Hackers are using a new technique to infect desktop PCs via compromised websites, while avoiding antivirus detectors, according to the SANS Institute.
SANS' ISC (Internet Storm Center) said on Thursday it has come across the attack on a compromised website, where an iframe was used to deploy various pieces of malicious code via Javascript. Iframes allow content from one website to be embedded in another website.
This technique in itself isn't new, but researchers found that the server deploying the malicious Javascript was heavily modifying it - "obfuscating" it - so as to be undetectable by antivirus detectors, the ISC said. Moreover, the obfuscations were generated randomly and on the fly, according to ISC handler Bojan Zdrnja.
"What makes this new is that the hosting website generates this code dynamically," he wrote in an analysis.
"Every time you request this web page it will use completely random names for all variables and functions... changing variable and function names even causes the payload information to change."
The technique makes the script code in effect undetectable by common types of malware scanners, Zdrnja said.
"Such heavy obfuscation makes signature-based detection much more difficult, if not impossible," he wrote. None of the antivirus programs Zdrnja tested were able to detect the modified code.
The code contains what Zdrnja called a "typical" set of exploits, making use of known vulnerabilities in ADODB, QuickTime, WinZip and other software.
The code also included a less well-known, but highly pernicious exploit for the NCTAudioFile2 ActiveX control, Zdrnja said.
"A fully working exploit was publicly released in April, and what's worse is that the affected ActiveX control is delivered with dozens of popular audio/video applications," Zdrnja wrote.
Comments
ParalaMedicalistik said: Ya I agree with laddy dragonthoughts comments It also seem 2 meself that ur seem 2B a bit behind in reporting these issues Not only McAfee had earlier detection but as well Aviras AntiVir Crawlers ClamAV Sophos AV F-Secure amp on amp on amp So get with it thar Matey
CaffeineAddiction said: Yah just obfuscating the variable names and subs wouldnt be enough That would allow it to get past a regular hash check but there are other ways of checking a file If the person who wrote the code left one unique comment or one unique piece of code one that either isnt used that often or at all in any other program it would be able to be found by just that one or two linesObfuscating malicious code is not a new idea and for some stuff it works like a charm but in order for it to be totally immune to anti virus it would almost have to morph at the byte level which is much more complicated
dragonthoughts said: This article claims that current anti-virus programs cannot detect itThis is in conflict with McAfees claim at vilnaicomvilcontentv1419that it could detect this since May 31st 2007 - 2 months before this article was published