A Noteworthy Report on Fast Flux Hosting

By Suresh Ramasubramanian
Suresh Ramasubramanian

This very interesting document was released by ICANN's Generic Names Supporting Organization (GNSO) for public comment yesterday. And it asks some fundamental questions while at the same time pointing to sources such as the Honeynet Alliance's reports on fast flux.

It also points out the benefits of "legitimate" fast flux — such as its use by content distribution networks, or by DDoS protection systems. An additional use is of course a simple attempt at using multiple A records with short (< 1 minute) TTL in a basic attempt to load balance.

It would be interesting to see what registries and registrars can do to suppress malicious fast flux — such as due diligence to prevent fraudulent registration of domains (most if not all malicious fast flux domains are registered using stolen cards, and chargebacks of course hurt registrars far more than the revenue from these, or at least I hope so), and proactive action by registries to block registration of fastflux domains.

A lot of the fast flux domains also — it must be noted — use Whois privacy as a default where it is available (and some registrars have a very bad habit of inserting absolutely fake addresses into the Whois records, for Whois privacy — where others list their own business address and a clear note on the nature of this Whois privacy). Some of that ugly mess of a discussion is quite likely to be relevant here as well.

Questions that get asked in the report — some are quite probably rhetorical, and most of these do have suggested answers in the report — are below:

By Suresh Ramasubramanian, Antispam Operations

Related topics: Cyberattack, Cybercrime, Cybersecurity, DDoS, DNS, Domain Names, ICANN, Malware, Networks, Policy & Regulation, Registry Services, Spam, New TLDs, Whois

Comments

One thought on Fast Flux Richard Golodner  –  Jan 29, 2009 11:40 PM PDT

After reading the GNSO report on Fast Flux one of the problems I can see is that it breaks the ability of a network that uses IP based ACL's to limit access to their online resources. With Fast Flux, the bad guys can now have access to targets that may have been previously blocked by these controls. Granted a company should have many layers of protection such as firewalls, IDS and IPS, but I see a lot of smaller customers, not ISP's, that rely on IP based ACL's as part of their line of defense. With Fast Flux, these companies will now need to step up their game and determine other ways to protect themselves.

With budgets being tightened in all aras of IT, security staff will have to be extra dilligent in reading logs, scanning their nets for unusual activity and monitoring or prohibiting where users are allowed to go on the Internet which should be done anyway, as the money to increase netork protection via hardware may not be there. Explaining this to upper level mangement may be difficult until they are compromised by Fast Flux guided bots.

I can't count the times I have been told that "it cant happen here" or "why would someone try and attack us?" Often times I find the answer to that question is because the bad guys could and did.

Fast Flux does have it's place in the industry and we all take advantage of it. Content delivery networks, computer updates all use it.

My question is what can be done at the domain registry level to make it more difficult to for the bad guys to use Fast Flux as a means of continuing their criminal enteprises?