On 6/1/2011 2:29 AM, Dave Howorth wrote:
John Andersen wrote:
On 5/31/2011 3:40 PM, Anders Johansson wrote:
On Wednesday 01 June 2011 00:24:22 Edwin Helbert Aponte Angarita wrote:
I think this is a security issue. An unprivileged user that knows that the system is maintained remotely using ssh and, perhaps, sudo, could keep attempting to use sudo until they gets it. They would first need to log in as the same user the admin was using. sudo won't do that for all users. It just remembers that you have already authenticated once, and won't force you to do it again until some time later.
I think the point Edwin was trying to make was assume you ssh into a remote machine _that is being used_ by an authorized users, and you use that person's login and then issue a sudo command.
The regular user sitting at that remote machine can then issue another sudo without knowing root's login (allegedly).
The whole point of sudo is to delegate authority to particular users. So the only person who should know that user's password and be able to login as them is that person themselves. If you're able to login as that person you *are* that person, in sudo's security model. Indeed the default is as ubuntu has it, that there is no root password; it's your own password that you use to gain rootly power.
That setup increases auditability and accountability. It's always possible to trace exactly who made some change and to know that it was them that did it knowingly. At least, that's the theory.
Cheers, Dave
This is all interesting (if not redundant) background information, but how is it germane to the problem at hand, where according to one report, (Edwin's) transient state information leaked from one login session to another, but according to my tests NO SUCH LEAK OCCURRED? -- _____________________________________ ---This space for rent--- -- To unsubscribe, e-mail: opensuse+unsubscribe@opensuse.org For additional commands, e-mail: opensuse+help@opensuse.org