Microsoft Admits XP Has Same Bug As Win2K

Arashtamere sends in a Computerworld story on a security flaw in the Windows 2000 pseudo-random number generator published by Israeli researchers earlier this month. Microsoft has now admitted that the flaw is present in XP too. Microsoft denies that the bug is a security vulnerability, since an attacker would have to have gained administrative access to a system before exploiting it. (The Israeli researchers point out that many common exploits provide admin access.) This stance apparently lets them off the hook for patching Win2K, which is in "extended support" mode, though it powers about 9% of US and EU business computers. Microsoft said that XP SP3, due in the first half of next year, will fix the bug. The company said that Vista, Windows Server 2003 SP2, and the new Windows Server 2008 are not vulnerable.

stupid

[Anonymous Coward]

if you already have admin access via another "exploit" why would you bother attacking via random number generator, seems like its a lot of fuss over nothing, Windows has alwayss been vunerable locally (luckily for admins whose users forget passwords etc) so the most worry is over a remote exploit which this flaw isnt. But iam sure some million dollar company will sell a solution for this, paranoia is a great sales tool in the murky world of snake oil, cough i mean computer security

I have to agree with MS on this one...

[Blakey Rat]

If you have admin access, the battle's already lost. What's the point of running a complex process to obtain their password when you have full access to everything on their computer? Might as well just drop in a keylogger and get the same info much easier.

THe paper refered to.

[leuk_he]

This article refers to this summary [buslab.org] of this paper [iacr.org]

I fail to see why you would need administative privelidges however. You would only need to run in the userspace of the process that did run the random number generator before. Having administrative privs would be nice to inject code into that userspace, but is not needed i think.

It can get even worse if from a public key part the random number that was used to generate it can be extracted, what was done in early ssl implementation attacks.

Re:I have to agree with MS on this one...

[xaoslaad]

Granted, I agree with this for the most part. However, it always seems like there is that one person that looks at a problem like this in a way that no one else had prior and manages something completely expected. It's only at the point that a virus is running amok across half the corporate networks in the world that we find out you did not really need administrative priveleges if you did x, y, z first...

History is full of examples, probably both within and out of the computing field where people thought that 'that' was impossible...

Re:I Post Anonymously

[heffrey]

Why is this flamebait? Surely the original post and the pathetic summary was flamebait?

As lots of people have commented, if you have admin rights you own the box.

Re:I have to agree with MS on this one...

[John Betonschaar]

If you have admin access, the battle's already lost. What's the point of running a complex process to obtain their password when you have full access to everything on their computer? Might as well just drop in a keylogger and get the same info much easier.

Most of the other ways to get to the passwords would leave a detectable trace, especially keyloggers. Or they need a reboot. If you're really after the user passwords, resetting them to something else is also not an option. AFAIK there is no other *easy* way to get a user's password from a locally exploitable Windows box, especially not if you cannot reboot it without being detected.

So in some cases, where a hacker with local access to a Windows box wants to have a user password without leaving a trace, an attack like this would be interesting.

I admit It's all a bit hypothetical... Still, it's not very nice to have a possible security hole like this and not patching it.

Naw. You just have to take a different approach.

[khasim]

Microsoft claims this is not a "security vulnerability" because the machine has to already have been cracked to exploit it.

That is not 100% correct.

It is still a "security vulnerability".

It just cannot be exploited to increase your access on that machine.

That we know of. Today. So the code still needs to be patched. Security is not an "either / or" situation. You have to reduce the effectiveness of threats.

Open crypto algorithms; no fix for Win2K

[compumike]

While in general I think open-source and closed-source software can coexist, I think this is a pretty good example of why anything related to crypto should be open. All of public key cryptography relies on the secrecy of private keys, not on the secrecy of the algorithm itself. And while they might have faithfully implemented the algorithm, who knows what kinds of arguments/whatever to the crypto functions might cause undesired results -- it's just too hard to test.

In any case, the thing that surprised me most from the article was that Windows 2000 users would be left out in the cold: "Because the company has determined that the PRNG problem is not a security vulnerability, it is unlikely to provide a patch [for Win2K]." Wow. Especially when it's something this easy to fix. This bug also solves any attacker's problem of trying to sort valuable from non-valuable information, since presumably any valuable information (credit cards used online, etc) will use encryption. And while someone suggested that a program should use its own random number generator, there is a problem because, in general, your application (not running as Admin) shouldn't have access to nearly the same amount of entropy sources (like network activity, GUI inputs, etc).

--
Educational microcontroller kits for the digital generation -- great gift! [nerdkits.com]

Re:I Post Anonymously

[cloakable]

It's flamebait because the GP didn't have to call people retarded, in order to get his or her point across.

They also could have worded this a lot more diplomatically than they did. So yes, the GP is flamebait.

Re:I have to agree with MS on this one...

[Anonymous Coward]

You can possibly compromise keys which were generated before you had admin access and you can comprimise keys which are created in the future without keeping a rootkit installed which might be detected and traced back to you.

Re:Open crypto algorithms; no fix for Win2K

[Anonymous Coward]

Windows supports thousands of different hardware configurations, in hundreds of different languages.

OTOH, Linux, OpenSSH and OpenSSL only run in Intel 80386 IBM branded hardware in Medieval English, so providing a more secure implementation is waaaaay simpler.

Re:I have to agree with MS on this one...

