Re: CVE request: opencryptoki insecure lock files handling

[Raphael Geissert <geissert () debian org>]

On Thursday 20 September 2012 09:10:14 Tomas Hoger wrote:
> Ok, so I think we need 1 CVE for the two insecure temporary file uses,
> unless we want to split each temporary file issue under a separate
> CVE.  I don't believe there's a real need to assign CVE for 2.4.1
> (which did not improve things on systems with world writable /var/lock)
> or 2.4.2 (which re-opens the attack for pkcs11 group members on systems
> with restricted /var/lock, but improves things on systems with world
> writable /var/lock).

I think two ids is more appropriate given that the issue isn't fixed in 2.4.1 
for systems with world writable /var/lock. 2.4.2, on the other hand, covers 
boths scenarios (given that pkcs11 group membership is already considered 
root-equivalent.)

Regards,
-- 
Raphael Geissert - Debian Developer
www.debian.org - get.debian.net