you are viewing a single comment's thread
view the rest of the comments
[–] 33 points 1 day ago (20 children)

What's this a reference to?

  • source
  • hideshow 20 child comments
  • [–] 114 points 1 day ago (5 children)

    https://www.theregister.com/security/2026/10/20/indian-certification-authority-emudhras-private-keys-compromised-one-of-them-was-the-number-4/5304312

    It was discovered that one of India's Trusted Certificate Authorities, eMudhra, tasked with signing website certificates to verify authenticity of HTTP traffic was using the number "4" as the signing key along with 7 other keys less than 250. The key is supposed to be a unique number hundreds of binary digits long, but with such a simple key anybody can spoof a certificate that they could claim was signed by eMudhra and CA-checking lists would accept that.

  • source
  • parent
  • hideshow 5 child comments
  • [+] -38 points 23 hours ago (3 children)
  • [–] 65 points 23 hours ago (3 children)

    4 has been breached. Update your browsers! We are using 5 now :)

  • source
  • parent
  • hideshow 3 child comments
  • [–] [S] 32 points 1 day ago* (last edited 1 day ago) (1 child)

    How do I begin? Basically, an India based company that does TLS certification published some comically bad private keys. One of which was just the character 4. Update all of your browsers btw, desktop and mobile. This is a very major low level security breech of the entire internet.

    It may take Linux distros days to patch it, and this includes millions of servers.

  • source
  • parent
  • hideshow 1 child comment
  • [–] 4 points 4 hours ago*

    Firefox, likely Chrome, and maybe other major browsers update certificate revocations independently of the main app. Things are more complicated with various command-line apps and daemons, programming languages and libraries, which may require updates to system-wide packages or, worse, libraries for individual programs (ahem Python).

    Also, private keys aren't published, that's their defining characteristic. Public keys are published, and of course they are used to check signatures made with particular private keys — in this case the ones generated by the security researcher.

  • source
  • parent
  • [–] 23 points 1 day ago (4 children)

    An ssl certificate provider used 4 as their private key. This should be hundreds of characters long. Somebody figured this out and was able to issue their own certificates based on it. This results in loss of trust in their certificates which is a big deal.

    It is compared as bigger than one of the largest security issues in IT the last couple of years since it forces everybody to patch their shit to not trust this certificate chain. And lots of stuff aren't patched that easily.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 13 points 1 day ago (3 children)

    I bet they used 0000000000…004

    See, hundreds of digits long!

  • source
  • parent
  • hideshow 3 child comments
  • [–] 9 points 23 hours ago (2 children)

    I mean …. Yes, that would be secure.

    Sorta.

    It’s be dumb, and a pattern based attack could potentially solve it easily … but technically much more secure than 4

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 17 hours ago (1 child)

    I get what you mean, but no, it would not be secure. The private key in question is really used as a number in a calculation, not as a string of digits. Leading zeroes would just be ignored.

  • source
  • parent
  • hideshow 1 child comment