For some hacks, everything old is new again
Hewlett-Packard has released its report on top security risks
for 2011, containing a wealth of statistical data about the threats facing us last year. What struck me from the report was the persistence of vulnerabilities in software — both in newly developed applications and in deployed applications that have not been patched — and how successfully they are being exploited.
The rate of disclosure for new vulnerabilities has been dropping off in recent years, and it is not hard to see why.
“People aren’t using the new vulnerabilities nearly as much as the old ones,” said Jason Jones, advanced security intelligence engineer for HP Digital Vaccine Labs. “The older ones are still working.”
:Anonymous hack of intell company Stratfor was all too easyGetting the most out of automated IT security management
That does not mean that new applications aren’t also vulnerable. Static code analysis of 359 Web applications found that most contained flaws, making them susceptible to common techniques such as cross-site scripting and injection. Insecure cryptographic storage, information leakage and improper error handling were very common.
The bottom line seems to be that speed to market still trumps security in software development. In the competitive market for Web applications, developers would rather be first than be secure, and quality assurance suffers. These problems are compounded by inadequate patching, which often gives vulnerabilities an unnecessarily long shelf life on our systems.
According to the report, new vulnerabilities disclosed in the Open Source Vulnerability Database peaked in 2006 at around 11,000, tapering off since then to about 6,600 in 2011. But “this decline in disclosure of new vulnerabilities doesn’t mean that vulnerabilities don’t exist or that software is suddenly secure,” the report warns.
Finding vulnerabilities often is more difficult in today’s more complex software, and underground market demands have put a premium on high-risk vulnerabilities. “Overall, these sets of data show that coding mistakes that lead to security attacks are widespread,” the report states.
These new vulnerabilities can be saved for high-value targets, because the old vulnerabilities continue to work well. One of the most popular exploit kits being used by criminals is Blackhole, which is hardly cutting edge, the report states. “The surprising fact about Blackhole’s success is that despite using known, patched bugs from 2010 and prior, it still achieved infection rates comparable to or better than other exploit kits.”
The end user is a part of the infection problem, of course. They too often invite malicious code onto the system. Symantec recently reported
on Japanese users being targeted by a group of 29 malicious mobile apps that steal information from host devices. Upon download the apps ask not only “Do you want to install this application?” but also ask specifically for full Internet access, access to all contact information and to a phone and identity information.
“The PC has trained us to click through everything,” said Joe Chen, director of engineering for Symantec’s security technology and response organization. Users see pop-up warnings and requests for authorization as barriers to functionality rather than useful security tools.
The obvious lesson here is that we already know how to improve our security. This will not necessarily be easy to do. Criminals have become adept at obfuscating and delivering exploits for known vulnerabilities, so that blocking attacks at the perimeter are not likely to be completely effective. We have to look deeper inside our systems to fix the flaws and watch for malicious activity.
But we should know what to look for. Patching and configuration management are still important, as is quality assurance in software development.
Of course, if we should succeed in eliminating the current generation of known vulnerabilities, a new generation will come along to replace them. But at least we can level the playing field by making the hackers work as hard to hack us as we are in defending ourselves.