[arin-ppml] IPv6 in the Economist

[Michael Loftis]

--On June 6, 2008 10:25:45 AM -0400 Dean Anderson <dean at av8.com> wrote:

> On Fri, 6 Jun 2008, Owen DeLong wrote:
>
>> Except you can't do name resolution if you turn off IPv4.
>>
>> I would say that's not full IPv6 support.
>>
>> I'd say that's minimal sort-of support at best.
>
> DNS is very severely broken in IPv6. This non-technical reason is that
> certain root operators want to keep their monopolies on anycast sales,
> and so (for technically inexplicable reasons), they have advocated
> mixing IPv6 and IPv4, and silenced dissent in apparent violations of

If you can get the tens of thousands of ASs to switch to v6 on a given 
0-hour then you must be closer to some God than any of us mortals.  A 
solution without mixed v4/v6 is impossible right now because there are 
still many people who are in the v6 world on an island compared to certain 
other destinations, this is improving, but it will be quite a while before 
one can say v6 connectivity is as good as v4, and I'm pretty sure most op's 
agree on that.  The only thing I can perceive that you want is for 
nameservers to answer with only v6 responses on v6 based queries, v4 on v4. 
And that's NOT the nameservers job.  It might be a resolvers job, but 
certainly NOT a nameserver.  A nameservers primary function is always to 
serve the requests of the client, be they for A, AAAA, NS, TXT, SPF, AFSDB, 
SOA, whatever RRtype the requestor desires.  There again I'm using some 
leaps of logic to come to that conclusion, so if my leaps are wrong, please 
enlighten us as to what you mean by v4/v6 mixing, and why it's bad, and how 
anyone could even possibly be able to effect a 0-hour switch.

anycast is a technical solution to providing more reliable, higher 
performance services, and has not a damn thing to do with address space 
depletion and IPv4 limitations. and therefore your statement about DNS 
being severely broken in IPv6 (I don't see this, I see a lack of deployment 
in ccTLDs especially and some TLDs, and a lack of support from registrars, 
but I've yet to experience any outright brokenness that can be blamed on 
DNS protocol, or implementations in common use)  And on what planet are the 
root operators selling anything?  They're not.  In fact they're donating 
huge amounts of equipment and labor just so you can make what appears to be 
an uneducated rant.

> anti-trust law.  So, there are no IPv6 root nameservers. Instead, one

Uh?
$ dig +short aaaa f.root-servers.net
2001:500:2f::f
$ dig +short aaaa a.root-servers.net
2001:503:ba3e::2:30
$ dig +short aaaa a.gtld-servers.net
2001:503:a83e::2:30

(OK that last one isn't a root nameserver at all, but more a proof)

So on what internet are you on?

> mixes IPv6 DNS records with IPv4 DNS records on the same nameserver.
> This totally unnecessary mixing creates stability problems for both IPv4
> and IPv6.  One has to remove IPv4 NS records to make room for IPv6
> records, so any effort to deploy IPv6 comes at the expense of IPv4
> stability. While bad enough, that isn't the worst part.

One only has to remove IPv4 NS records if one wants to remove IPv4 support, 
plain and simple.  They're not mutually exclusive.  Not by a long shot.

>
> What's worse is that the DNS resolver implementations are broken as
> well. One can't just create IPv6 root nameservers because the resolvers
> don't do the right thing--there is no IPv6-specific resolver which could
> use different root nameservers for IPv6. IPv4 and IPv6 have to be mixed
> at the roots on down.  Until this is fixed, IPv6 won't really be very
> useful or else both won't be stable.  Altering and updating resolvers on
> every computer is a very time-consuming job to say the least. So, I
> think IPv6 won't be taking over in 3 years, and IPv4 won't be going away
> in 3 years.

And you've spouted that OSI CLNS as a solution....Except last I checked the 
same problem would exist, worse, there's almost nothing outside of routers 
that actually supports OSI CLNS.

And we all KNOW v6 won't suddenly take over in 3 years, and we all KNOW v4 
*won't* be going away in 3 years.  I don't think anyone but net.kook's have 
ever said anything even remotely like that.  We're all going to be stuck 
with both for a long time.

>
> Its probably 10+ years to fix the resolver problem, and so a long time
> before IPv6 could be ready for stable deployment outside a lab.  In that
> time, I'd say we could go to OSI CLNS instead, and have much less risk.
> The good news is that one can work on both IPv6 and CLNS simultaneously
> as completely separate stacks.  Keeping CLNS separate from IPv4 this
> time will improve the process of development, and improve deployment
> stability later.

The only modern platform I'm aware of with a broken v6 resolver is Windows 
XP specifically.  And AFAIK it's only broken in the stance it can't use v6 
transport, for places stuck with Windows boxes they can use a IPv4 RFC1918 
DNS server that straddles the v4/v6 boundary until such time that Microsoft 
fixes it.

And yes, we're all very aware it could take many years before everything is 
stable in v6.  *HOW* does going with OSI CLNS help the matter at all? 
Other than divert resources.  You'd still need all the other pieces. 
Having an agreed upon protocol and addressing scheme to go with it is only 
the smallest part.  The hard part is all the software that *must* know 
about it, how to represent it, how to ask the OS to use it, how to find 
destinations on that address space/protocol.  Are there any OSI CLNS client 
stacks?  Seriously, the routers are a big problem, but far far far larger 
is the client and servers that will talk over a given protocol because 
there are many more implementations of those, and many more users of that 
software, and many more deployments of that software and hardware.






--
"Genius might be described as a supreme capacity for getting its possessors
into trouble of all kinds."
-- Samuel Butler