Dark Reading is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Partner Perspectives  Connecting marketers to our tech communities.
SPONSORED BY
12/21/2016
11:00 AM
Malwarebytes Labs
Malwarebytes Labs
Partner Perspectives
50%
50%

Explained: Domain-Generating Algorithms

Cybercriminals use domain-generating algorithms to prevent their servers from being blacklisted or taken down.

A domain-generating algorithm (DGA) is a program or subroutine that provides malware with new domains on demand or on the fly. Kraken was the first malware family to use a DGA (in 2008) that we could find. Later that year, Conficker made DGA a lot more famous.

The DGA technique is in use because malware that depends on a fixed domain or IP address is quickly blocked, which then hinders operations. So rather than bringing out a new version of the malware or setting everything up again at a new server, the malware switches to a new domain at regular intervals.

An example of DGA in practice is C&C servers for botnets and ransomware. If we were able to block these or take them down, we would cut the link between the victims and the threat actor. Bots would no longer be able to fetch new instructions, and machines infected with ransomware would be unable to request encryption keys and send user data.

The constant changing of the domain for the C&C server is also sometimes called “domain fluxing” or “fast fluxing,” which actually is a reference to an older technique based on abusing the DNS load balancing system.

How It Works

To better understand how these algorithms work, let’s look at the requirements they have to fulfill:

  • The routines have to generate domains that are predictable to both sides of the communication chain.
  • The routines have to be as unpredictable for security researchers as possible.
  • The domain registration fee has to be low, given the huge amounts of domains that will be used.
  • The need for speed can be enormous.
  • The registration process has to be anonymous or at least untraceable.

To achieve predictability, yet remain hard to research, DGA routines use a few building blocks:

  • The seed (base element)
  • An element that changes with time
  • Top level domains (TLDs)

 

Image courtesy of Cisco Blog

The seed can be a phrase or a number -- practically anything that the threat actor can change at will and that can be used in an algorithm. The seed and the time-based element are combined in an algorithm to create the domain name, and this “body” will be combined with one of the available TLDs.

Note that a time-based element need not be the date and time. It can be something else that varies with time -- for example, the trending topic on Twitter in a certain country at the moment of the connection. Actually, something that is difficult to predict is preferred, as this makes it harder for researchers to register certain domains ahead of time and intercept traffic or do a takeover.

Another trick to throw off countermeasures is to not use all the domains that the algorithm produces, but only certain ones. This will drastically increase the number of domains necessary to register by researchers if they plan to intercept the traffic.

When it comes to TLDs, .xyz.top, and .bid are popular at the moment. This is due to the low costs and quick availability because the registrars allow automated and anonymous domain registrations.

Cybercriminals use domain-generating algorithms to prevent their servers from being blacklisted or taken down. The algorithms produce random-looking domain names. The idea is that two machines using the same algorithm will contact the same domain at a given time, so they will be able to exchange information or fetch instructions.

Comment  | 
Print  | 
More Insights
Comments
Newest First  |  Oldest First  |  Threaded View
AI Is Everywhere, but Don't Ignore the Basics
Howie Xu, Vice President of AI and Machine Learning at Zscaler,  9/10/2019
Fed Kaspersky Ban Made Permanent by New Rules
Dark Reading Staff 9/11/2019
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
Write a Caption, Win a Starbucks Card! Click Here
Latest Comment: This comment is waiting for review by our moderators.
Current Issue
7 Threats & Disruptive Forces Changing the Face of Cybersecurity
This Dark Reading Tech Digest gives an in-depth look at the biggest emerging threats and disruptive forces that are changing the face of cybersecurity today.
Flash Poll
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2019-4147
PUBLISHED: 2019-09-16
IBM Sterling File Gateway 2.2.0.0 through 6.0.1.0 is vulnerable to SQL injection. A remote attacker could send specially-crafted SQL statements, which could allow the attacker to view, add, modify or delete information in the back-end database. IBM X-Force ID: 158413.
CVE-2019-5481
PUBLISHED: 2019-09-16
Double-free vulnerability in the FTP-kerberos code in cURL 7.52.0 to 7.65.3.
CVE-2019-5482
PUBLISHED: 2019-09-16
Heap buffer overflow in the TFTP protocol handler in cURL 7.19.4 to 7.65.3.
CVE-2019-15741
PUBLISHED: 2019-09-16
An issue was discovered in GitLab Omnibus 7.4 through 12.2.1. An unsafe interaction with logrotate could result in a privilege escalation
CVE-2019-16370
PUBLISHED: 2019-09-16
The PGP signing plugin in Gradle before 6.0 relies on the SHA-1 algorithm, which might allow an attacker to replace an artifact with a different one that has the same SHA-1 message digest, a related issue to CVE-2005-4900.