Re: [Full-disclosure] Leveraging pam_env to steal DSA keys

[Peter van Dijk <peter@xxxxxxxx>]

On May 31, 2011, at 12:48 AM, paul.szabo@xxxxxxxxxxxxx wrote:

>> ... http://7bits.nl/projects/pamenv-dsakeys/pamenv-dsakeys.html
> 
> Seems to me that CVE-2010-3435 may allow users to determine also:
>  password in /etc/lilo.conf
>  secret in /etc/bind/named.conf /etc/bind/rndc.conf /etc/bind/rndc.key
>  bits of /etc/ssh/ssh_host_dsa_key /etc/ssh/ssh_host_rsa_key
> which should all be protected.

- lilo.conf commonly has whitespace around '=', pam_env does not tolerate that
- bind configs don't even use '=' and are often indented, pam_env does not 
tolerate indents
- RSA appears to be uninteresting in that the amount of bits we can lift is not 
sufficient to make an attack feasible (this is in the article!)
- the DSA host key certainly is a target

If I understand correctly, Debian (and, I presume, Ubuntu) put this bug on low 
priority precisely because there were very few practical applications that they 
knew of.

Cheers, Peter
_______________________________________________
Full-Disclosure - We believe in it.
Charter: http://lists.grok.org.uk/full-disclosure-charter.html
Hosted and sponsored by Secunia - http://secunia.com/