Slashdot Log In
U.S. Gov't To Use Full Disk Encryption On All Computers
Posted by
timothy
on Thu Dec 28, 2006 10:48 AM
from the double-secret-probation-rot-13 dept.
from the double-secret-probation-rot-13 dept.
To address the issue of data leaks of the kind we've seen so often in the last year because of stolen or missing laptops, writes Saqib Ali, the Feds are planning to use Full Disk Encryption (FDE) on all Government-owned computers. "On June 23, 2006 a Presidential Mandate was put in place requiring all agency laptops to fully encrypt data on the HDD. The U.S. Government is currently conducting the largest single side-by-side comparison and competition for the selection of a Full Disk Encryption product. The selected product will be deployed on Millions of computers in the U.S. federal government space. This implementation will end up being the largest single implementation ever, and all of the information regarding the competition is in the public domain. The evaluation will come to an end in 90 days. You can view all the vendors competing and list of requirements."
This discussion has been archived.
No new comments can be posted.
The Fine Print: The following comments are owned by whoever posted them. We are not responsible for them in any way.
Full
Abbreviated
Hidden
Loading... please wait.
But why? (Score:2, Funny)
Re:But why? (Score:5, Insightful)
Parent
A stake through the heart of non-commerical linux (Score:5, Interesting)
Now the hackles being raised are that this means we can't use Macs and maybe not linux since there are no acceptable enterprise-worthy full disk encryption systems. If you know of some, expecially for macs please reply with details below. But the term "acceptable" and "enterprise-worthy" matter a great deal. You can't just go installing full disk encryption based on some open source solution that might or might not get updated to work with the next version of say debian or fedora in a timely way. It has to have a method of key escrow that is usable. etc...
Hence people are looking to windows.
Another raging argument is what full disk encryption means. Surely something like mac's built in encryption of home directories and if need be combined with secure virtual memory would be sufficient to protect anything but very critical information. The answer we are hearing is No and "maybe". We are beinf pushed to use Entrust which all users I have heard from say is a disaster. There's going to be huge data recovery issues. And I don't see it as likely that Entrust will always be assured of working across OS upgrades
Personally I'd prefer to see encryption done in a transparent hardware layer.
In the long run this going to be good for the branded commerical OS, and the Linuxes backed by commerical vendors. The reason is that in the end you'd have to be pretty stupid to encrypt your whole disk with anything not supplied by the OS vendor because it simply has to work right under all circumstances and there simply has to be one person you can call when it fails. It woul dbe intolerable to have to have the OS vendor say well it's not our problem and the encryption vendor saying they are trying to work with the OS vendor to figure out why the kernel upgrade broke it.
And when it does break after you hit the "Software update" button or worse corporate HQ pushes the update overnight to your computer there is no failsafe mode! the computer won't boot. Corprorate HQ can't even contact your computer to undo the problem after the reboot. you can't even donwload a patch from the vendor or let them know it was broken. You can't even look up their phone number. Nor can you go to your neighbors computer to download a patch since his machine is broken too.
Other arguments people are unsure of
1) is home directory encryption enough
2) what about removable media?
3) what about FAT tables?
4) boot tracks?
5) virtual memory?
The fact that this order is zero tolerance with no asseement of risk seems to prove it is ill conceived.
It's a stake through the heart for all non-comercial linux
Parent
Re:A stake through the heart of non-commerical lin (Score:3, Interesting)
Not necessarily. You're assuming that this gigantic government-mandated undertaking is going to work. I think that is a mistake.
Ask yourself how many times such major overhauls have ever worked right, when the Feds are in charge. The FBI botched a big upgrade, the IRS is still botching theirs, the FAA botched theirs
Re:But why? (Score:4, Insightful)
On the other hand, they're losing laptops full of veteran's records on a monthly basis. Either they need to take better care of the data, or they need to put tighter controls on who has access to the data.
