1

I came across this article last night regarding SSL certificates and issues with the serial number not being long enough. Here is a status report from GoDaddy.

According to their csv file, our sites are not affected, but I did find that our serial number for one of our sites is 63 bits instead of 64 like it should be. The other two are 64 bits.

According to the research I’ve done, the impact this would have is that GoDaddy might revoke the certificates, but until they do that, there’s very low security risk with this issue.

How can I verify if this will be an actual issue for us?

Here is some information on our certificates. I didn't post specific details of my site, so if you need additional information, please let me know and I'll update the question.

Version V3
Signature algorithm sha256RSA
Signature hash algorithm    sha256
Issuer  
  CN = Go Daddy Secure Certificate Authority - G2
  OU = http://certs.godaddy.com/repository/
  O = GoDaddy.com, Inc.
  L = Scottsdale
  S = Arizona
  C = US
valid from  ‎Wednesday, ‎April ‎25, ‎2018 8:25:31 AM
valid to    ‎Monday, ‎May ‎25, ‎2020 8:53:55 AM
subject 
  CN = www.example.com
  OU = Domain Control Validated

I apologize in advance for my lack of knowledge of this. I am a developer who has also been tasked with managing our web server as well.

1 Answer 1

2

It's perfectly valid for a serial number to be shorter than 64 bits; this can happen because the certificates use a variable length encoding (DER). The issue in this incident is whether the serial numbers contain at least 64 bits of entropy, i.e. whether they were generated from at least 64 bits of randomness. This can legitimately result in numbers where the highest bits are zero. For some certificates however, GoDaddy (and others) apparently used a method that caused the highest bit to be predicatable, thus effectively destroying one bit of entropy.

The CSV you linked is called "Final Serial Number List" in the bug report, which indicates that you're unaffected if your certificate is not included. However, the discussion about this group of incidents is still ongoing, and as far as I understand, there is still disagreement about the correct interpretation of the relevant standard (the Baseline Requirements). You can follow the debate here: https://groups.google.com/d/msg/mozilla.dev.security.policy/S2KNbJSJ-hs/F8AS4MNVCAAJ

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .