Once China has caught up, the really poor
> countries are next in line. IPv4 can't sustain our growing
> communication needs, it's as simple as that.

Then they'll be happy to move to IPv6. You won't have to mismanage IPv4
to encourage them.

> > I'm going to have to look into IPv6, to find out what's wrong with
> > it, that you would need to disrupt Ipv4.

> I recommend reading a good book about it.

Its kind of hard to find a good book that hasn't been obsolesced by the
continuing changes to IPv6, judging by the RFC index.

I've been following the IPv6 DNS changes for 10+ years.  I can't say
that all of IPv6 suffers these same problems. The DNS changes seem to
work, but not well. Indeed, I think that is the slogan for IPv6: "It
works, but not well".

Here's one DNS example: Recently, a Country code (cctld) domain operator
tried to add IPv6 AAAA record support to their cctld name service. If I
recall correctly, they found that they could only have 6 A records, and
3 AAAA records before they ran out of the (IPv4) packet size limitations
(512 bytes).  Of course, the crazy question is "why mix IPv4 DNS and
IPv6 DNS?"  No one ever did that with IPX, Appletalk, DecNet, SNA, or
anything else.  Was this a stroke of genius?  No. The one certain answer
is that mixing is good for the existing root and TLD operators, since
they can continue using their existing servers (cheap), and they don't
have to worry about competition for a new set of IPv6 root and TLD
server operators; No one else will be taking over their profitable
franchises.

There is no technical reason that IPv6 DNS should talk to an IPv4
nameserver.  No more so than an IPX nameserver needs to talk to an IPv4
nameserver.  So, there is a lot of cruft and limitation in IPv4 DNS that
would be quite good to remove from IPv6.  None of that happened. So,
IPv6 DNS is a clusterf*ck of IPv4 limitations imposed on IPv6 computers,
and the only reason for that is so that certain people keep their
franchises.  IPv6 is designed in someways to profit some people.  
That's part of the reason it sucks.

The other part is that IPv6 seems to be so unstable (due to continuous
changes) that nothing works very well.  Its so bad that people don't
choose IPv6 for off-net things they use RFC1918 space for. The DOD was
supposed to move to IPv6. That seems to have stalled. Why?

The IPv6 problems have nothing to do with lack of connectivity to the
rest of the network.  If that were the case, there would be a ton of
users migrating their RFC1918 spaces to IPv6, and then complaining that
they still had to use IPv4 with their ISP. But ISP get few or no such
complaints.

You're right that IPv4 will have problems with address space. We've
known that for 20 years.  People were supposed to be working on that
problem. But instead of honest science, they put their own interests
ahead of the science of a better protocol suite.  I'm not certain what
the solution is for IPv6. I think we can fix it and make it work.  Just
about anything can be made to work. On the other hand, maybe we have to
start over with more honest people. But there is no reason to damage
IPv4 in the meantime.



> >> My assumption of what would happen when we run out (in the absense
> >> of new policies) is that if a request comes in for more address
> >> space than is available, the request is denied A request for a
> >> smaller amount of address space that can still be accommodated
> >> would be honored. If you want your policy to work like that, you
> >> should say that in the proposal.
> 
> > Well, this happens with or without rationing.
> 
> You didn't specify that in your proposal.

It doesn't need to be in the proposal because it's not relevant to
rationing.  ARIN, I think, handles this situation already quite well. In
any case, its not a change particular to rationing.

> > None of your arguments make any difference under rationing or not
> > rationing. So, they don't have any relevance to rationing.
> 
> If you are an ISP that is going to connect a million customers in the
> next few months, you need a million addresses. If you only get 800000,
> 80000, 8000 or 800, you have a problem. The severity of the problem
> may differ slightly, but in each case, you have a problem that you
> don't have if you get that million addresses. By giving out address
> space while it lasts we don't start creating problems before we have
> to so people have more time to be proactive.

This also happens in March 2010 without rationing. Rationing mitigates
the problem by creating smaller stoppages and a predictable slowdown. 
Businesses can deal with predictability.


		--Dean


-- 
Av8 Internet   Prepared to pay a premium for better service?
www.av8.net         faster, more reliable, better service
617 344 9000