Recent publicity for online hacking groups such as Anonymous and Lulzsec has seemed to show that nobody is immune from attack on the Internet. Once targeted, it seems that these groups are capable of breaching security systems and retrieving data, including identity information, from the most secure systems. However examining the true nature of these attacks shows a different story.
Many of these attacks rely on outdated systems, old vulnerabilities and failure to use best practices. As an example, Anonymous' penetration of HBGary Federal's systems in January this year started with a simple SQL injection. These types of attacks have been known for well over 15 years and most modern web frameworks protect against these attacks by default! Other attacks, such as Lulzsec's attack on Sony, were simple DDOS attacks.
Protection against these attacks should be standard and part of most security strategies. If that is the case, then how did these attacks work?
When examining the targets, it is unlikely that Lulzsec chose to DDOS online games such as Minecraft and League of Legends while ignoring larger and more prolific targets such as World of Warcraft and Second Life. Instead, what we see is the result of a breadth-first attack, where simple attacks are tried against a large number of targets and only the successes are publicised. This contrasts to a depth first attack, where a target is chosen and increasingly complex hacking attempts are made until the system has been penetrated.
A breadth first attack can also be seen in password brute force attacks- instead of trying thousands of passwords to login as one user, one password is used to try login as thousands of users. Such methods overcome the standard "three attempts before locking a user out" defences and instead require more global analysis of login patterns. The same is seen on a global scale with hacking attempts - the weakest link in the chain will be discovered and will be compromised.
About the Author
Robert Layton is a member of the Internet Commerce Security Laboratory (ICSL) at the University of Ballarat. Robert completed his Bachelor of Computing with first class honours before moving into a PhD in cybercrime attribution. He has submitted his PhD, which developed techniques for automatic and unsupervised authorship analysis. These methods are useful for clustering documents by authorship, leading to attribution.