 Computerworld is New Zealand's only specialised information systems fortnightly.
Subscribe now for $97.50 (24 issues) and save more than 37% off the cover price!
 Get the latest news from Computerworld delivered via email.
Sign up now
 Subscribe to Computerworld's
RSS newsfeed here and get news stories as they break.
Microsoft has now fixed a widely reported flaw in its Internet Explorer browser that had been used by attackers over the past few weeks to take over the PCs of unsuspecting users.
The flaw was one of four IE bugs fixed this week in Microsoft's regularly scheduled software update, which also addressed some of the problems caused by Sony BMG Music Entertainment's XCP copy protection software.
Although attacks based on the vulnerability have not been widespread, it is important that IE users now install the patch, says Neel Mehta, team lead of Internet Security Systems's X-Force group. "It's not of epic proportions," he says. "But isolated attackers here and there have used it to install malware."
Security experts had known about the flaw since May, but on November 21 hackers with a UK organisation called Computer Terrorism posted sample code that showed it to be much more serious than originally thought. Within days that sample code was adapted and being used by attackers, prompting many security experts to erroneously predict that Microsoft would rush a patch ahead of its December update.
Microsoft has rarely released such out-of-cycle patches since beginning its monthly security update process two years ago, says Stephen Toulouse, security programme manager with the Microsoft Security Response Centre. "We've done it three times and they were, all three, Internet Explorer updates."
Like the rest of the world, Microsoft had not realised that the IE bug could be used to run unauthorised software until the Computer Terrorism code was posted, Toulouse says.
At that point, the company evaluated the potential threat posed by the malware against the possibility of releasing a buggy patch and decided to include fix the problem within an IE update that had already been in the works for December.
"We have to be very, very cautions about the amount of testing coverage that we're getting," Toulouse says. "In this case the right thing for us was to do was to get it through the testing process to make sure ... it was of the high quality we were used to."
The bug concerns the way IE processes the "Window()" function in JavaScript, a popular scripting language used by Web developers to make their sites more dynamic. It affects IE users on Windows XP, Windows 2000 and Windows 98. In order to exploit this problem, attackers must first trick users into visiting a maliciously encoded website, which has helped prevent the bug from being more widely used.
Microsoft fixed this problem, along with the other three IE bugs, in one of two security updates.
With the software release, Microsoft has also taken steps to mitigate the danger associated with Sony's XCP (Extended Copy Protection) digital rights management software. Sony was forced to recall millions of CDs last month following a series of blunders that related to this program. Because of its use of "rootkit" techniques normally associated with spyware or viruses, XCP itself was considered spyware by many security vendors. And Sony's XCP uninstaller installed a buggy ActiveX control that could be used by attackers to take over a user's PC.
As planned, Microsoft has now updated its Malicious Software Removal Tool to disable XCP's rootkit features and identify and remove a malicious Trojan program that takes advantage of the rootkit's cloaking techniques, Toulouse says.
The IE security update also includes software that disables the Sony's buggy ActiveX control. "With Sony's permission, we're preventing that ActiveX control so you can't be attacked," Toulouse says.
A second update, assigned Microsoft's less severe security rating of "important," fixes a problem in the Windows 2000 kernel. This bug could help an attacker to circumvent Microsoft's user privileges mechanism and perform unauthorised tasks on a PC.
Typically, this flaw could not be exploited remotely, as it requires that the attacker gain access to the targeted computer's keyboard, says Steve Manzuik, security product manager with the company that discovered the bug, eEye Digital Security in its advisory.
|
