Slashdot Mirror


Highly Critical Hole Found in IE

dotpavan writes "Eweek reports on a highly critical MS Internet Explorer hole found by Secunia Research's Andreas Sandblad. The vulnerability is due to the processing of the "createTextRange()" method call applied on a radio button control. From Secunia, "The vulnerability has been confirmed on a fully patched system with Internet Explorer 6.0 and Microsoft Windows XP SP2." The vulnerability has also been confirmed in Internet Explorer 7 Beta 2 Preview (January edition) though it could be avoided by turning off Active Scripting, as suggested by Microsoft Security Response Center blog. How would this put MS in the market, hit by the ever-growing shots of vulnerabilties? And would the divorce of IE7 from Vista's Windows Explorer help?"

6 of 336 comments (clear)

  1. Yet further evidence... by Anonymous Coward · · Score: 1, Interesting

    Yet further evidence that IE7 and also likely Vista and all other 'new and improved' products rolling out of Microsoft will be nothing more than business as usual.

  2. Re:Why are IE security flaws even reported anymore by caffeination · · Score: 2, Interesting

    Not quite true. Mostly because of the sheer amount of lazy bastards reading Slashdot while they should be working, a high proportion of this site's visits are through Internet Explorer. Even if they will use some newfangled firebird or netcraft when they get home, this hole matters to them *now*.

  3. Re:It's funny by Zocalo · · Score: 2, Interesting

    Also, I note that there is no mention as yet (there is another story on the way) of the highly critical security flaw found in Sendmail which also had a proven potential for remote and local exploitation and arbitrary command execution. Actually this is potentially quite interesting; with remotely exploitable problems with both IE and Sendmail announced at almost the same time, I wonder which one we are going to see exploited by the blackhats first? Admittedly there are already updated packages for most Linux distros and commerical UNIX versions, plus a new release of the software (no offical Sun patch for Solaris yet though) which is going to tip the results a little, but still...

    --
    UNIX? They're not even circumcised! Savages!
  4. Easy formula by EraserMouseMan · · Score: 2, Interesting

    A simple math analogy will demonstrate the formula for /. sentiment. A negative multiplied by a negative equals a positive. Hackers hacking Microsoft == good news. Hackers hacking Firefox == bad news. Any good tech company can easily turn evil simply by an association with Microsoft.
    GoDaddy == Good.
    GoDaddy * Microsoft == Evil

    In the same vein (but totally against any mathematical logic), any company (including evil ones) that are associated with Open Source and/or Linux automatically become good.

    Oracle == Evil
    Oracle * Linux == Good
    China == Evil
    China * OSS == Good

  5. Repeating themes on slashdot by amightywind · · Score: 1, Interesting

    Why do stories like this even make it to Slashdot anymore?

    Why do they mod you flaimbait? This is a good question.

    1. Microsoft security problems are one of a handful of topics that appeal to the slashdot priesthood. It is really quite an ecclectic group: global warming, crank science, amateur space programs, criticism of the Bush administration... These are confortable subjects that reaffirm their views.
    2. This forum mainly came about to resist Microsoft and promote free software. It is interesting that over time slashdot has come to promote open source and ridicule free software.
    3. Like freeway chases, Microsoft security problems are entertaining. Most will never be effectively exploited, but there is always a chance that it will be 'the big one'.
    --
    an ill wind that blows no good
  6. Re:IE 7 in Vista would have been safe by kbielefe · · Score: 2, Interesting
    the security model in NT-based systems is much richer than that in Linux-based systems

    I beg to differ, unless you qualify that with default. Even then, there is little difference in capability in actual practice, as you pointed out. The security model in Linux has almost always been as rich as you want it to be. Process and role based access control has been available and used in Linux for several years in systems where that level of control is desirable, and has even crept into default installations of some server and even desktop distros in the last couple of years.

    For example, all the applications that connect to the internet on my home desktop already have similar restrictions to the IE7 restrictions the grandparent pointed out, and are probably more configurable and transparent. There are also several other layers of security that will probably prevent an attacker from ever getting to that point. Now you can say you've heard of an "ordinary" user process switching to an even less privileged user account.

    Admittedly, it wasn't easy to set up, but it is very easy to use and maintain. When I first made the changes, my wife didn't even notice a difference, and she couldn't see what the big deal was. I'll be very interested to see if Microsoft can manage to make it effective, easy enough for the average joe to install, and transparent enough that the average joe won't get annoyed and turn it off. I don't see how they can do it without limiting the extra security features to their own products in very inflexible configurations.

    --
    This space intentionally left blank.