Parent
IT'S ABOUT FREAKING TIME! (Score:4, Insightful)
Large corporations that deal with private data from their customers should also be required to use full-disk encryption as well. In fact, I recommend some form of encryption for sensitive data to everyone.
Parent
Re:But why? (Score:5, Informative)
And, you'd be the first one to cry to the f*&king heavens as soon as the Government let YOUR secrets out in the open. Or when a government, controlled by a political party other than your chosen favorite, screwed up in a major way when Intelligence is released into the wild.
Find a government on the planet that does as you desire, I'll show you mythology. Only those seeking the downfall of a political system, or governing body require that body to release all its secrets. When that body is your government, then you meet the definition of "Traitor".
Whether controlled by Republicans, Democrats, Libertarians (mythological political party), The Raving Loons of Parump, the government must keep secrets and protect select information from release until such a time that its release is no longer a harm to the citizens and country.
Parent
Re: (Score:3, Insightful)
Then you give them Ultra-speedy flash memory HDDs and lots of Ram
And only then can you find back that speed feeling you had when you first launched Win95...
BTW, you forgot something
"each file is decrypted, scanned" then encrypted again into a secure memory heap with a random location in Ram then reinterpreted and decrypted from memory by the CPU for processing "and then viewed" on a secured, shielded screen that itself is decrypting the secured data transmis
Eh. (Score:5, Insightful)
On the other hand, this is just a bandaid on their terrible information policy...The reason that they have to encrypt a zillion machines is because they store sensitive personal data on a zillion machines. Then there are multiple operating systems, levels of security, etc. All this means that compromising one machine will still be pretty easy, because when you have encryption on the crappy desktop in the mailroom where everyone surfs porn, you stop taking it seriously.
They could kill the whole problem by centralizing their data stores, and developing some secure web interfaces across enhanced encryption. That way, instead of trying to encrypt every machine, you could encrypt 50 data centers and control access locally...Hell, if I were the government I'd push all my software needs toward think clients and terminal services anyway...The average user doesn't need more, and that makes all your security problems more managable.
Re: (Score:3, Insightful)
Re: (Score:3, Informative)
Because not every government employee has access to high bandwidth connections, especially if they are stationed outside the US. Disconnected operation is essential.
Re: (Score:3, Interesting)
If you work as a low-level US diplomat in Peru, do you really need to carry around the complete medical records of 20 million veterans?
Additionally, you can get a tolerable bandwidth connection anywhere on the planet - We now have these things circling the Earth far above, sort of artificial "satellites", if you will. Some of them have the
Re: (Score:3, Insightful)
No, I don't suppose I would.
Now, sitting in that same vehicle in Iraq, why exactly would you need to know the complete VA medical benefits history of Frank Brown, age 84, current location (and residence), Chelsea WI?
Offline access to (nonclassified) maps, great. Offline access to somewhat sensitive information directly relevant to your mission, fine. Offline access to data that has no u
Re:Eh. (Score:5, Informative)
It's not 'dragging this stuff home', it's people who go out in the field to do their job - One simple example is FEMA. When they go to a disaster they take along thousands of laptops in order to register people who need aid. There isn't a LAN they can "SSH into" and they can't phone this stuff in. Another example might be the IRS who would visit individuals and businesses to perform audits.... The list goes on.
Parent
Re: (Score:3)
Or, roughly translated: "Let's have an armed revolution and a new consitution!"
That's what you'd need in order to implement a FAIR flat tax.
Re:Eh. (Score:4, Insightful)
Parent
Re:unpopular data/facts, not "personal data" (Score:4, Informative)
Parent
Don't lose your pass-key (Score:4, Funny)
Re:Don't lose your pass-key (Score:5, Informative)
The Air Force currently requires ( in addition to the use of a "Smart Card" plugged into the machine to gain access ) a 15 char password consisting of 3 caps, 3 lower, 3 numbers, and 3 special char ( the rest is up to the user ), no proper names, dictionary words, more than 3 letters or numbers in sequence ( back or forward ), must not be the same or simular to your last 25 passwords, and you must change it every 90 days.
