Vista Protected Processes Bypassed

Anonymous Hero writes "Security Researcher Alex Ionescu strikes again, this time with a proof of concept program that will arbitrarily enable and foremost disable the protection of so-called 'protected processes' in Windows Vista. Not only threatening Vista DRM and friends, it's also another step towards hardened and even more annoying malware. Normally, only specially signed processes made by special companies (decided by Microsoft) can be protected, but now the bad guys can protect any evil process they want, including the latest version of their own keylogger, spambot, or worm, as well as unprotect any 'good' one."

In related news

[tinkertim (918832)]

A spokesperson for Microsoft was quoted as saying :

This is only an issue if you're downloading and watching porn. You should be watching only wholesome media, like "What About Bob [wikipedia.org]", instead.

Re:In related news

[_KiTA_ (241027)]

A spokesperson for Microsoft was quoted as saying :

        This is only an issue if you're downloading and watching porn. You should be watching only wholesome media, like "What About Bob", instead.

People are modding this as flamebait, but I've seen far, FAR too many IT professionals take that stance with Spyware / Malware. I've seen a system get all sorts of nasty winlogon-enabled Spyware within minutes of being hooked up to a network, with no action on the user's part. Not only that, in a world where banner ad companies can get infected with trojans [out-law.com] the idea of people only getting infected if they're doing something "shady" on their machine is utterly absurd.

Re:In related news

[tinkertim (918832)]

People are modding this as flamebait, but I've seen far, FAR too many IT professionals take that stance with Spyware / Malware. I've seen a system get all sorts of nasty winlogon-enabled Spyware within minutes of being hooked up to a network, with no action on the user's part. Not only that, in a world where banner ad companies can get infected with trojans the idea of people only getting infected if they're doing something "shady" on their machine is utterly absurd.

It was a joke, just a joke and only a joke.

The link given is to Microsoft Bob, which Microsoft gave up on shortly after launching it and (according to Wikipedia) later admitted the product was their single largest failure in their company history.

You'd need to remember Bob in order to appreciate that Vista is well on its way to being "Bob 2".

I suppose any joke could be taken as flamebait lol, but really, its just a joke. Better put in /. terms :

its funny, laugh. .. or perhaps not, since I had to explain it :)

Re:In related news

[erroneus (253617)]

I rather liken Vista to WinME. But every time I say so, someone chimes in saying Vista is the best thing Microsoft ever did or that Vista sales have set new records here or there or somewhere.

Vista goes way ot of its way to reduce functionality for the user in order to make content providers happy. Think of what that really means. Company A sells something to Consumer A but that something is disabled in order to make Company B happy. Company B is happy because they can continue their old business model and maintain their dominance if and when they finally move into new business models when they feel ready. Meanwhile, companies C, D and E through M move to create, innovate and design new things only to be prevented by both Company A and Company B. Depending on how this is done and how much evidence can be produced, this is illegal behavior.

Re:

[PingXao (153057)]

It's the same way with spam. Too many people are content to say it's only a problem if you're not using spam filters. They completely ignore the point that the spam exists in the first place and is transmitted hither and tither across the net, stealing bandwidth far and wide.

Re:In related news

[cduffy (652)]

The only infection my home Windows system has ever had came from a MySpace page my wife was browsing. Both of us appreciate good porn, and use that system for viewing it -- and, as I said, the only infection we've ever had was from MySpace.

The parent is not necessarily too uptight to admit surfing porn.

Re:In related news

[StinkyGeek (1035888)]

I have to ask. If both you *and* your wife enjoy porn, how do you find time to post on /.?

Re:In related news

[LighterShadeOfBlack (1011407)]

You're wrong. The "collective observations of thousands of admins" is in fact little more than assumptions and anecdotes perpetuated by people such as yourself.

Do a significant proportion of porn sites have malware? Probably.

Is there a greater risk of getting infected by malware when surfing for porn than doing "wholesome" surfing? Perhaps.

