HOWTOs please! [LWN.net]

HOWTOs please!

Posted May 13, 2008 22:28 UTC (Tue) by man_ls (subscriber, #15091)
In reply to: HOWTOs please! by jch
Parent article: Cryptographic weakness on Debian systems

Great info, thanks!

Although I don't understand step 3: "Generate new host keys for ssh". I'm purging my system of DSA keys, and I find that by default my systems identify using RSA keys (as seen in ~/.ssh/known_hosts). This part should be safe then. Why regenerate all host keys? This will only create a new set of (still insecure) DSA keys, which as it seems are not used anyway.

Yes, this specific step 3 leads to a lot of work I'm too lazy to do: erasing all known_hosts files and recreating them. And this is on a small home LAN; on a big network I can imagine it must be a real pain.

(Log in to post comments)

HOWTOs please!

Posted May 13, 2008 22:40 UTC (Tue) by joey (subscriber, #328) [Link]

The new host keys won't be insecure if you've upgraded openssl before generating them.

Tomorrow's version of openssh-server (in unstable) will automate the host key regeneration, as
well as blocking authentication using weak keys, which will remove some of the pain if you can
stand to wait 8 hours for it to reach the mirrors, and are running unstable. (I hope these
enhancements will later be pushed into stable?)

This document has fairly complete instructions BTW, until the official www.debian.org page
goes up: http://wiki.debian.org/SSLkeys

HOWTOs please!

Posted May 14, 2008 1:34 UTC (Wed) by bboissin (subscriber, #29506) [Link]

> What's wrong with DSA keys anyway?
Even if they aren't weak, they are compromised if they were used (the ssh client) in a
affected system (debian, ubuntu). According to the DSA this is due to the fact that "Digital
Signature Algorithm relies on a secret random value used during signature generation".

HOWTOs please!

Posted May 14, 2008 11:12 UTC (Wed) by nix (subscriber, #2304) [Link]

Well, yes, but that's true of just about any encryption system's keys. The whole point of them
is that they're meant to be unpredictable (hence random). If the randomness is bad, so is the
key, always.

HOWTOs please!

Posted May 14, 2008 13:40 UTC (Wed) by tialaramex (subscriber, #21167) [Link]

Presumably (please, someone who remembers how this actually works chime in!)

Suppose you create two signing keys, R (using RSA) and D (using DSA) on your nice RHEL 4
machine

It seems like Debian's security people are saying that if you copied these good keys to a
Debian system, and then used them to sign documents, the R key is still fine but the D key is
compromised by the signing process, due to it having poor entropy. That is, a sophisticated
attacker (or a script kiddie with software developed by someone else) could use your old
signatures generated on Debian systems to figure out your DSA private key.

It's certainly true that DSA's signature method explicitly requires unique cryptographically
secure random numbers for each message to be signed. But I don't know whether these numbers
protect the DSA private key, or just the signature itself. If the former, then Debian's
advisory is correct. If we don't know, then we must act as if it is correct and invalidate the
affected keys.

Although DSA and RSA often appear interchangeable to us as end users, they are quite different
in terms of their mathematical properties, so it could easily be true that this problem
affects only DSA.

Re: DSA vs RSA

Posted May 15, 2008 10:21 UTC (Thu) by ldo (subscriber, #40946) [Link]

As I remember, DSA was invented back when the US government was trying to restrict the use of strong cryptography. The key thing about it was that it was supposedly usable only for digital signatures, not for encryption. This was later proven to be false.

And yet people still use DSA today. So why bother any more? Why not just stick with RSA for both encryption and signing? Isn't this compromise a reason to stop using DSA altogether?