Announcement

Collapse
No announcement yet.

Cheap GPUs render strong passwords useless

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • Cheap GPUs render strong passwords useless

    Get an extra 10% off with code READ10 when you spend $/£/€ 7 or more.T&Cs apply PC Pro is the UK's number one IT monthly magazine, keeping readers up to speed on the latest technology developments since 1994.Perfect for Keeping readers up to speed on the latest technology developments since 1994.


    How a cheap graphics card could crack your password in under a second

    Read more: How a cheap graphics card could crack your password in under a second | PC Pro blog http://www.pcpro.co.uk/blogs/2011/06/01/how-a-cheap-graphics-card-could-crack-your-password-in-under-a-second/#ixzz1OSrYu2Bx


    I was pointed in the direction of a blog posting talking about the use of GPU processors to launch brute-force attacks on passwords. GPUs are extremely good at this sort of workload, and the price/performance ratio has changed dramatically over the past few years. What might have seemed impossible even 36 months ago is now perfectly do-able on your desktop computer.

    In this report, the author takes a fairly standard Radeon 5770 graphics card (you’ll find it on our A-List under Value Graphics Card), and uses a free tool called ighashgpu to run the brute-force password cracking tools on the GPU. To provide a comparison point with the capabilities of a standard desktop CPU, he uses a tool called “Cain & Abel”.

    The results are startling. Working against NTLM login passwords, a password of “fjR8n” can be broken on the CPU in 24 seconds, at a rate of 9.8 million password guesses per second. On the GPU, it takes less than a second at a rate of 3.3 billion passwords per second.

    Increase the password to 6 characters (pYDbL6), and the CPU takes 1 hour 30 minutes versus only four seconds on the GPU. Go further to 7 characters (fh0GH5h), and the CPU would grind along for 4 days, versus a frankly worrying 17 minutes 30 seconds for the GPU.

    Now, I cannot imagine anyone managing to mandate a nine-character, mixed-case, random-character password on an organisation. But if you did, and you weren’t hanging from a tree by the end of the first working day, the CPU would take 43 years versus 48 days for the GPU.

    He then went on to add in mixed symbols to create “F6&B is” (there is a space in there). CPU will take 75 days, GPU will take 7 hours.

    What does this tell us? well, the stark reality is that even long and complex passwords are now toast. If you think you were being wise by forcing users to have randomisation in their passwords, then think again. It is utterly futile.

    Yes, you can force your users to have a 15-character password consisting of random numbers and letters, and throw in punctuation as well. This is great as an idea, but we know that most users think that a password like “Barry1943Manilow” where 1943 was the year he was born, is complex and hard to remember. Is an IT manager really going to manage to get the CFO to log in using “fR4; $sYu 29 @QwmQz” without the combination ending up on a Post-it note in his wallet? Or stuck to the side of the screen? Because anything much less than this is going to be open to attack over the next few years.

    A GPU of the type used by this chap is not unusual or high end. It is standard-issue stuff. Indeed, I have just sat through the AMD presentation here at Computex in Taiwan, and they made a big deal about putting GPU power into netbooks offering 500Gflops, without denting its 12-hour battery life. And that’s shipping within months.

    All I can say is this: you have been warned. It is time to think long and hard about password security, and how you do your authentication. This has crept up on us in the background, and we really haven’t been paying attention. Nor has Microsoft, frankly, who should be having a whole raft of alternative, hardened solutions in place ready for its business customers to roll out.

    What are the solutions? To be honest, I’m not sure. A combination of TPM, biometrics, passwords and maybe something else entirely new will be needed. But it’s clear that a complex password that users will actually accept for day-to-day authentication, and keep secret, might be history.
    Cool ****.

    This guy: http://blog.zorinaq.com/?e=43
    has 4 Radeon HD5970s (8 GPUs) which can do 33.1 billion MD5 password hashes/sec.
    "The issue is there are still many people out there that use religion as a crutch for bigotry and hate. Like Ben."
    Ben Kenobi: "That means I'm doing something right. "

  • #2
    oh, it's MD5. Wouldn't SHA1 or SHA512 take a lot longer? And unless you have access to the actual hash of the password along with the salt, you wouldn't be able to solve it this fast because it takes time to make each guess. If you have physical access to a system which is what is required to get the hash and salt, then you can already consider your data to be toast.
    If there is no sound in space, how come you can hear the lasers?
    ){ :|:& };:

    Comment


    • #3
      People use passwords less than 24 characters?
      “As a lifelong member of the Columbia Business School community, I adhere to the principles of truth, integrity, and respect. I will not lie, cheat, steal, or tolerate those who do.”
      "Capitalism ho!"

      Comment


      • #4
        SHA1 isn't much better. SHA256 or ideally 512 would be best.

        And yeah, basically this is only useful if you have access to the physical system, or you have the password hash otherwise (like hacking in to the database, a la Sony).
        "The issue is there are still many people out there that use religion as a crutch for bigotry and hate. Like Ben."
        Ben Kenobi: "That means I'm doing something right. "

        Comment


        • #5
          SHA3 is coming in 2012.
          "The issue is there are still many people out there that use religion as a crutch for bigotry and hate. Like Ben."
          Ben Kenobi: "That means I'm doing something right. "

          Comment


          • #6
            SHA1 isn't much better. SHA256 or ideally 512 would be best.