Is a malware infection reason enough to presume that they got it from browsing porn and/or piracy-related sites? Not in the slightest in my experience. If you've got differing experiences that prove me wrong, by all means collate your data and present your findings because I and I'm sure many other people working in admin or IT roles would love some hard numbers on the nature of malware sources online. Until then I'll have to assume the "observations of thousands of admins" you speak of are in fact nothing more than your own pre-conceptions.

Re:

[Master of Transhuman (597628)]

I'll tell you, personally I think porn sites don't need malware. They KNOW what you're there for - they don't need to slap adware on your system to get you to come there. I've always had some spyware protection back when I was running mostly on Windows 2000 and XP, and I surfed porn sites frequently (albeit with Opera originally and later Firefox, more than IE, so my exposure to ActiveX was minimal) and I very rarely got any spyware according to my utilities.

Basically ANY sleazy commercial outfit will slap

Didn't we see this before...

[NecroPuppy (222648)]

With that OS protected space in Windows ME?

I clearly remember being called to help a friend with a spyware/malware problem, discoverng he had ME, and going out to buy a copy of XP to replace it.

Re:Didn't we see this before...

[FutureDomain (1073116)]

I clearly remember being called to help a friend with a spyware/malware problem, discoverng he had ME, and going out to buy a copy of XP to replace it.

Well, it looks like you might be doing it again. Helping a friend with a malware problem, finding out that he has Vista, and buying a copy of XP to replace it.

this is just an another step

[imbaczek (690596)]

...to start considering Vista as an usable OS.

Wait, wait...

[kripkenstein (913150)]

A typical process cannot perform operations such as the following on a protected process:
[...]
Access the virtual memory of a protected process

It's been a while since I knew squat about operating system internals, but aren't processes supposed to not be able to access other processes' memory anyhow? I assume, then, that this means that 'protected processes' are special in that they are also protected from any 'supervisor'-type processes, not just run-of-the-mill? In that case, are 'protected processes' meant to protect the kernel from itself, in some sense?

Most likely I am missing the point here, and can't understand TFA accordingly. Somebody please set me straight.

Re:Wait, wait...

[Guilly (136908)]

There are ways, using the windows API, for any process run with Debugger privileges (any Administrator really) to read,write,terminate,create threads, etc in any other process. This was true in Windows 95 and still is in XP and probably Vista, except for protected processes.

It's not like they can just create a pointer and address the other memory space but using the API they can achieve the same thing.

This is what allows programs like xfire to inject into your game process or (as they mention in TFA) allows Warden to peek inside all processes to see if they are evil.

No, debuggers can't have special privileges

[DeadCatX2 (950953)]

When you start a process, you start it with a certain set of privileges. If you're logged on as administrator, your calls to CreateProcess [microsoft.com] can start processes with a different set of privileges.

When you make a Windows API call to something like CreateRemoteThread [microsoft.com], you need a handle to the process you're interested in. If the right security bits aren't set (and they get set by the call to CreateProcess), CreateRemoteThread returns unsuccessfully.

Anyway, what could you do to give debuggers special privilege

Re:

[randyflood (183756)]

I could be wrong, but I think Windows (2000, XP) generally allows processes running under the same user to look at each other's memory and such. This is useful when you want to debug a program or whatever. It's generally designed to protect users from each other, rather than protect users from themselves.

Ever since DOS

[Original Replica (908688)]

I miss the days when I gave my computer commands not suggestions. This whole "protected area" stuff just pisses me off.

Re:Ever since DOS

[by Anonymous Coward]

I miss the days when I gave my computer commands not suggestions.

You are becoming nostalgic, Deny or Allow?

It's really Melinda's fault

[ColdWetDog (752185)]

Want your missing is the higher social value of interacting with your computer on a more equal basis. Just like women, Computers are complex, pretty, expensive and inscrutable. Just like women, they are best handled with suggestions, not commands.

So get off your old, tired, 20th Century horse and get with the new paradigm.

Just a suggestion of course.

