Facebook says the worm's impact is minimal. "Only a very small percentage of Facebook users have been affected, and we're working quickly to update our security systems to minimize any further impact, including resetting passwords on infected accounts, removing the spam messages, and coordinating with third parties to remove redirects to malicious content elsewhere on the Web," a Facebook spokesperson said in an e-mail. "Users with up-to-date antivirus software are generally well protected from this and similar viruses."
The company has posted information on its security page to help users.
Craig Schmugar, a McAfee Avert Labs researcher, published a warning about the new Koobface variant Wednesday and said that the situation is likely to get worse before it gets better.
"It's important to note that spammed links leading to Koobface are likely to come from infected friends, reminiscent of early mass-mailing worms," he said. "The safe-computing practice created more than 10 years ago still applies today, which is not to open any unexpected e-mail attachments, even if they are from someone you know."
Spammed links generated by Koobface lead to various compromised host sites that appear to serve a video. The user is then presented with a fake error message saying that the version of Adobe Flash installed on his or her computer is out of date. The message prompts the user to download an update.
The update, of course, is malicious software. It can easily by changed by those behind the attacks to exploit any of a variety of security vulnerabilities. At the moment, it installs a proxy server called tinyproxy.exe and a service called Security Accounts Manager that loads the proxy server at startup. According to Schmugar, the server listens on TCP port 9090 and scans all HTTP traffic to intercept search results for the purpose of ad hijacking and click fraud.
In October, Fortinet security researcher Guillaume Lovet published a report indicating that those behind the Koobface worm had taken to hosting their fake video code on compromised shared Google Reader and Google Picasa pages to exploit users' trust in the Google brand.
About two weeks ago, Facebook won $873 million in damages from a spammer in the largest judgment under the 2003 Can-Spam Act to date.
"Everyone who participates constructively in Facebook should feel confident that we are fighting hard to protect you against spam and other online nuisances," said Facebook director of security Max Kelly and deputy general counsel Mark Howitson in a joint statement last month. "We will continue to invest in this area by improving our technical safeguards and devoting significant resources to finding, exposing, and prosecuting the sources of spam attacks." Facebook didn't immediately respond to a query about whether it had collected any of its damage award yet.