Private use of non-RFC1918 IP space

Måns Nilsson mansaxel at besserwisser.org
Wed Feb 4 17:09:22 CST 2009


--On onsdag, onsdag 4 feb 2009 17.44.20 -0500 Ricky Beam <jfbeam at gmail.com>
wrote:

> On Tue, 03 Feb 2009 20:29:36 -0500, Skeeve Stevens <skeeve at skeeve.org>
> wrote:
>> I agree... I'd love to know where they got that from... who even wrote  
>> it?
> 
> I see you've never done business with EDS.  They've been using 1/8 for
> over a decade.  Also, over the years, I've seen a number of universities
> and supercomputing facilities number nodes out of 1/8 -- however, those
> systems are never supposed to see the internet anyway, so they could
> technically number them however they want.  Personally, I've used 1/8 in
> lab setups.

Last time I built a supercomputer (as in a cluster of run-of-the-mill
servers) RIPE gave us a /21 -- I wanted a /20 for the expected upgrade but
was told I had to reapply. The compute nodes need to read files from AFS
from CERN or another university, so NAT'ing them is so not an option. 

-- 
Måns Nilsson			M A C H I N A

Sign my PETITION.
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 186 bytes
Desc: not available
URL: <http://mailman.nanog.org/pipermail/nanog/attachments/20090205/8bf8939c/attachment.bin>


More information about the NANOG mailing list