Re:Ever since DOS

[Udo Schmitz (738216)]

I miss the days when I gave my computer commands not suggestions. This whole "protected area" stuff just pisses me off.

So, is using a Vista PC like talking to the bomb in Dark Star?

biting the hand that feeds you

[kv9 (697238)]

He [Alex Ionescu] is also a Microsoft Student Ambassador and is representing the company on campus as a Technical Rep.

not for long, I bet.

New Meaning for "Genuine Advantage"

[BoRegardless (721219)]

Genuine Advantage seems to now benefit the bastards too.

possible silver lining

[Trailer Trash (60756)]

Could this technology be used to make a file copy command for Vista that isn't dog slow? Just wondering...

Surprising really?

[loconet (415875)]

If you build a house out of hardened excrements, it is still a house built out of shit even if you paint it pink.

Again?

[Proudrooster (580120)]

VISTA hacked again? In about three years I predict this OS will actually be usable due to helper apps which allow end users to use the computer as they see fit, instead of how MS and friends think you should use it. DRM is such a waste of human resources, but I guess this is the game we have to play.

Bill Gates wants more cheap labor [infoworld.com] to waste of useless software [theinquirer.net]. What a waste of human intellect and talent. How about making the computer RUN faster, be more intuitive, and reliable?

Good, now MS cant dictate software advantage

[plasmacutter (901737)]

all DRM issues aside, i'm surprised nobody has brought up new antitrust charges, especially in europe, for this idea that microsoft is allowed to deny a company the ability to use process protection.

by doing that they give incumbents an advantage over others and are using their OS to exapand monopoly interests into other sectors.

Good idea, bad implementation.

[Animats (122034)]

"Protected processes" are a reasonable idea. They're certainly better than putting video and audio processing in the kernel as part of the DRM system. But apparently Microsoft botched the implementation.

Microsoft has for some years allowed processes to do too much to other processes. Things like "injecting" a DLL or thread into a running process from the outside, or "hooking" system calls, are inherently security problems. In the Windows world, normal processes can do that to each other. This tends to be overdone, with too much "hooking" of system calls and such, a tradition from the DOS era. The UNIX/Linux world doesn't have that tradition. Fortunately.

In the Linux world, the things you can't do to a Microsoft "protected process" are roughly equivalent to the functions of the PTRACE [linuxgazette.net] call. In SElinux, the mandatory security system controls which processes can use PTRACE on which other processes. [12.110.110.204] So SELinux already has "protected processes", but with a better security model.

If we have to have DRM, protected processes aren't a bad idea. But what you want is for them to be compartmented, not privileged. They should be running in a compartment which prevents other processes from attaching to them, but they don't need the privilege of attaching to other processes. So the video decoder can be protected, but doesn't have enough privileges to act as an aimbot for some game. The security system for a game should be able to lock the game processes into a compartment which other processes cannot enter, preventing cheats. Enforce separation, not privilege.

Re:Good idea, bad implementation.

[Spy Hunter (317220)]

Protected processes are a terrible idea, and they have no analog in Unix. You have misunderstand the purpose of protected processes. It has nothing to do with protecting processes from each other for better security. It is *only* about protection from the *user* for media. Protected processes cannot be written by anyone but Microsoft and "trusted" partners (theoretically) and are supposed to be immune from tampering by every user, even one with the highest possible administrative rights. No Unix has this concept, because it is retarded. It removes your own control over what your computer is doing and hands it to Microsoft and a few "trusted" companies which are allowed to write protected processes.

This is how it's done

[by Anonymous Coward]

The tool needs to be run with elevated privileges (otherwise it will not work). It decompresses a 848 bytes driver and loads the driver. The driver does nothing but set bit 11 (ProtectedProcess) of the Flags2 bitfield (offset 0x224) of the corresponding _EPROCESS structure of the process to be modified. However, this requires the neccessary rights to load and install a driver...and as we all know, once being in kernel mode there's no real protection against malicious code...

