[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]

Re: PAM questions (fwd)

Joseph S. D. Yao wrote:
> How about:
> 	a support module that marks the start time of the next module
> 	the "active" module that needs a fixed time "in" it
> 	a support module that waits until a certain time (or a random
> 		time, fixed distribution) from the START TIME of the
> 		previous module - the previously marked time

I guess the time should be measured in CPU time actually used and not
real time elapsed. This seems to be a good way of satisfying things in
a way that leaves libpam unchanged.

Sun's position (which necessarily seems to reflect that of all of the
subscribed DCE "vendors" too) seems to be that it is most appropriate
to leave the pausing to the application's discretion.

It is not at all clear to me that this is the right thing to do.
Perhaps someone out there can see why this would be best?

My personal opinion, again, is that the libpam pam_authenticate()
etc..  function should really take care of this. Do people have any
objections to me putting such a feature (perhaps as a compile-time
option) into libpam that will ensure that failed authentication
etc. calls take some time?



[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index] []