bugtraq - Re: IE ActiveX 0day?

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite for Android: free password hash cracker in your pocket

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date: Mon, 18 Sep 2006 09:44:20 -0700
From: Alexander Sotirov <asotirov@...ermina.com>
To: bugtraq@...urityfocus.com
Cc: "Hayes, Bill" <Bill.Hayes@....com>
Subject: Re: IE ActiveX 0day?

Hayes, Bill wrote:
> It looks like the flaw is a buffer overflow and not a memory corruption
> error.

Actually, the vulnerability is an integer overflow which leads to a buffer
overflow which leads to memory corruption.

KeyFrame(npoints, ...)
{
	buf = malloc(npoints*16)

	copy stuff into buf
}

What is your definition of memory corruption? How can a buffer overflow not be a
memory corruption error?

> FrSIRT claims that by sending a specially-crafted argument to the
> DirectAnimation.PathControl" (daxctle.ocx) ActiveX object, a local or
> remote attacker can cause a memory corruption error that leads either to
> a Denial of Service (DoS) condition, execution of arbitrary code. As a
> workaround, FrSIRT is recommending disabling Active Scripting in the
> Internet and Local intranet security zones.  This will obvioulsy break a
> number of pages.

There is a better workaround than disabling ActiveX - set the kill bit on the
DirectAnimation.PathControl CLSID. Nobody uses that control anyway.

> Symantec SecurityResponse blog states that Symantec researchers have
> determined that the flaw in the  DirectAnimation Path ActiveX Control is
> in fact a buffer overflow instead of a memory corruption error. Symantec
> researchers now believe that the buffer overflow occurs "when IE tries
> to instantiate a certain DirectionAnimation COM object as an ActiveX
> control."  The blog note says that remote execution of arbitrary code is
> possible.

This flaw is not an object instantiation bug. It happens when you call the
KeyFrame method of the object with a large argument.

Alex