Where Do Attackers Have Most of Their Fun?
Hacking into a system and environment is almost always carried out by exploiting vulnerabilities in software. Only recently has the light started to shine on the root of the problem of successful attacks and exploits, which is flaws within software code. Most attack methods described in this book can be carried out because of errors in the software.
It is not fair to put all of the blame on the programmers, because they have done exactly what their employers and market have asked them to: quickly build applications with tremendous functionality. Only over the last few years has the market started screaming for functionality and security, and the vendors and programmers are scrambling to meet these new requirements and still stay profitable.
Security Does Not Like Complexity
Software, in general, is very complicated, and the more functionality that we try to shove into applications and operating systems, the more complex software will become. The more complex software gets, the harder it is to predict properly how it will react in all possible scenarios, which makes it much harder to secure.
Today’s operating systems and applications are increasing in lines of code (LOC). Windows operating systems have approximately 40 million LOC. Unix and Linux operating systems have much less, usually around 2 million LOC. A common estimate used in the industry is that there are between 5–50 bugs per 1,000 lines of code. So a middle of the road estimate would be that Windows 7 has approximately 1,200,000 bugs. (Not a statement of fact; just a guesstimation.)
It is difficult enough to try to logically understand and secure 40 million LOC, but the complexity does not stop there. The programming industry has evolved from traditional programming languages to object-oriented languages, which allow for a modular approach to developing software. This approach has a lot of benefits: reusable components, faster to market times, decrease in programming time, and easier ways to troubleshoot and update individual modules within the software. But applications and operating systems use each other’s components, users download different types of mobile code to extend functionality, DLLs are installed and shared, and instead of application-to-operating system communication, today many applications communicate directly with each other. The operating system cannot control this type of information flow and provide protection against possible compromises.
If we peek under the covers even further, we see that thousands of protocols are integrated into the different operating system protocol stacks, which allows for distributed computing. The operating systems and applications must rely on these protocols for transmission to another system or application, even if the protocols contain their own inherent security flaws. Device drivers are developed by different vendors and installed in the operating system. Many times these drivers are not well developed and can negatively affect the stability of an operating system. And to get even closer to the hardware level, injection of malicious code into firmware is an up-and-coming attack avenue.
So is it all doom and gloom? Yep, for now. Until we understand that a majority of the successful attacks are carried out because software vendors do not integrate security into the design and specification phases, our programmers have not been properly taught how to code securely, vendors are not being held liable for faulty code, and consumers are not willing to pay more for properly developed and tested code, our staggering hacking and company compromise statistics will only increase.