[Terrasque]

This is how I read it :

"At the moment we know of no way to abuse this bug without already having obtained Administrative access."

I will almost bet money that there is a smart bugger out there which find a way to abuse this.
That we don't know of a fearsible attack right now is no excuse not to fix the bug IMHO.

Re:I have to agree with MS on this one...

[mosch]

If you truly agree with MSFT, then you should quit working in computers right now, for everybody's sake.

Many corporate computers have local admin accounts that are likely to share a user/password combo across large numbers of machines. A keylogger might not get you these credentials, but the ability to crack these credentials could get you admin access to a huge number of other computers.

It is people like you who make sure that security consultants will never want for work.

Re:I have to agree with MS on this one...

[webmaster404]

The fact though still remains that Windows is a proprietary, closed-source operating system. If it was open-source much like Linux or BSD, the bug would have been fixed sooner and you could patch your own system, if MS doesn't see it is a security threat it won't get patched. Also, who is to say that crackers haven't found the bug out earlier? If it was Linux, the potential would be very minor for widespread devastation due to differing kernel versions and different patches for different kernel versions. About the only way for a sure-fire attack on the Linux kernel is to attack a distro without any patches but even the most popular distro still has 3 versions still receving support (7.10, 7.04 and LTS) (Ubuntu) and that would make an attack very hard if only 2 of the 3 had it and a patch was released quickly. Its the danger of a propriatary operating system, you never know who knows what and even if you will receive a patch, Linux you can audit the code yourself and rely on the community if you so choose.

Re:stupid

[smallfries]

It does if the data is accessed during the period that you have admin access. The process using the data has to manipulate the keys at some point, and if you can access their memory space then any security is toast. This is exactly how the drm on the new drm for blueray / hddvd was cracked.

This was the point of palladium, that the keys would be locked up inside a separate box, segregated from the processor. Each process would only manipulate opaque handles to the keys.

One nice aspect of this attack is that if you gain admin access after key generation, but before the entropy pool is refreshed then you can play back the state of the random number generator to recreate the keys after the fact. But this just extends the window slightly, you still need an exploit to get admin first.

This is Why Open Source is Good.

[Stephen Samuel]

If this bug was in RedHat 5.2, there would be no issue about getting this critical bug fixed. If nothing else, I could just fix it myself -- and put the necessary patches to the source packages on my website.

No worries about whether or not it's even legal to fix a machine that I'm using to run my business.

Re:THe paper refered to.

[MoogMan]

As the winsock TCP/IP stack randomises it's TCP sequence numbers, I would suggest that it's very likely that it uses a PRNG output directly, and therefore is at risk of being spoofable.

Theoretically, one would need knowledge of just one TCP sequence number, and then it could generate the future sequence numbers coming out of the box. Therefore one would be able to hijack TCP/IP sessions *much* faster and easier than before.

Anyone know to the contrary?

Re:I have to agree with MS on this one...

[RightSaidFred99]

Yeah, because Microsoft doesn't know what they're talking about. This is a PRNG flaw, it doesn't help you "get credentials" in terms of getting Windows logins/passwords. For Christ's sake. Once you have access to the machine, you can theoretically access any encrypted data on the machine because you can get the session keys for e.g. SSL sessions. But, of course, since you already have admin access you could do this any of various other ways anyway.

You use the same password on other machines

[goombah99]

Knowing someone's password can be handy. Most folks use the same password on multiple machines or entire networks. Moreover they seldom change them.

Re:Open crypto algorithms; no fix for Win2K

[lgw]

Does 7 years sound like a long time for a computer product to be in service to you? What platform do you work with again?

Re:Open crypto algorithms; no fix for Win2K

[TheAwfulTruth]

Exactly how many Linux distributers support 7 year old versions of their OS?!?!?

(Well MAYBE Debian...)

Most of them crap out after 12 months!

Re:I have to agree with MS on this one...

[El Lobo]

Hmm... so if somebody writes M$ that makes him obviously a troll? OK, so 5/6 of the posts here are trolls then if you are right.

Re:I have to agree with MS on this one...

[Ash Vince]

I know twitter/erris posts regularly but not that often.

What the hells has that got to do with anything?

Plenty of people call microsoft M$. Personally I prefer calling them MicroShite but that is my preference.

Twitter also occasionally makes some valid points in some of his posts but who cares about facts when you can just slag someone off without taking the time to exercise your brain.

I am not saying the Linux is perfect, it pisses me off just as regularly as Windows does but at least with Linux I can do something about it like commit a patch. With Windows I might as well just lump it as there is nothing I can do to help remedy the situation.

This is what a lot if Windows and Apple fan boys miss. Linux does not annoy as many coders as we feel we can remedy the things about said OS that annoy us, whereas with windows there is the feeling of complete powerlessness to fix problems even if you can isolate exactly what causes them. For anyone who programs computers, relies on them to work and encounters the same bug regularly this becomes tremendously frustrating as it prevents true self-reliance.

To come up with a car analogy it is like having to pay for a cab regularly because your car is constantly in the auto shop getting repaired for an issue you could fix if only it would not void your warranty.

Re:I have to agree with MS on this one...

[Legume]

Hmm... so if somebody writes M$ that makes him obviously a troll?

Yes.

OK, so 5/6 of the posts here are trolls then if you are right.

Yes.

Terms like M$, Linuzzz etc. amount to petty, schoolyard name-calling. Useful dialog is only diminished by them.