The net result is that most people are writing it down and storing it in some easy to access place. Previously, we had an 8 char pass that required 2 caps, 2 lower, 2 special, 2 numbers... It was short enough that you could actually remember it.
Parent
PS... (Score:4, Interesting)
Parent
Re:PS... (Score:4, Insightful)
"What?? I can't change the clock on the PC? How am I supposed to test this function that generates a string based on the time?"
"What? I can't defrag my own harddrive?"
"What? I can't create a folder in C:\?"
The SDC is good, but damn...some of us need a little more.
Parent
Re:PS... (Score:5, Insightful)
I hate to sound like a dick, but....good!
By being forced to develop your software as a restricted user, you're forced to ensure that your software will run with restricted user privileges. You're forced to use the proper means of determining the user's home directory, their temp directory, etc. You're forced to use the HKCU registry to store any registry items. You're forced to make the software multiuser-capable.
That's the way it should be. If most software had been written like that from the beginning, Windows would probably be a lot more secure for the general population because they would be able to comfortably run as a restricted user and know that all their software would Just Work.
So while it may be more painful as a developer to run as a restricted user, the pain does have a rather substantial payoff. Hopefully that'll make the pain a bit more bearable.
Parent
Re: (Score:3, Insightful)
Uh, bad example. Good design would normally dictate you prototype this function as follows:
char* generate_string(time_t time);
Now you can call it using the output of time(), or in a unit test, try a bunch of different time_t values. What, were you going to have your unit test keep changing your system time?
Then, if you area always going to call it using the current time, simp
Re:Don't lose your pass-key (Score:5, Insightful)
Parent
Re: (Score:3, Informative)
Then you have decent physical security as well. Don't get me wrong, it's not perfect but it is still very effective.
Re:Don't lose your pass-key (Score:4, Funny)
Parent
Why Full-Disk?? (Score:2)
Are they just concentrating on a Windows-only solution that will lock out OS X and Linux??
As a government employee, I know there are a lot of people where I work who want to keep their Macs.
Re: (Score:3, Informative)
Because software frequently puts sensitive data in files outside your home directory.
Are they just concentrating on a Windows-only solution that will lock out OS X and Linux??
Linux supports full disk encryption. If OS X doesn't, well, it should, since home-directory-only encryption is not particularly secure.
Re: (Score:3, Informative)
However as other contributors rightly pointed out,
Re: (Score:3, Funny)
Re: (Score:3, Informative)
Are they just concentrating on a Windows-only solution that will lock out OS X and Linux??
From the requirements:
Truth be told, this doesn't really say that much ... 'It is important if you support multiple' - what does that mean?
Re: (Score:3, Insightful)
should be? You gonna personally guarantee that every possible Linux and Mac application store all of their information in the same place? If we're talking "should be"'s, then there wouldn't be this problem in the first place, because no sensitive data should be stored on laptops that walk out of buildings. "Should be" is what causes the
Re:Why Full-Disk?? (Score:4, Insightful)
All that said, I think that giving a contract like this to a commercial vendor developing proprietary software would be... unfortunate. Funding addition of missing, necessary features to TrueCrypt would be a one-time expense (rather than one which scales with the number of systems deployed), and would benefit the private sector as well.
Parent
Sid by side competition? (Score:5, Funny)
Halliburtons new encryption subsidary.
Founded in 2006 by some guy who read a book on encryption.
Re: (Score:3, Insightful)
What gets me is that PGP isn't competing for this DoD bid. Of all the FDE solutions I have used, I like PGP's because it offers not just a PKI, but an open, standardized PKI that has stood the test of time. This is not to say that other FDE software isn't good. Safeboot, SecureDoc, DriveCrypt, and CompuSec are all very good solutions too.
