Re: OpenSSH & S/Key information leakage

[Robert Bihlmeyer <robbe () orcus priv at>]

flaps () dgp toronto edu (Alan J Rosenthal) writes:
[quote reordered & trimmed]

> A login prompt for a non-account looks like this:
>
>       login: flomp
>       otp-md5 175 at2078 ext
>       Response: 
>
> So far, so good.  But press return once or twice to get "Login incorrect"
> (or make a new conection), and then do
>
>       login: flomp
>       otp-md5 220 at0624 ext
>       Response: 

Oops.

But if a system mandated a common seed for all accounts (one that changes
regularly) all login attempts will show that. For the sequence number, the
fake response could use a number that is dependent on the login name and
the seed, e.g. the lower bits of MD5(login + seed + host-secret)

Would using the same seed for all (real) accounts lose us security? My
intuition is no, but this needs to be thought over more.

> If OPIE didn't tell you the password number, for example, it would be quite
> hard to use.

You can keep the last used number on a slip of paper in your wallet
(according to one's threat model and set-up keeping the OTPs there as
well may be appropriate).

Logged-in users can query their seq#, if they are in doubt. The login
process can also tell you the current number if you try to use an
older OTP. This does not affect security, because an attacker that
knows an older OTP obviously has snooped on a previous successful
login attempt and thus knows that this account exists.

-- 
Robbe

Attachment: signature.ng
Description: