Merely storing -- without opening -- a malicious PDF file can trigger an attack that exploits the new, unpatched zero-day flaw in Adobe Reader, a researcher has discovered. Didier Stevens, a researcher and IT security consultant with Contrast Europe NV, today released a proof-of-concept demonstration that shows how a file infected with the Adobe flaw can trigger a new attack when the machine uses Windows Indexing Services. And the user doesn't even have to open or select the document.
In addition, Stevens last week released a proof-of-concept demonstrating how PDF files could be exploited with minimal user interaction -- just saving it to the hard drive and viewing it in Windows Explorer.
But this latest attack vector is more risky, he says, because the user doesn't have to do anything with the file at all. "It requires no user interaction, and for the Windows Indexing Service, it can lead to total system compromise [with] privilege escalation," Stevens says.
Windows Indexing Service is an operating system-level feature that provides an index of files on the system.
Adobe first reported the buffer overflow flaw in Adobe Reader and Acrobat on Feb. 20; initially, security experts advised disabling JavaScript in order to defend against any attacks that exploited the vulnerability. But after further research on the vulnerability, it turns out disabling JavaScript isn't always enough. "The vulnerability is not in the scripting engine, and, therefore, disabling JavaScript does not eliminate all risk," Adobe said in a blog post.
Adobe plans to release updates to address the vulnerability starting this week: By March 11, it will update Adobe Reader 9 and Acrobat 9, and by March 18, Adobe Reader 7 and 8 and Acrobat 7 and 8.
Stevens' latest attack demonstrates how an infected PDF sitting on a Windows XP SP2 machine -- with Windows Indexing Services and Adobe Acrobat Reader 9 running -- gets "indexed." Then the malware is executed, allowing for a privilege escalation attack. Stevens recommends disabling the indexing services' ability to index PDFs.
No comments:
Post a Comment