Hi,

I'm a bit concerned about the protection afforded by the PEM format to
private keys against offline brute-force attacks. PEM seems to use a
decent KDF, but uses a fixed iteration count of 1. Am I correct in my
understanding that this cannot be changed without breaking the format?
PEM is pretty convenient, and way more widely used than (say) PKCS12 so
it makes sense to defend it well.

One might argue that if an attacker has a copy of a private key file
then they have already won. I think this is true at the limit, but IMO
it is still worth protecting them - from the moment they are taken it is
a race between the attackers ability to brute force and the key owner's
ability to detect and revoke/replace the keys. Anything that handicaps
the race in favour of the defenders is worth considering

-d
______________________________________________________________________
OpenSSL Project                                 http://www.openssl.org
Development Mailing List                       [email protected]
Automated List Manager                           [email protected]

Reply via email to