Adobe Won't Patch Latest PDF Zero-Day Until Jan. 12
Adobe doesn't plan to patch the newest critical vulnerability in its PDF viewing and editing software for another four weeks, even though attack code has been publicly released.
Wed, December 16, 2009
Computerworld — Adobe won't patch the newest critical vulnerability in its PDF viewing and editing software for another four weeks, even though attack code has been publicly released.
In an update yesterday to the security advisory it issued Tuesday, Adobe set the patch date as Jan. 12, 2010, which is also the next regularly-scheduled quarterly security update for Adobe Reader and Adobe Acrobat. Most of the advisory was dedicated to confirming the bug -- which the company had first disclosed late Monday -- and providing instructions for blacklisting the JavaScript API call that contains the flaw.
Other security experts have urged users to disable JavaScript in Reader and Acrobat to protect themselves until Adobe ships a fix.
Adobe's plan to patch next month means that most users will be at risk until then, as researchers have already crafted a working, if not always reliable, exploit for the vulnerability.
HD Moore, the creator of the open-source Metasploit penetration testing framework, and chief security officer for security company Rapid7, said yesterday that he had worked up attack code for the Reader/Acrobat zero-day bug. "We managed to get the exploit sorted out faster than I thought," said Moore in an e-mail to Computerworld yesterday. "It's now available through the online update feature of Metasploit."
He also announced the exploit module's availability via his Twitter feed .
The exploit is not 100% reliable, however. In a reply to a tweet asking why the Metasploit exploit module wasn't working for one user, Moore explained : "tricky bug and some non-english locale's overlap with the heap spray, we should have an update soon."
Unlike other penetration testing frameworks -- such as CANVAS, which is developed by Miami Beach-based Immunity -- the open-source Metasploit doesn't charge for its exploits. That lets hackers, as well as legitimate security researchers, get their hands on working attack code.
The appearance of an exploit on Metasploit increases the chance that widespread attacks will begin, said Joshua Talbot, a security intelligence manager in Symantec's security response team. "An exploit in Metasploit lets hackers build on that code," said Talbot, referring to the leg-up attackers receive from Moore's work.
For his part, Moore defended Metasploit's practice of providing working exploits to anyone. "Since the bug is 1) public and 2) widely exploited, we feel that adding an exploit module is the right thing to do, as it provides a safe way for folks to verify that their mitigation efforts actually work," Moore said in a Tuesday e-mail.
