ICANN ICANN Email List Archives

[gnso-ff-pdp-may08]


<<< Chronological Index >>>    <<< Thread Index >>>

Re: [gnso-ff-pdp-may08] RSA Summary Data

  • To: Rod Rasmussen <rod.rasmussen@xxxxxxxxxxxxxxxxxxxx>
  • Subject: Re: [gnso-ff-pdp-may08] RSA Summary Data
  • From: Eric Brunner-Williams <ebw@xxxxxxxxxxxxxxxxxxxx>
  • Date: Fri, 26 Sep 2008 07:43:28 -0400


Rod,

Yinon's used "country determined by the domain registrar's location". I suspect he means "registrar" in a sense other than what is customary in ICANN policy discussion, possibly meaning the location of the registrant (or the technical or administrative contact). As dubious as the value of that particular representation may be, taking the location of the registrar of record for the location of a domain registered through that registrar seems peculiar to me, unless the point is the legal regime to which the registrar is subject, which is necessarily outside the scope (sorry) of the ajurisdictional (or crypto-Californian jurisdiction via ICANN's choice of incorporation) nature of GNSO policy.

There is something I didn't expect to see in Yinon's data, a distribution by registrar not simply defined by total market share. In the registries question, other data showed com/net/cn/org/info/biz or there abouts (from memory), or to a first approximation, total market share. I suppose the existence of a village in Romania and other actual clusters of people engaged in network exploits may be manifested in the marketing footprints of various registrars, rather than what barriers they present to attempts to acquire a resource at less than cost.

Because we work in domains, not "attacks", could you add a note that translates Yinon's metrics into metrics we can use?

Eric



Rod Rasmussen wrote:
Folks,

Finally got permission from the powers-that-be at RSA that this can be released and included in the public report appendices - with attribution to RSA of course. More body-of-knowledge stuff to add to the overall picture here. Thanks to Yinon Glasner of RSA for pulling this data together from their databases - I know all too well how hard it can be to get summary data from a database designed to run anti-crime operations!

Best,

Rod

Rod Rasmussen
President and CTO
Internet Identity
1 (253) 590-4088

Begin forwarded message:

I hope this is of use, although the numbers are quite crude. I'll start with some definitions and constraints related to the data: - Most of the figures here are aggregated, and include Fast-Flux attacks, Rock attacks and other phishing attacks which are hosted on Botnets - "Phishing attacks" – a "phishing attack" was counted as a unique URL that targets a specific brand. For example – a single Rock Phish domain, as you most definitely know, usually targets several brands. Under the same domain, 5 different URLs can host attacks against 5 different brands. So – we counted it as 5 distinct Rock Phish attacks. - In some cases I counted domains rather than "attacks". A domain is, well, simply a domain… J
-          Data range is January 08 – June 08
- In general, I think it will be very useful for you to say that in usual months, Rock and/or Fast-Flux attacks constitute more than 50% of all phishing attacks *_And now, the figures:_* 1. The number of Rock Phish *domains* detected:
Jan
        
403
Feb
        
443
Mar
        
450
Apr
        
420
May
        
461
Jun
        
190

2. The number of Fast-Flux / Rock Phish / Botnet phishing *attacks* detected:
Jan
        
770
Feb
        
1158
Mar
        
1408
Apr
        
1647
May
        
1866
Jun
        
608

3. The number of brands targeted by Fast-Flux / Rock / Botnet-hosted attacks – 40 4. The list of top-20 registrars where Fast-Flux / Rock / Botnet phishing domains were registered in 2008 + *the number of* *attacks* hosted on these domains. Top-20 registrars were measured by the number of *attacks*, not domains. Now.cn
        
1031
Eurodns.com
        
809
123-Reg
        
536
WildWest Domains
        
505
Key Systems
        
502
HKDNR
        
406
Melbourne IT
        
325
PublicDomainRegistry
        
305
Dot.tk
        
291
Register.com
        
273
Network Solutions
        
264
Today Nic
        
251
eNom
        
161
Nic.St - ST Registry
        
150
Yahoo
        
137
Tucows
        
133
COMMUNIGAL COMMUNICATIONS LTD
        
117
IA Registry
        
115
Bizcn
        
96
Online SAS
        
5. The list of countries where Fast-Flux / Rock / Botnet phishing *attacks* were hosted (country determined by the domain registrar's location): United States
        
2413
China
        
1481
Germany
        
482
Luxembourg
        
472
United Kingdom
        
452
Hong Kong
        
406
Tokelau
        
228
India
        
196
Australia
        
173
Sao Tome Principe
        
149
Afghanistan
        
148
Israel
        
139
France
        
128
Canada
        
114
Netherlands
        
83
Ukraine
        
63
Spain
        
56
Belize
        
41
Kyrgyzstan
        
39
Turkey
        
36
Switzerland
        
25
Taiwan
        
22
Belgium
        
19
Sweden
        
17
Estonia
        
16

I hope this helps, *Yinon Glasner* | Fraud Intelligence Business Analyst, OTMS | Identity and Access Assurance Group | **RSA, The Security Division of EMC** | Phone: +972.9.9728158 | Please note that my email address has changed to _yinon.glasner@xxxxxxx <mailto:yinon.glasner@xxxxxxx>_






<<< Chronological Index >>>    <<< Thread Index >>>

Privacy Policy | Terms of Service | Cookies Policy