Security watchers are bracing themselves to respond to the activitation of the huge botnet created by the Conficker superworm.
The malware has created a network of infected PCs under its control estimated at 9m or even more, according to the latest estimates — dwarfing the zombie army created by the infamous Storm worm, which reached a comparatively paltry 1m at its peak in September 2007.
Variants of Conficker (aka Downadup), which began circulating in late November, exploit the MS08-067 vulnerability in the Microsoft Windows server service addressed by Redmond with an out-of-sequence patch last October.
The malware also infects removable devices and network shares using a special autorun file. The worm uses social engineering trickery so that users on Windows machines looking to simply browse the contents of a memory stick may be tricked into selecting an option that actually runs a malware payload and infects their PC.
Some variants are programmed to spread across machines in the same local area network. Weak passwords in corporates have therefore aided the distribution of the worm.
The multiple infections techniques - none of which, incidentally, feature email — has fuelled the prolific spread of the worm. It’s been years since any worm has spread so widely. In many ways the Conficker worm epidemic represents a return to the bad old days of worms such as Nimda, Blaster and Sasser.
It only takes one rotten apple
In the case of Conficker, security watchers reckon the fact that the worm only needs to hit one infected machine in a network to spread goes a long way towards explaining its success. Slow patching, particularly in corporates, has also contributed to the epidemic.
“We haven’t seen this type of advanced worm in many years,” Eric Schultze, CTO of patching firm Shavlik Technologies told El Reg. “It’s successful because once a single machine is infected in a corporate environment, it can spread itself to all of the other corporate machines, whether they’ve been patched or not.
“In terms of damage it can do, some reports say the worm is a dud but I believe that it’s simply ‘sleeping’ and may be woken up at a future date to execute some set of evil instructions. Even if never executed, the worm turns off the windows update service and blocks access to many security vendor websites [blocking uptake of new antivirus signatures].
“To many, these actions alone may be considered malicious.”
Net security firm Sophos reckons that business users have been harder hit than consumers by the spread of the worm. The malware has caught some firms on the hop because they haven’t rolled out patches, it figures.
Theories on why we haven’t seen a worm of this type for three or four years are thin on the ground. It may be that writing such a worm (even if it pinches parts of its code from Metasploit, the open-source penetration testing tool) is simply too much like hard work.
“It’s more effort to write malware that exploits a new vulnerability than, say, regular executable malware that is emailed or shoved on web,” said Graham Cluley, senior technology consultant at anti-virus firm Sophos. “If email or web attacks work just fine, then why go to extra effort?”
“These guys aren’t doing it for intellectual challenge or showing-off. Money is the motive.”
Despite the noteable lack of network worms over recent years the approach — much like spreading computer viruses using infected email attachments — has always been an option for miscreants.
“Hackers never completely abandon old tricks,” Cluley continued. “They can always dust them off and use them again. For example, there was a huge increase is infected email attachments last year year. It’s a danger to think we have any particular attack strategy licked.”
Cluley, like other security researchers, credited Microsoft for releasing a clean-up tool in January after publishing a patch in October, while noting the software giant bears significant responsibility for creating the security vulnerability that allowed the worm to spread in the first place.