Pierluigi Paganini January 28, 2012

In the last decade we have witnessed an extraordinary evolution of those agents technically defined as the term “malware” in several respects as:

1. functional characteristics
2. developing process
3. distribution channels
4. field of use

In the past we used to differenciate the malware in groups whose members were deeply identified by distinct behaviors, hence the origin of terms such as viruses, Trojans and Worms. Today, a clear-cut categorization is almost anachronistic because of the vagueness of the behaviour of each agent and the ability to encompass the main features of the categories listed above.

We are facing a Darwinian outlook of malware evolves parallel to the technology evolution and to the development of defense systems. The medical parallelism is the perfect example, malware such as pathogen agents become increasingly resistant to forms of defense. Similar to what happens in medicine, malware can be combined each other, not for spontaneous action, but with precise intent of the developers.

New malware morphs into different shapes unattended by humans, but how is it possible?
Well consider that a malware is primary a program, an application that could be executed inside the target environments, this means knowing final objective it is possible di write specific code to implements the desidered functionality.
Being that malware is software with a good analysis and development it is possible to make it reusable, modular and scalable.
The concept is easy, common to ordinary software ingineering, a malware intended as basic platform that could be extended using specific modules developed for requested targets. Of course this approach open to unpredictable scenarios, open evironment that help open new forms of cooperative and marketing, as I wrote in my article on the new services made available by the developers of malware.

I have reade of a “practice” that has been observed: the file infector that accidentally parasites another e-threat. A virus infects executable files that could be itself another virus or worm, becuse both are executable file.  The described mechanism could be intentional or not but in both cases the interaction results really dangerous for the victims because in sometimes the attacks methods results combined. Personally I believe that randomness of the event is limited, today who produce a malware knows very well which its final purpose and is able to predict the iteraction of the agent with the target environment and with other virus.
Let’s image a governative agency that develope a new cyber weapon to steal information from an hostile enemy environment, for sure one of the requirement is to avoid that other malware will interact with it, this is possible for example avoiding what in development gerg we call “code injection”.

Many agents have built in spreading mechanisms via common channel as P2P, chat applications or social networks, but they can also benefit of the replicating vector of other malware. With the terms Frankenmalware it is refered the cross infection between malware that take place “spontaneously”, when a virus infects by mistake another piece of agent and ends up using its capabilities to spread.
A good example is the case of the Virtob spreading module that has been observed to infect has been found infecting Bifrose backdoor Trojan and worms like Mydoom.

In summary the combination of agents produces the following side effects:

• malware more effectively by combining features of original components, such as information gathering module and code such as rootkits. Eradicating these threats is undoubtedly more expensive.
• changing of malware signature produced as resultantof processes such as cleaning action by an antivirus, or for the cross injection of pieces of the source code of other infector. The malware will result in unaltered but its signature results changed, but detection processed are based on signature recognition with the result that a malware can not be easly identified.

So far we have dealt with functional aspects related to characteristics, distribution channels and developing process but we have ignored the fields of employment. Over the years, agents such as viruses and trojans have found wide commitment to fraud and intelligence operations, we can say that cybercrime and defense were the main proponents of the stunning achievements.
Malware, by its nature, operates in adverse conditions, or in environments made ​​hostile by the presece of defense systems and therefore it must require a truly effective process of development. I find it extremely fascinating field of research in this area especially for the ongoing commitments that researchers involvement in the search … is a constant struggle and he who hesitates is lost.

Pierluigi Paganini

References

http://www.malwarecity.com/blog/virus-infects-worm-by-mistake-1246.html