The Philosophy of Protection

[The Living Fractal (162153)]

I think history has shown that no matter how hard you try you cannot create a doorway in software protection and only expect to let those you want get through. The nature of software today is so fluid that it's possible to make your way through the door by imitation, brute force, social engineering, etc. Microsoft does not seem to grog this. Neither do DRM propenents. Information will find a way to get through, around, over and above, and beneath all obstacles.

So what do you do? Well, one thing you don't do is provide special security rights to only certain approved software.

The only true answer is open software and education. People who don't know how to use their computers will be attacked. They will be compromised. If you can't control yourself on the internet and local networks, you will lose the right to control your computer because someone will take it from you. If you run unknown and untrusted programs, you face the risks. Your online habits help determine your exposure. If you absolutely must visit 'free porn', warez, social networks like MySpace, etc websites, then do so with caution tempered by proper education on how to isolate your important, sensitive data, from the rest of the crap you are willing to lose. You are better off simply not visiting sites of that nature. But if you are going to, at least understand how to keep yourself safe. Because no software written today is going to be able to do it for you. There will always be software out there capable of getting around it.

In the end, to the wolves go the slowest, weakest sheep. It's natural. Don't be one of them.

Looks like 32-bit

[figleaf (672550)]

I would like to see him do this in 64-bit.
32-bit allows unsigned code in kernel mode for legacy reasons so its much more easier to inject into 32-bit processes.

Re:Other OSes

[by Anonymous Coward]

No, this feature is available only in Windows Vista.

Re:Other OSes

[diegocgteleline.es (653730)]

No, other operative systems don't have this stupid notion of "protected processes", not even XP has it, only vista.

Re:Other OSes

[I(rispee_I(reme (310391)]

Actually, Windows versions as early as 2000 use a whitelist method of "protecting" processes: If the process name matches a hardcoded list [seclists.org], then task manager will refuse to kill it. This is so broken it's ludicrous- simply rename your process to any of the ones on the list, and it becomes unkillable. Programs such as PSkill will kill all processes, regardless of name.

Re:Can't beat em, join em?

[Fallen Kell (165468)]

The problem with this is that the said paid hackers get better pay working on the exploits on their own and selling them in the black market. A lot of exploit code goes for $5000 a pop to the people who use it, and there are plenty of buyers (and it is not like they can't sell to multiple people, and make N*$5000 for a single good exploit). Heck, something like the above would easily sell hundreds or possibly thousands of times for $5000 a pop. Can most software companies afford to pay hackers the $300,000-500,000 a year that a good one could easily make off a single exploit?

You think so?

[Fallen Kell (165468)]

Do you really think so? Why would MS pay someone $300,000-500,000 when they have people who get $70,000 that could simply scan the code itself? They won't upset their current pay scales and pay grades to place "hackers" into their business units. For one, many of those "hackers" are hackers because they have a record of conduct that does not work in a normal business environment. Be it social, societal or other issues (potentially and not limited to criminal and trust issues). In fact, some people many not

Re:You think so?

[sqlrob (173498)]

Right, like those code scanners that preemptively found the second ANI bug after the first was found. Those code scanners?

Re:Can't beat em, join em?

[misleb (129952)]

Sure, but what kind of employees do these people make? And will they have the same motivation if they are being paid to do it? It is highly variable. You're little website is one thing, but if you're microsoft, you have a lot to lose. Maybe the hacker just wants to get on the inside to get better info for future illicit hacks... or worse, put in backdoors.

-matthew

Re:Can't beat em, join em?

[ultranova (717540)]

You're little website is one thing, but if you're microsoft, you have a lot to lose. Maybe the hacker just wants to get on the inside to get better info for future illicit hacks... or worse, put in backdoors.

Why would anyone bother putting in more backdoors to the OS equivalent of Goatse ?

Re:Can't beat em, join em?

