
Participating domains return a set of 10 IP addresses for each query with a varying DNS cache time-to-live (TTL) of less than 150 seconds, forcing the addresses to be refreshed after no more than two and a half minutes,” explained Crowder and his report co-author Noah Dunker, director of security labs for RiskAnalytics. “Over time, hundreds or thousands of IP addresses are used. This technique is designed to bypass IP address blocking solutions while still maintaining the advantages of a highly-available network.
Source: Fast Flux Taken To The Next Level With Zbot Botnet
I continue to believe that monitoring outbound DNS queries is one of the most important and fruitful techniques a company can employ against malware.
If you have good DNS hygiene that’s precisely what should happen with malware like this (and most other kinds as well).
If you have a properly administered network, your hosts should have your internal DNS servers as the only source of DNS. And if they’re interacting with any other DNS server (or trying to) that should raise immediate interest.
On a tight network, a DNS query that goes anywhere but to your internal DNS server should be treated as a serious problem. It’s either a configuration error, or something trying to call home.
If it’s the former, get it fixed so it’s not messing up your detection capabilities. And if it’s the later, you now know about some potential malware that you didn’t before.
DNS is an endless treasure of security information. Lock it down and harvest it.