Slashdot Mirror
SecurityFocus On MS Security "Hole"
friday2k writes "There is an interesting writeup at SecurityFocus that puts the latest security 'hole' in XP into perspective. It is a worthy read and should remind us all of the real issues out there." And it collects into one place much of the flak I caught after posting about the claimed security hole opened by the XP Recovery Console.
I mean, if I wanted to hork data off of a system I had full physical access to, I'd just grab the drive, stick it in my pocket, and walk out whistling "Jimmy Crack Corn and I Don't Care."
Now I can't get that song out of my head!
Anybody else stunned that Slashdot posted an article about MS that didn't involve an explanation as to how they're incompetant?
.. but he is right about the physical security. Not long ago I walked a client several hundred km away through an OpenBSD boot via floppy so he could change his forgotten root password. I don't hear the masses screaming for Theo's head because this is possible.
Trolling is a art,
This appears to be a problem using the win2k recovery console on a winxp install, not the XP console.
And all it allows you to do is copy files around. Whoopty do. Pop in a linux boot floppy with ntfs support and do the same thing, only easier (because the win2k recovery console doesn't support wildcarding; lame.)
Once the general populace knows about a problem, the media has to say something, because how would it look if they didn't report on a new trend? Suddenly everybody "knows" about the problem, even though it does not exist.
I can't say that I don't give a fuck. I've just run out of fuck to give.
... who still thinks the Registry is a bad thing?
(comment to be taken lightly. Should irritation persist, chill.)
"Derp de derp."
Jornalistic integrity? Man which world do you live in?
I'm with the author on this one. I dislike MS as much as the next guy, but I'd WANT a recovery disc to dump me at a prompt if the data files were corrupt. If the files on the drive are THAT important, they should have been encrypted anyway...and if I was the admin of the box, they would already be encrypted.
I have nothing to worry about.
News flash: this is expected, and desirable, behavior. The Win2k RC can't read the XP registry, so it thinks it is a corrupted Win2k installation. When it can't verify the SAM, it bails out to the console. Administrators want this behavior. If you have an installation on which some third-party driver has hosed the registry, the Recovery Console will allow you to attempt to fix it. That's what "Recovery Console" means.
No recovery console does not mean to bypass the password set by the administrator. It means to recover data that has been lost due to reason "foo".
While I don't see it as being that big of a deal, you could do it w/any OSs bootdisk I suppose (or even a LILO prompt on a Linux machine) I think it is an odd bit of information that should be known.
Media organizations know they get eyeballs when their audience is afraid.
Ignorant and afraid of terrorists? Watch Fox News.
Ignorant and afraid of hackers? Read Wired, or WinInformant.
Maybe we should be afraid of ignorance, instead.
Laugh at my Lisp and I keeell you.
This isn't a security flaw.
This is desired administration behavior. The Win2k disc can't deal with the WinXP registry properly, so it goes straight to recovery mode. Recovery mode is pretty much useless to begin with, and you can't really do anything to a system in recovery mode
Besides, if you can physically walk up to the computer in question and boot it from a CD in your pocket, your security problem doesn't come from Windows - it either comes from a BIOS that doesn't support changing the boot order, or it comes from between your ears.
does XP Recovery Console run on Linux?
That's patently untrue. It's a well-known fact that Microsoft's security problems are not due to exposure alone.
Microsoft's development model is fundamentally flawed from a security perspective, because it squarely places featureset additions above security. The corporate culture at Microsoft is and always has been more about gaining marketshare than about anything else.
It seems that there are differences in security, above and beyond the monopoly domination Microsoft enjoys. How many ISPs use FreeBSD to run their servers? Hmm.. I wonder if there's more to it than just speed and the fact that FreeBSD is Open Source.
I'm not alone in my assesment. There's this security guru named Bruce Schneier. Perhaps his name has crossed your desktop at some point. He's contemplating getting a Mac, because he is tired of hassling with security problems on his Windows machines.
