In honor of Ms. Hubbards pontification, I am starting a little archive:
This archive will be for people to send me (so we don't clog up NANOG)
Totally Clueless Allocation Policies of providers. So all you consultants,
send me your examples of hosed allocation policies from providers, large
and small. You don't have to name your clients, but please do name the
providers. I will summarize and forward to Kim and anyone else who cares
at the end of two weeks and maybe we'll get an idea of who allocates
correctly and who doesn't I'll start off with two examples:
1)
Billion dollar company. MCI is their upstream. 5 divisions on one campus.
Each has a mailgate, proxy, webserver, and router. Total need per
division: 4 addresses. Total allocation: 5 x /24 . % Used: 1.6 % Waste:
98.4 MCI's response when we said take em back: "Why?"
2)
Small office. PSI DDR LAN solution. 1 office, 2 PC's, one router. Total
need: 3 Total Allocation: 1 x /24. % Used: 1.18% % Waste: 98.82 PSI's
response when he said "All I need is a /29": "This is the way we do it.
It's just easier for us."
BTW Kim, I've gotten 4 responses in two hours on how to nail down all your
addresses so a ping sweep shows them in use. Might just be a LITTLE more
prevalent than ya'll think, eh?
Go back and sweep MCI and PSI and Sprint et. al. and SEE what they really
use. Then follow your own guidelines and maybe the startups won't bitch so
much...
You can bet when people start paying for addresses there gonna be pretty
pissed when they aren't fully routable.
Anyone care for the summary in two weeks? Mail me privately.
Eric
I'm beginning to think there is a market for a device which has 1 Ethernet
port and responds to any RANGE of addresses, so you can scam Internic into
thinking you have 100% utilization of your address space, right off the
bat...