RE: for folks tracking DDOS sources or reading the GRC attack log

[woods () weird com (Greg A. Woods)]

[ On Monday, June 25, 2001 at 20:12:11 (-0400), Richard A. Steenbergen wrote: ]
> Subject: RE: for folks tracking DDOS sources or reading the GRC attack log
>
> Well since I don't think you can argue Canadian cable modems follow
> different patterns, you seem to have contradicted yourself...

Well, I can argue that some Canadian cable modems do follow different
patterns.  Although GT are absolutely terrible at maintaining their
network assignments, I know for a fact that chunks of both
GROUPTELECOM-BLK-5A and GROUPTELECOM-BLK-6A are assigned to operating
cable modems.  I'm also fairly certain that when the cable provider I
refer to finally gets their own assignment from ARIN that it won't
likely be from 24/8 (though it might -- it's anyone's guess at this
point since I don't the application has even been made yet).

Then of course there's NETBLK-GTE-CABLE-DUKE-ADSL.  Is that an oxymoron,
or just an example of a contradiction to your argument?

Making assumptions about the type of last-mile connection in use by some
IP address based solely on its classical prefix (eg. 24/0) is just never
going to be accurate.  Making assumptions about where a classical prefix
is routed geographically is going to get you in real trouble.  From what
I can see 24/8 can be found on many continents, never mind in many
countries.

-- 
                                                        Greg A. Woods

+1 416 218-0098      VE3TCP      <gwoods () acm org>     <woods () robohack ca>
Planix, Inc. <woods () planix com>;   Secrets of the Weird <woods () weird com>