[arin-ppml] How bad is it really?

michael.dillon at bt.com michael.dillon at bt.com
Thu Jul 15 04:08:48 EDT 2010


> Indeed "Abuse"   is such a loaded word to use for a possible mistake.
>  I  would suggest the phrase  "Fraud and Abuse Process"  be replaced
> with:    Fraud and  Policy Violations Reporting Process

What is wrong with calling a spade, a spade?

It is in fact, the "Data Cleansing Process" because its goal is
to cleanse the database of errors and ommissions which have
accumulated over the years.

> Still,  at a certain extent,  certain mistakes are just negligence..
> ARIN orgs  have a responsibility to make some reasonable efforts to
> follow policy and keep their assignment records accurate.    It
> cannot be excusable (really) to leave a stale record in place for THAT
> many years.

You have no idea how hard it is to keep databases in sync in a large
company that has grown through acquisition after acquisition. It takes
a lot of effort and regular data cleansing audits. 

In addition, customers get migrated to new infrastructure, old address
blocks get repurposed. It is very easy to get into a situation where
you can identify that an IP address is in fact in use, but the record
is incorrect because it identifies a different use. Now you know that
you shouldn't delete the record, but in the absence of proper records
it is hard to know what the address block is being used for.

In addition, this is *NOT* the Internet addressing organization. IP addresses
are not exclusively for use on the Internet. It is perfectly legitimate
for an organization to use IP addresses on networks which do not peer
with the public Internet or exchange traffic with the Internet. So you
and I have no way of knowing whether that address block is in fact
unused. We can only say that it has a high probability of being
incorrectly recorded, but the outcome may be to change the record 
to show that it is used for infrastructure or some management network
or some internal lab that is intentionally not routes to the Internet.

--Michael Dillon




More information about the ARIN-PPML mailing list