Late 90's? OpenSSH was released December 1, 1999, but the first real release featuring substantial OpenSSH-project written code, including support for protocol version 2 was released June 15, 2000. Also, the theoretical vulnerability you mentioned was protocol-based, not limited to OpenSSH. And what is the non sequitur about OpenSSL doing there? You missed some other completely unrelated look-alikes, such as OpenSS and OpenSSO.

Axe to grind or lazy?

That the OpenSSH dev's were not responsible for the flaw introduced by the modifications made by the Debian team.

What are the sources for this article? Is it, perhaps, this case from November?

http://www.cpni.gov.uk/Docs/Vulnerability_Advisory_SSH.txt

Yes, it's the case from November. The technical paper describing the research was published this week at the IEEE Symposium on Security and Privacy. The paper is now online here:

http://www.isg.rhul.ac.uk/~kp/SandPfinal.pdf

Cheers,

Kenny

Hi, I was referring to this attack by CORE in 1998, which allows insertion of data into a session:

http://www.coresecurity.com/content/An-attack-on-CRC-32-integrity

The article isn't clear that this wasn't OpenSSH flaw per se, although it certainly was based on the SSH code that later became OpenSSH. It was a flaw in the SSH protocol version 1 itself, not a particular implementation.

Since the attack Albrecht et al are discussing is a protocol problem, it is fair to compare it to other protocol problems SSH has had in the past. It's good they found this problem, and it's good to see the SSH protocol is getting more secure over time.

-- Nate