13 ms·
This is just one of many reasons one should switch from RSA to ECC.
by lisper 10y ago
This is just one of many reasons one should switch from RSA to ECC.
- dc352 10y agoI don't say switching is a bad idea but not sure it would prevent this kind of attack :)
- tptacek 10y agoThe secret scalar in ECDH doesn't need to be prime; it's just a random number.
- lisper 10y agoYes, exactly. To generate an RSA key you need a complex mechanism wherein many vulnerabilities can hide. To generate an ECC key all you need is a source of entropy. To be sure vulnerabilities can hide in entropy sources as well, but it's a lot easier to swap out an entropy source than it is to swap out a random prime generator.
- api 10y ago... and yet again this shows that complexity is inherently BAD in secure systems.
- dc352 10y agoECDH doesn't give you authenticity, i.e., you can't get a certificate for that. When you look at ways to generate ECC keys, there are classes of vulnerable numbers for which you need to test.
- tptacek 10y agoCould you be more specific about the "vulnerable numbers" we're talking about? The curve base point is a parameter for ECDSA as well. I agree that any protocol that required you to generate curves or even base points on the fly would have similar concerns, but we generally don't use those kinds of protocols.
- dc352 10y agoMy err - rubbish wording as I was thinking primes and talking curves. Still, while I'm punching here a wee bit above my weight - I should read the 186-4 again - there are some tests to verify the strength of the prime. Is it enough for a similar classification? - I don't know...
- tptacek 10y agoThe prime field is part of the definition of the curve. Everyone interoperating on that curve shares it. It may have been generated weirdly --- the Curve25519 prime sure was! --- but all that tells you is what curve they're using. It's true that the particular curves a system supports could be a kind of fingerprint, but that's a banal observation, equally true of the ciphers they support, or the compression algorithms.
- zeveb 10y ago> ECDH doesn't give you authenticity, i.e., you can't get a certificate for that. ECDH doesn't, but ECDSA does. That's why we have Curve25519 & Ed25519, respectively.
- kuschku 10y agoAs long as you don’t use any of the NIST curves, yes.
- lisper 10y agoRight. I was going to add, "and curve25519 in particular" but decided that sounded too biased.