What is funny is that FDE s
Re: (Score:3, Insightful)
People need to understand this. Government rules, regulations and procedures disqualify most possbible bids. Only those companies *specialized* in government contracts get these jobs. In addition, the margins on these jobs are so small, that larger companies have a huge advantage in the bidding process. Throw in several layers of lawyers and you end up with a system several realities removed from any semblance of a market.
Hey, Government! (Score:5, Funny)
Bidding starts at $47 Million.
This is my job... (Score:4, Interesting)
The June 23 White House memo had a 45-day deadline. Everyone has already blown the deadline.
Big props to WinMagic for their marketing. They've been all over the government computer press for the last 1-2 years with press releases and random mentions that make it appear they are the only workable solution. As a result, the agencies that jumped on the bandwagon in time to meet a (seemingly common) end of year deadline have grabbed their SecureDoc software and started installing. My experience with it has been semi-OK. Given that the software is touching every single file on every machine that leaves our physical space, the number of screwups has been acceptable at less than 2%. Our most widespread problems have mostly been a result of insufficient server capacity to deal with all the machines being encrypted at the same time within the last couple of weeks. Whether that was a result of us going cheap on the server side or WinMagic promising that the servers could handle a bigger load than is actually the case, I don't know. I suspect it's a bit of both. Still, things are slowly working out, even if our frontline support staff is going to wind up losing, literally, a month of productivity to the project.
A bunch of the requirements on that DOD checksheet are being ignored by civilian agencies. With no PKI infrastructure in lots of places, plenty of things have to be done "hands on" and the ability to do things like silent installs is out the window.
A bunch of the names on that vendor list are just resellers and of little interest to the slashdot crowd. What's more interesting is the list of products that do the job. THAT list is much, much shorter.
I haven't heard of anyone doing their encryption in hardware, which irritates me. I use hardware-encrypted drives at home and I was looking forward to doing the same thing at work. There is a widespread rumor in my agency that 2 or 3 generations of computer refreshment down the road, we'll transition to encryption in hardware. I hope so.
We've been doing this for 5+ years now (Score:5, Informative)
At that time we found just 5 vendors who were qualified to deliver (after an initial pre-qualification round), and we invited them all to a specially setup testing lab: Of these 5 vendors, 3 were selling pure snake oil (encrypt the partition table and/or root directory only), it took less than 5 minutes to break into each of these.
Nr 4 seemed a lot better, but after 20 minutes work I found the crucial 'compare password, JE decrypt' sequence in the driver, and we were in.
Only the final entry (from a german company) had understood how you design a product like this:
First you encrypt, using your preferred symmetric key algorithm (AES-256 these days?), all sectors on the disk. You use some form of hash of the logical sector number as a salt when encrypting, this makes each block unique, even those that contain the same 'FDFDFDFD' freshly formatted pattern. The key you use for this is the master disk key, it is a random number generated during installation.
Next you make a small table, with room for at least two entries: User and admin.
The user entry can be modified as often as you like (we default to slightly less than once/month), while the admin key/password is constant, but unique to this particular PC.
Each password (user/admin) is used as the key when encrypting the master key, which means that there is no way, even for the crypto architect, to recover the master key without knowing at least one of these passwords. (The passwords are never stored anywhere on the disk of course!)
The admin key/password is saved both as a printout and on disk on a secure system (without any form of network connection), so that you can use it each time a user manages to forget his/her user disk password.
There are lots of nice to have features as well, one of the more important is the ability to use a challenge/response setup to safely regenerate a user password remotely, without ever having to transmit the relevant admin key. This does require some kind of side channel to verify the identity of the user who owns the particular laptop: We use a combination of RSA's SecureID cards and the user's cell phone for this (each user has such a card to be able to use the corporate VPN connection which requires strong authentication).
Terje
start your own company (Score:5, Interesting)
I work for a multinational corporation with more than 10 K laptops
Just wanted to give you a reality check:
If you work for a company like that and know this technology to the level you are describing in this post, you should leave your employer to start your own company providing this solution. There's no way you're getting paid at a multinational corporation as much as you would make in your own (successful) company. If you had launched your company back when you had performed the aformentioned evaluation, you'd probably have enough progress with your own product to pitch it in this govt. bidding process.
