On 2013 Jan 11, at 15:13 , Corey Quinn wrote:

> Right.
> 
> The problem with that is that as the admin, you can't enforce that users set 
> a passphrase, or remove it later in time from the private key.  The idea with 
> two factor is that they definitely have the device / application you sent 
> home with them, without being able to disable it for convenience.
> 
> You can also add the passphrase to your ssh keyring so you don't have to type 
> it each time, which is neither here nor there. :-)

You can also write your SecurID PIN on your keyfob, or put on a sticky on the 
back of the credit card style token.  That's how I used to see the tokens 
literally handed out.

There is no security so foolproof as to avoid the problem between keyboard and 
chair.  

A SecurID token is nothing but a fancy way of validating a seed.  That seed is 
rehashed each minute and the user has to type in the hash (if you'll let me 
abuse the mathematics behind the PRNG), but it is no longer a secret how the 
seed translates into the displayed number.  The whole "something you have that 
cannot be copied" is a myth today.  The real goal of multi-factor 
authentication is to limit the damage of a compromise.

The more I work in security, the more I am discouraged by the options in 
authentication.  We are facing increasingly sophisticated attacks, even if done 
by script kiddies, and a repeat of the SecurID attack seems likely in the 
future, along with techniques to bypass the longer PIN.  At this point, I feel 
that getting user cooperation is our best bet for security.  If the users want 
to cooperate, if we make it not a burden for them, they'll follow good 
practices much more readily, and even help us with efforts to validate that 
they are complying.  If we make it a nuisance, like SecurID does with the one 
time use (for users who need to login to 20+ systems at once to repair a major 
app outage), they will bypass things just to get their job done.

Sorry for the rant.

----
"The speed of communications is wondrous to behold. It is also true that 
speed can multiply the distribution of information that we know to be 
untrue." Edward R Murrow (1964)

Mark McCullough
[email protected]




_______________________________________________
Discuss mailing list
[email protected]
https://lists.lopsa.org/cgi-bin/mailman/listinfo/discuss
This list provided by the League of Professional System Administrators
 http://lopsa.org/

Reply via email to