[ARIN-consult] Community Consultation on IRR Route Validation. attorney client communication.

[Tauber, Tony]

On 3/31/15, 5:47 PM, "Hannigan, Martin" <marty at akamai.com<mailto:marty at akamai.com>> wrote:


On Mar 31, 2015, at 4:52 PM, Gert Doering <gert at space.net<mailto:gert at space.net>> wrote:
Hi,
On Tue, Mar 31, 2015 at 08:28:08PM +0000, Hannigan, Martin wrote:
AS 32787 and AS 20940 will continue to use and expand use of RADB for the foreseeable future.
So, what do you suggest how the gaping security problems of RADB can be
solved?

The same as we would when they arise in any other IRR? Point fingers of shame, get commitments to fix or go elsewhere?

A couple of weeks ago I posted in this thread:
http://lists.arin.net/pipermail/arin-consult/2015-March/000597.html

to ask which Standards would be used to achieve what was in the original proposal:
https://www.arin.net/participate/acsp/suggestions/2015-3.html

  *

2015.3: Tie Route Objects in IRR to Netblocks of RIR Database

One way to do this might be RPSL authorization such as described in RFC2725<https://tools.ietf.org/html/rfc2725>.
I believe something like that is what RIPE does, but haven’t had any response to my particular query.
It is something that can happen where the address assignment is aligned with the RR (which ARIN didn’t do with their existing RR).   I don’t see a way for RADB or operator-based RRs to do it, but I may lack imagination.

Tony

Now, I understand that you can buy a maintainer object, and thus protect
your own address space - but as a user of the RADB, how can I see which
objects are legitimate, and which (paid-for!) maintainers just put in
stuff that does not belong to them?

How will that differ in an ARIN IRR?

Best,

-M<
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.arin.net/pipermail/arin-consult/attachments/20150401/a2c7dc12/attachment.html>