Not trying to criticize you. Just trying to inspire people.
Seth
Parent
It takes more than knowledge to start a company... (Score:3, Insightful)
Maybe the GP has all those skills and is willing to take the risk, maybe he doesn't. The point is though that the lure of making more money, or having more control ove
Re:start your own company (Score:4, Informative)
Parent
Re:start your own company (Score:5, Interesting)
Before taking a one-year sabbathical (91-92) which I spent in the US, writing networking code, I had a company that sold terminal emulation/file transfer software. I sold enough licenses to make it one of the top 5 bestselling norwegian programs. During the last year the norwegian IRS grabbed 83% of every Krone I invoiced my customers.
At that point I realized that I'd much rather work less and spend more time with my wife & kids, so I closed the company.
I still write/optimize code, but always because I enjoy it, not to make money. (Sometimes I do get paid as well (in addition to my regular salary), but that's not the important part.)
Re. "know this (crypto) technology": I want to know a lot more than just crypto, and the job I have, which is a sort of IT Fire Brigade Chief, means that I get to work on all sorts of interesting technology, including everything that's new, as well as everything that doesn't perform as well as it has to. The Full Disk Encryption requirements I mentioned in my first post were obvious to me at the time, but not to most of the vendors unfortunately.
I spend my leisure time on orienteering http://orienteering.org/ [orienteering.org], which is the perfect thinking person's sport.
I'm also the Scandinavian coordinator of the Confluence project http://confluence.org/ [confluence.org]
Check google for my other interests!
Terje
Parent
Re: (Score:3, Insightful)
Not initially, but I believe the current version does so.
The boot sequence is to load (from a reserved area) the FDE sw which first tries to verify that it is running in plain unprotected DOS mode, then it takes over the keyboard hw so that it can read keystrokes without risking a trojan/keylogger attack.
After getting the password/passphrase it uses this to decrypt the user entry which contains the master disk key: If this doesn't succeed it goes into a sw timeout loop, taking prog
FDE Requires Gov't ID Card (Score:3, Insightful)
All gov't employees will at some point get an ID card similar to the Common Access Card. This will have a number of public keys on it. One of which probably decrypts their workstation.
The U.S. gov't is building the capacity to issue millions of smart cards on their own. See this: http://www.fcw.com/article94813-06-07-06-Web [fcw.com] There was a proper publicly available contract up for bid for this project but it wouldn't surprise me if it has been pulled in favor of a no-bid award.
Before anyone says, "Well it should be a secret! What if the terrists get a badge?!" There are two things to remember.
1. Lots of bad people have proper ID in their country of choice. Identification has little if any relationship to their activities. The failure points remain the usual human factors out in the field.
2. There's no need for secrecy in the production environment. Every half-decent perso system/PKI properly manages such an obvious point of failure. If a Visa-certified card plant can manage to keep track of 10's of millions of cards anyone can. It's not rocket science.
I for one welcome our fully encrypted overlords.
Right and Wrong and Gravity (Score:4, Insightful)
I can however confidently predict that since a very large number of people are involved in making the decision, the worst possible product will be chosen.
So it won't be TrueCrypt, or something decent - it'll be something like the latest commerical version of PGP.
000000? (Score:3, Funny)
Re: (Score:3, Insightful)
Well, this should be fully analyzed to see whether it's actually going to protect anything, or whether it's just "Something must be done! This is something my brother who runs this one company told me about, therefore we must do it!" For instance, laptops are involved in the majority of data loss cases. If someone suspends a laptop and sets it down somewhe
Re:Doomed to software failures... (Score:5, Funny)
That reminds me, whatever became of that ARPANET thing they were all talking about way back?
Parent
Re: (Score:3, Informative)
loss of speed.
http://www.freeotfe.org/docs/index.htm