[sjames (1099)]

That's MS's big problem. A LOT of people WANT them to fail because they're MS. Because fundamentally, a computer and it's OS is supposed to do what the user wants, not what Bill Gates, the RIAA and the MPAA want it to do. There are enough people out there who know how to hack it up so it actually does do what they want. The more pragmatic ones WANT MS to fail because that's how to crack the content they want.

Once the hacking is accomplished, a significant number of people will then abuse that code to get other people's computers to do what THEY want rather than what Bill wants (doing what the user wants is simply not up for discussion).

The real beauty here is that the "bad guys" are turning the OS's own features against the creator (the other bad guys). The divine appropriatness of that is simply irresistable.

Re:Why do they even bother?

[cyphercell (843398)]

no it's worse than crap when it can be exploited so easily. I read it as malware can become a "protected process", as in protected processes that the administrator doesn't have control over.

Re:Why do they even bother?

[cyphercell (843398)]

after a $b investment over five years from the dominant player in operating systems, yes "The WOW starts Now!"

Re:Why do they even bother?

[Rodness (168429)]

I agree.

The problem with Microsoft is not so much one of bugs as it is a problem with their general design philosophy.

Such as providing mechanisms for your own developers to bypass the security of the entire system to make some friggin media clips play more smoothly. News flash, idiots: if you provide two paths through security, a strongly checked path and a weakly checked path, you incentivize attackers to take the weak path! And if you provide those hooks for your own developers to bypass security, then attackers can use them too!

They were probably praying that no one would ever figure out that those hooks were there... and security by obscurity is very, very poor design.

My inclinations against myself or my family running vista just got a +1 Justification.

Re:

[Master of Transhuman (597628)]

You're one hundred percent right - and the reason is simple: security doesn't make Bill any money, whereas "featuritis" - and deals with big content providers - does.

Microsoft needs to be put out of business. Now. They have all the brains and social conscience of Enron.

Re:DRM in Vista is misunderstood

[jomas1 (696853)]

The addition of DRM in Vista allows you to play DRM-encrypted files on your computer. Without this feature, you would not be able to play DRM'd songs. Now at least you have the choice.

You can't possibly mean what you just wrote. Vista's DRM is needed to play DRM-encrypted files? Why can XP and Windows 2000 play encrypted files?

You're joking, right?

[MarkByers (770551)]

> Why can XP and Windows 2000 play encrypted files?

The ability to play some DRM'd files was also added to XP and Windows 2000. I assume you already knew that though...

Re:

[jomas1 (696853)]

> Why can XP and Windows 2000 play encrypted files?

The ability to play some DRM'd files was also added to XP and Windows 2000. I assume you already knew that though...

Ok so your original quote that suggested Vista's DRM, which is clearly different when compared to XP's and 2000's DRM mechanisms, is somehow a good thing was wrong? Or were you trying to say that some type of DRM is necessary? If the latter, then I don't know yet if I disagree. I can't however understand why you would criticize Ionescu for enlightening us to the flaws in Vista's security/DRM strategy. Ionescu did not make Vista any less secure than it was a week ago. He's simply let some of us know tha

Re:

[Original Replica (908688)]

yes, it would make a nice tool for you to administer your systems. or for anyone out there to "administer" for you.

Re:Source code

[cyphercell (843398)]

no one is a low life for holding on to their code. this guy just cracked the one of the strongest features of Vista. A system that took five years and a billion dollars to produce. About two months after public release and this guy has broken the "heightened security" wide open. If Symantec wants the code they should pay for it or figure it out themselves. Symantec doesn't give me anything for free. If you're using Vista, then you're an early adopter and need to deal with that, just thank this "low life" for providing you with a binary tool you can use if you get into trouble.

Re:Source code

[eddy (18759)]

Seems to contain a compressed buffer with a .sys driver that is decompressed with a call to RtlDecompressBuffer and hidden away by writing it to the alternate stream "%SystemRoot%\system32\drivers\crusoe.sys:drmkaud. sys", and then there's a registry update to load the driver.

Someone who cares should write out the compressed buffer and disassemble that.