[75035] in North American Network Operators' Group
Re: Big List of network owners?
daemon@ATHENA.MIT.EDU (Tom Vest)
Thu Oct 28 14:18:37 2004
In-Reply-To: <200410281725.i9SHPwxg021357@turing-police.cc.vt.edu>
Cc: Joe Abley <jabley@isc.org>, John Underhill <stepnwlf@magma.ca>,
nanog@merit.edu
From: Tom Vest <tvest@pch.net>
Date: Thu, 28 Oct 2004 14:17:14 -0400
To: Valdis.Kletnieks@vt.edu
Errors-To: owner-nanog-outgoing@merit.edu
On Oct 28, 2004, at 1:25 PM, Valdis.Kletnieks@vt.edu wrote:
> On Thu, 28 Oct 2004 13:12:39 EDT, Joe Abley said:
>
>> Poke around the ftp sites of the four RIRs until you find address
>> registration data. Don't expect to see a single dump format across
>> RIRs.
>
> For bonus points, does anybody have a good estimate of what percentage
> of the registration data doesn't match reality, due to missing SWIPs
> and the infamous "allocated to a reseller who allocated to a
> re-re-seller
> who..." issues? (Not talking actively hijacked, just all the "forgot
> to file the paperwork" allocations...)
We're working on this question at the operator (ASN) level for a couple
of projects. I can't produce a list immediately, but there seem to be
at least 600-700 ASNs that were consistently routed between Oct 01 and
Oct 03 that have no easily matchable whois data in any registry.
Probably the best you can come up with the the converse; the percentage
of operators who take the (varied kinds of) trouble to identify
themselves broadly to the community, thereby making themselves at least
implicitly available for large-scale event management, etc. I think if
you sum up the unique users of various extra-whois tools (nsp-sec,
INOC-DBA, Jared's NOC list, etc.), you come up something like 3-4k
operators. For those 3000+/- you can be reasonably confident that their
whois data is correct; the other 15.5k actively routed ASNs (much less
the routed netblocks, and less still the idled ASNs and netblocks) are
anyone's guess...
Tom
