• 𝕸𝖔𝖘𝖘@infosec.pub
      link
      fedilink
      arrow-up
      2
      arrow-down
      4
      ·
      7 hours ago

      It’s not, though. And thinking that it is impossible is why DES, for example, was “translatable” by the NSA for decades. Never assume something is impossible just because it’s difficult.

      • Constant Pain@lemmy.world
        link
        fedilink
        arrow-up
        8
        arrow-down
        1
        ·
        6 hours ago

        It is. It is practically impossible to guess the file names. You telling otherwise means you don’t have sufficient knowledge on the matter.

      • Grendel84?@tiny.tilde.website
        link
        fedilink
        arrow-up
        5
        ·
        5 hours ago

        @01189998819991197253 @ConstantPain

        Security isn’t binary, it’s a spectrum. You apply the level of security that is appropriate for each situation.

        Of course it’s *possible* to brute force it, but by the same logic you could brute force jwt tokens, or api keys, or even ssl certs.

        It’s literally *impossible* to apply “max security” to everything, so you have to prioritize.

        What happened was unconscionable, but insisting uuid are mathematically breakable isn’t helpful, and can make it worse.

        • 𝕸𝖔𝖘𝖘@infosec.pub
          link
          fedilink
          arrow-up
          1
          arrow-down
          4
          ·
          7 hours ago

          The scope isn’t if they’re crackable (which, if course, they’re not, since they’re not encrypting anything). The scope is if using UUIDs as filenames in this publicaly accessible db a good way to hide the files. And the answer is: no it is not, because a computer powerful enough can guess all possibilities in a matter of minutes, and query them all against the db to discover all files stored within.

          • FooBarrington@lemmy.world
            link
            fedilink
            arrow-up
            3
            ·
            2 hours ago

            Aside from the fact that a strong enough supercomputer won’t exist for decades, you’re not limited by the speed of UUID generation. Even if you had an infinitely fast supercomputer, it wouldn’t speed up your brute force attempts, since you’re limited by the speed of the backend. Wherever Tea stores their images, that server has only a limited capacity for responding to requests, far less than the speed with which you can generate UUIDs. That’s a hard cap - you won’t try guesses faster than that.

          • Constant Pain@lemmy.world
            link
            fedilink
            arrow-up
            11
            arrow-down
            1
            ·
            6 hours ago

            The powerful enough computer doesn’t exist, and will not exist for some time. And even if it exists, it can’t query the web server fast enough to have meaningful effectiveness.

            So, for all intents and purposes, it’s impossible. Period.

            • bamboo@lemmy.blahaj.zone
              link
              fedilink
              English
              arrow-up
              5
              ·
              5 hours ago

              Thank you for bringing sanity to this thread. At this point, I have to assume that this person is trolling? That or they’ve been vibecoding too long?