Why different keys for -testing and non-testing?

[Panu Matilainen]

On Sat, 17 Jan 2009, Steve Grubb wrote:

> On Saturday 17 January 2009 11:59:09 am Jesse Keating wrote:
>> We should also make import of keys an auditable event.
>>
>> Are not all rpm actions audited?
>
> No. What I'm talking about is perhaps defining a specific audit event type
> that would signify that a key was imported and where it came from. I've seen
> cases where rpm tries to download keys from the network. This is one of the
> few security sensitive actions that is not put into the audit system.

FWIW, rpm >= 4.6 will never try to download keys from the network unless 
explicitly told to do so with --import. Only 4.4.x has the "feature" to 
automatically try and fetch unknown keys from public keyservers, and from 
4.4.2.1 onwards that's disabled by default (and every vendor has disabled 
it in their packages for affected versions anyway)

 	- Panu -