As confusion continues over who is responsible for a zero-day Windows vulnerability -- Microsoft or Mozilla -- the former said it sees no need to patch Internet Explorer (IE) while the latter promised to fix Firefox, even though it blames its rival for the problem.
On Tuesday, researchers started arguing over a bug that allows attacks against IE users -- but only those who also have Firefox installed on their PCs. Thor Larholm pinned the blame on IE, and said that while Firefox registers the FirefoxURL protocol used in the proof-of-concept exploits, Mozilla's browser was an innocent bystander.
"There is an input validation flaw in Internet Explorer," said Larholm. Specifically, he said that IE fails to escape quotation marks, as well as other characters, such as commas.
"Internet Explorer is to blame for not escaping 'quote' characters when passing on the input to the command line," Larholm said later Tuesday. "I agree that Firefox could have registered its URL handler with pure DDE instead and thereby have avoided the possibility of a command line argument injection, but IE should still be able to safely launch external applications."
Other security experts, including Thomas Kristensen, chief technology officer at Danish vulnerability tracker Secunia, said otherwise. "This is in fact not an IE issue, it is a Firefox issue," Kristensen claimed in an e-mail to Computerworld.
Things didn't become any clearer yesterday as Mozilla's chief security officer, Window Snyder, promised a fix for Firefox, but said it is patching only to protect its customers, not because it sees Firefox as the culprit. "This will prevent IE from sending Firefox malicious data," Snyder said in a posting on Mozilla's security blog. "Other Windows programs may also be vulnerable to bad data being passed from IE, although we are not aware of any at this time."
Although Snyder never directly pinned responsibility on IE, she came close by describing the problem. "Any Windows application that calls a registered URL protocol without escaping quotes may be used to pass unexpected and potentially dangerous data to the application that registers that URL protocol," she said. "This could result in a critical security vulnerability."
For its part, Microsoft issued a one-sentence statement through a spokesman. "Microsoft has thoroughly investigated the claim of a vulnerability in Internet Explorer and found that this is not a vulnerability in a Microsoft product," read the statement.
Other researchers waded into the brouhaha. Today, Jesper Johansson -- who, like Snyder, once worked for Microsoft but is now a security program manager at Seattle-based Amazon.com -- again argued that IE is not to blame. "It is quite clear really: IE does not validate the URL string, nor does it ever make any promise to do so [emphasis in original]."
Instead, he said, the flaw lies with Firefox. "It is clear from [Windows'] documentation that it is incumbent upon the application to validate the URL string," he said on his blog. "If the application can accept, and process, dangerous commands through its protocol handler, as Firefox does, it is even more critical that the application take care to validate the URL before processing it."
Some see both sides, and put the onus on both parties. "I think it's a shared issue, really," said Roger Thompson, CTO of Exploit Prevention Labs. "If you go to an exploit site directly with Firefox, it doesn't work; only if you go to the site with IE. That's why I think IE is at least part of the issue."
Even more confusing is that while Mozilla plans to patch Firefox -- tentatively in a 188.8.131.52 release expected to be out at the end of this month -- Firefox users are not at risk. Snyder made a point of stressing that only Windows users who have both browsers installed, and are using IE if (or when) attacked, are in danger.
That diminishes the threat, said Thompson. "[The entire argument] is out of hand," he said, "until someone starts using [the exploit]. There's a huge difference between what might happen and what is happening."
While proof-of-concept code has been published -- some researchers have claimed that they've not been able to produce an exploit using Larholm's code -- there are no reports of any in-the-wild use. And even if there is, the danger may be minor. "If anyone is geeky enough to use Firefox, it's probably their main browser, so they're unlikely to visit the site in IE," said Thompson.