            I believe HC was talking about the runtime of the hash - IIRC MD5 is much faster (to compute) than any of those.

            Comment


            • #7
              This only affects passwords to turn on a computer, right? Not like Internet passwords?
              Captain of Team Apolyton - ISDG 2012

              When I was younger I thought curfews were silly, but now as the daughter of a young woman, I appreciate them. - Rah

              Comment


              • #8
                Originally posted by OzzyKP View Post
                This only affects passwords to turn on a computer, right? Not like Internet passwords?
                All passwords.
                "The issue is there are still many people out there that use religion as a crutch for bigotry and hate. Like Ben."
                Ben Kenobi: "That means I'm doing something right. "

                Comment


                • #9
                  They can bypass the protection against multiple failed entries?
                  In Soviet Russia, Fake borises YOU.

                  Comment


                  • #10
                    Originally posted by Oncle Boris View Post
                    They can bypass the protection against multiple failed entries?
                    No. This is more useful if, say, you hack in to a server and steal 100M user accounts with encrypted passwords. You can then decrypt the passwords and get them all.

                    Or if you have physical access to a computer, you could extract the encrypted password from the disk then crack it.
                    "The issue is there are still many people out there that use religion as a crutch for bigotry and hate. Like Ben."
                    Ben Kenobi: "That means I'm doing something right. "

                    Comment


                    • #11
                      There are three solutions to this that I'm aware of:
                      1. Password lockout (assuming that the attacker doesn't have the password file and so much check the passwords online) - after three (or however many) attempts the user must jump though some hoops to reset their password or unlock their system, e.g. if I fail to log into my mutual fund account five times then I need to call in and have the system unlocked. The problem with this solution is that it's open for DoS attacks - all an attacker needs to do to muck up your system is to get their hands on some usernames, after which they can keep locking out the users.

                      2. Password delays, e.g. you can only attempt a login every five seconds. One version assumes that the attacker doesn't have access to the password file, and simply puts the login thread to sleep for X seconds after a failed login (or after three failed logins or whatever). The other version also works for offline attacks: rather than hashing the password with SHA512, instead hash it a hundred times with SHA512. This puts a burden on the login system, but logins are relatively rare so this shouldn't be too problematic; meanwhile the offline attacker will find that it takes a hundred times longer to crack the passwords.

                      3. Passphrases. Require extremely long passwords without minimal restrictions on requiring special characters etc. For example, my Truecrypt partition uses a SHA512'd passphrase as its password; when generating a passphrase I take an easy to remember phrase ("Mary had a little lamb") and change a few works to stymie a dictionary (or nursery rhyme encyclopedia) attack ("Raoul had an obese goat"). (In practice I'd choose a passphrase at least three times as long as "Mary had a little lamb.") This way the password is long enough to have quite a bit of entropy, but at the same time is capable of being memorized.

                      Ideally (2) would be combined with (3).
                      <p style="font-size:1024px">HTML is disabled in signatures </p>

                      Comment


                      • #12
                        It might be possible to write a program to strengthen systems with inadequate password protections. Let's say my online bank password has a maximum of 15 characters with certain restrictions (e.g. requires one uppercase, one lowercase, and one number); then all I need is a program to generate salted SHA512 hashes. When I need to login I enter my arbitrarily long passphrase into the password generator, and it outputs 15 String64 characters and truncates the rest of the hash. The program would iterate through salts until it generated a hash that met with the password requirements, which in most cases shouldn't take too long (all password restrictions I've seen are of the "at least one of" variety, with only lowercase, uppercase, number, and special characters being in the "at least one of" category; the only time there would be a problem is if the password is obscenely short, e.g. at least one uppercase, one lowercase, one number, and one special character in a password with at most 6 characters, but in that case your password is going to suck no matter what you do so you might as well make it human memorable.)
                        <p style="font-size:1024px">HTML is disabled in signatures </p>

                        Comment


                        • #13


                          I suspect passphrases will just be the next big thing.

                          There's also the security question process. All of my banking sites now require I answer one of 5 or more security questions I set up before gaining access to the system. Stuff like "What was your first job?"
                          "The issue is there are still many people out there that use religion as a crutch for bigotry and hate. Like Ben."
                          Ben Kenobi: "That means I'm doing something right. "

                          Comment


                          • #14
                            Be quicker than remembering my password.

                            Comment


                            • #15
                              In Europe I have a little item, where I have to put in my card + password and a pass number (generated every time I need to get the number) and then I send the number onto the site.

                              It should be time dependent, but maybe not. But if it is... it requires a number in memory, a physical card in my possession, a specific time, a specific number, and a special apparatus.

                              JM
                              Jon Miller-
                              I AM.CANADIAN
                              GENERATION 35: The first time you see this, copy it into your sig on any forum and add 1 to the generation. Social experiment.

                              Comment

                              Working...
                              X