Read the EFF's Fair Use FAQ
In contrast, I know SQL Slammer was reported day-of. In this case, a free patch was available six months prior to the worm. And let's face it: if the patch is available but not applied, it's not Microsoft's, Oracle's, Linus's, or any other vendor's fault--only the SysAdmin in question.
One major difference was that SQL Slammer took out several networks, where Oracle did not have such impact.
To \.'s credit (and I'm going mostly off memory), but big critique was on the DB admins, not on Microsoft.
I totally agree on this - I've been doing Win2k installs for a few years now, and I'd have had to totally scrap god knows how many systems if it weren't for the recovery console.
And the fact that you can use the Win2k boot CD to log in without a password isn't a bug, or even a security hole, it's simply the fact that MS didn't require a password to use the Console in Win2k.
What do the critics want MS to do? Recall and patch every single Win2k boot CD?
sig:- (wit >= sarcasm)
Perhaps you missed the point he was trying to make. While the "its a feature, not a bug" argument is valid in many cases, this is not one of them. The whole argument can be ended with the simple fact that you need physical access for this "exploit". As mentioned in the article, and as anyone who follows computer security knows, once an attacker has physical access to a machine its game over. With that as a given, administrators WANT tools that allow them access to a system like this, its been included in systems back to the VMS days that I know of, and probably older.
I believe the rational way to view these types of articles is to look at what they're saying and actually stop to think about it, rather than flying off on blind tangents about bias. While it may be true that the author often defends Microsoft for whatever reason, this particular article is based on solid points that make a very compelling point on this specific issue.
If they reported _every_ M$ bug on Slashdot all the good articles would get pushed off the front page.
As opposed to now, when all the good stories getting pushed off the front page by reposts, you mean?
sig:- (wit >= sarcasm)
PHYSICAL SECURITY. This is the first tenet of network security. Prevent the box from being accessed by those who should have no access. This tenet, however well implemented, is absolutely useless if the baddies that mean your network harm are INDSIDE the network, which in 75% of cases is true. It's a sad-assed day indeed when your own employees are the evil that is supposedly lurking outside the firewall.
If this is a hole then so is the fact I can mount your ex2fs /home partition from a boot floppy and ftp all the filez there to whereever I want them to reside. Actually the linux "hole" is worse, as it has infinitely more powerful command-line tools available to a bootflopper.
/]% rm /*/* on my way out. Know your enemy, he's probably a family member.
People fear the Internet and what a hax0r could do to their PC, but (as this article proves) give me physical access to your machine and I could do more damage to you than 99.999% of crackers ever possibly could - and that's only because I'm not enough of a bastard to [root@localhost
-Mark
I've just found a huge bug in Linux security! If you boot from a Linux boot disk, then you can mount the hard disk and read files off it! Linux security all over the world is compromised! No server in the world will ever be safe again!
Oh, and anyone who disagrees with this, or tries to use some kind of 'logic' or 'rational argument' to disagree is a Linux apologist.
Actually, this 'hole' is worse the one in Windows. Windows config data is stored in the registry, which is binary and so is much harder to manually edit than the plain-text files in /etc/ on a Linux box.
I am TheRaven on Soylent News
Problem is, we're talking about Windows XP, so Mr. Pfeil is wrong.
Assuming one can get Admin access to the installed OS (re-installing OS destroys access to EFS-protected files), resetting the password on WinXP in a Workgroup (as opposed to changing it) destroys access to DPAPI-protected keys, and hence access to EFS-protected files.
Win2000 EFS is vulnerable to this sort of attack, but not WinXP.
With WinXP, an attacker should endeavor to crack the user's password rather than change it to a known value. Even so, this attack can be mitigated by a) using strong passwords, and b) using SYSKEY to protect the SAM from offline attack.
Other notes:
1) EFS was principally designed to protect data when the hardware has been compromised, so the premise of this whole comment is wrong.
