[23431] in North American Network Operators' Group
Re: domain IN-ARPA.ADDR
daemon@ATHENA.MIT.EDU (bmanning@vacation.karoshi.com)
Thu Mar 18 08:11:50 1999
From: bmanning@vacation.karoshi.com
To: yjin@CS.UCLA.EDU (Jin Yixin)
Date: Tue, 16 Mar 1999 02:51:36 -0800 (PST)
Cc: nanog@merit.edu
In-Reply-To: <Pine.SOL.3.96.990315175621.6818E-100000@cheetah.cs.ucla.edu> from "Jin Yixin" at Mar 15, 99 06:02:07 pm
Errors-To: owner-nanog-outgoing@merit.edu
>
> I was trying to find a way to check whether an IP address is used by some
> host without actually probing it. DNS sounds like a good way to go, but
> I'v heard that domain "IN-ARPA.ADDR" is not well maintained and updated.
>
> Is that true? If so, how un-dependable it could be?
>
> Thanks a lot
>
> Yixin Jin
> UCLA, Computer Science Dept.
the rumors of poorly maintained in-addr zones are just that.
the data suggest that the forward tree was better maintained
in, roughly 1993-1996/7. Since then there has been higher
noise injection in the forward tree and the inverse tree has
more accurate data. Rough numbers indicate that regardless
of the database in question (whois, rwhois, DNS forward, DNS inverse,
any of the iRR databases (ripe, RA, MCI, ANS and some of the newer
ones like arin and apnic)) the data, as a whole, has about 40%
inaccurate data. the reasons why are many but can be boiled down to
at least two points:
- centralized repositories "distance" people from
the resources they have been delegated to manage.
- there are few operational requirements for maintaining
the accuracy of the data.
YMMV.
--bill
