legacy /8

George Bonser gbonser at seven.com
Sat Apr 3 12:15:21 CDT 2010

> -----Original Message-----
> From: Mark Smith
> [mailto:nanog at 85d5b20a518b8f6864949bd940457dc124746ddc.nosense.org]
> Sent: Friday, April 02, 2010 11:09 PM
> To: George Bonser
> Cc: John Palmer (NANOG Acct); nanog at nanog.org
> Subject: Re: legacy /8

> Y2K was a bit different though - there was no alternative other than
> fixing it. "Carrier grade NAT" didn't exist for YY fields, so there is
> no actual deadline for IPv6 like there was for Y2K. I think that is
> what is leaving room for people to think they don't have to deploy it
> soon, or that it has failed.
> Regards,
> Mark.

I think part of the problem with v6 is that it is still sort of a moving
target.  There seem to be a lot of ways of doing this or that and it
seems new ones come out a couple of times a year that are incompatible
with other ways of doing things.  Things such as automatic address
assignment and communicating with v4 from v6 are just a pain in the hips
at the moment, there is not ironclad, cast in stone, "this is how it is
done, period". 

But if you look at other technologies, it was the network operators that
dragged them into play through necessity.  Nobody had to tell anyone
that they must put MPLS into their network or VRFs or BGP or whatever.
The operators dragged those technologies in because they needed them.
For the vast majority of operators, the only problem v6 is going to
solve (admittedly a huge problem) is address depletion and a lot of the
other "features" are going to go unused by most people but have to be
dealt with (turned off with a big hammer in many cases).  We have 128
bits of addressing, you would have thought someone might have decided
early on to come up with something simple like ... the first 32 bits are
your ASN and that is your "net block".  Have another facility that is
not connected to the first?  Get another ASN.

But this is all water long under the bridge.  At the moment we are stuck
with what we have.  We really don't have the choice at this point but to
hold our noses and use v6 and that is going to cause a lot of problems
in a lot of places with a lot of vendors.  Actually, to many operators a
depletion of v4 addresses is a good thing because it serves as a barrier
of entry to competition.  I would guess that people who have more than
enough v4 space currently are glad to see it hard to obtain and also
glad to see nobody moving to v6.  It just means more pie for them.

More information about the NANOG mailing list