2) EFS is one layer of defense-in-depth. It should be combined with strong passwords, SYSKEY, and proper recovery key management.
3) Windows XP Key security is discussed here.
4) EFS does not support keys on removeable devices as of WinXP.
In a domain, the Administrator account for the forest root domain is the recovery agent. Additional recovery agents can be assigned through the domain group policy object. The certificates are self-signed if no CA (Certificate Authority) is configured. Any recovery agent should export the private key to removable media and lock it up in a secure place and keep another secured copy off site. Delete the copy from the forest root's first domain controller.
On a stand alone server or workstation (Not a member of a domain), a self signed certificate is generated for use and the local Administrator account is the recovery agent. The private keys for the administrator and your own user account can be exported to a floppy or other removable media and deleted off the computer. Another copy should be kept in another secured location in case the first gets burned down, stolen, corrupt, etc. Make sure the floppy isn't in the laptop carrying case, otherwise, the theif will have your private key when he takes the whole bag.
Another important thing to note is that the document is decrypted in memory and a clear text copy isn't put on the drive. A hacker going through your drive, looking for deleted temp files will be wasting time. If you want to be extra paranoid, configure windows to clear the page file at shutdown.
For more reading:
Click Here
If you really want to learn this stuff, read this book. I found it to be extremely educational and was the only book to explain certificate server to me effectively.
Click Here
-Lucas
Windows NT and 2000 MCSE
Yea a stupid error was made and several sites reported on it. I am supposed to feel bad to bill or do what Tim Mullen says and "Give Bill a Break"? No I won't be giving Bill G. a break. Maybe if more articles are written which say how bad MS software is MS might actually have to be accountable one day.
So you're all for more articles making a big deal out "security holes" that aren't "security holes" at all?
Ever heard the fable about the boy who cried wolf? You should not support Microsoft-bashing for the sake of Microsoft-bashing when there's nothing behind it, it only lowers your own credibility. Focus on Microsoft's real problems.
NO CARRIER
Indeed. And not only featureset but usability and user-friendliness factor are also placed above security issues. :)
As a result we have a dominant OS that's insecure and a secure OS that's mostly unusable by anyone who is not a third generation sysadmin. In all that rush no one had the time to write an OS that's is BOTH secure and user-friendly. Flame away
If they reported _every_ M$ bug on Slashdot all the good articles would get pushed off the front page.
Gotta leave room for all the articles about toasters modified to run linux and whatnot.
::.. check out some Cell Phone Reviews
It doesn't matter how many users it has because they users won't be looking for security holes in the first place. So if you put 10 Windows users in a room, none of them would know much about these things. Putting 10 Linux users in a room, and you increase the chance that you'll find a real hacker. I'm a Windows user myself, so I'm not trying to sound like an elitist bastard. I haven't even uncovered any security holes in my life.
But it is difficult to determine this case, as there are a lot of questions and too few answers.
Let us instead look at a piece of software where the numbers are reversed - where Microsoft's product has only a small part of the market.
I am talking about the open-source Apache HTTP server, vs. Microsoft's IIS.
Apache has 60-70 per cent of the web server market. IIS has less than 30 at the moment. Yet, despite these figures, Apache has had far fewer known security issues than ISS. How does this fit with your question? Obviously, there are a lot more eyes on Apache due to its large market share?
So how does IIS come out so crappy when it comes to security?
I think we can come to the conclusion that your "it's not as frequently used so very few are looking for security holes"-like statement simply does not make sense. It is a myth. FUD?
Clever signature text goes here.
The idea was to use a Win2K disk on a WinXP box and the Win2K thinks it is a "corrupt" install.
After seeing WinXP in action, I would tend to agree with the Win2k disk on its assessment...
well, I'll let you pick which end
/. that server.......
Law #3: If a bad guy has unrestricted physical access to your computer, it's not your computer anymore.
I wonder if we could