Internet users have been hit by a widespread Web attack that has compromised thousands of Web sites, including Web pages belonging to the Wall Street Journal and the Jerusalem Post.
Estimates of the total number of compromised Web sites vary between 7,000 and 114,000, according to security experts. Other compromised sites include Servicewomen.org and Intljobs.org.
Cisco Systems’ Web-tracking subsidiary, ScanSafe, started following the incident two days ago, said Mary Landesman, a senior security researcher with Cisco.
Related Story: How to not have your Web site hacked like Sony’s
This server tries to install software on Web visitors’ computers. If it’s successful, the software gives the criminals a way to remotely control their victims’ PCs.
Security researchers are still gathering data on the attacks, but suspect hackers used what’s known as an SQL injection attack to trick the Web sites into running database commands, which ultimately gave the hackers a way of installing their malicious HTML.
All of the infected sites appear to be using the Microsoft Internet Information Services Web-server software running with Active Server Pages, according to researchers at Sucuri Security.
Sucuri put the number at 114,000.
Not all affected sites are thoroughly compromised, Landesman added. On some very large Web sites, attackers were able to install their code only on certain pages, rather than the entire site.
Often, hackers get limited access because they break into a partner site — an ad company, for example — that is allowed to post on certain parts of the larger company’s Web site.
On the Wall Street Journal, for example, only a small number of pages that displayed real estate ads were hit.
The Wall Street Journal had no immediate comment on the incident, but a spokeswoman said she was looking into the matter.
Antivirus vendor Sophos said it had spotted the attack on the Jerusalem Post’s Web site.
HP and Microsoft have released a free tool called Scrawlr that helps users check their Web sites for SQL injection vulnerabilities.
“The SQL injection attacks that allow the systems to be compromised are occurring due to vulnerabilities in third-party web applications and do not demonstrate vulnerabilities in Microsoft software,” said Microsoft spokesman Jerry Bryant via e-mail.
“We do offer guidance for developers on how to code applications so they are protected against SQL injection,” he added.
Although these mass Web attacks have become relatively common in the past three years, this incident appears to be the worst since a large number of WordPress-based sites were hacked in April, said
Andre DeMino, a co-founder of the Shadowserver malware-tracking group.
In that incident, Network Solutions LLC was confronted with a large-scale infection of WordPress-driven blogs. The company acknowledged that other sites it hosts had been compromised.
“We have received reports that Network Solutions customers are seeing malicious code added to their Web sites, and we are really sorry for this experience,” said company spokesman Shashi Bellamkonda in a blog post. “At this time, since anything we say in public may help the perpetrators, we are unable to provide details.”
The next day after the post, another Network Solutions spokesperson declined to get more specific or answer questions, including queries about what moves the company was making and how many sites had been affected.
“At this time, we believe this is affecting a subset of our hosting customers,” said Susan Wade, director of corporate communications at Network Solutions. “For now, it’s difficult to make a conclusive statement or provide more details publicly.”
The same server was involved in the earlier attacks against Network Solutions-hosted blogs.
According to the StopMalvertising blog, the attacks planted a rogue IFRAME on the hacked sites to shunt users to the attack server. That server then launches multiple exploits, including an attack kit of ActiveX exploits and three more that take advantage of Adobe Reader vulnerabilities, against visiting PCs.
Several browsers, including Microsoft’s Internet Explorer 8, Google’s Chrome and Mozilla’s Firefox, displayed warnings when users were redirected to the attack site.