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

Re: next version .44 or .5?



Missed a few days, and a whole new 0.1 release.  ;-)

> > * if no one objects (and I have time), I will add a new libpam
> >   function: pam_fail_delay(pamh, microsec). It will take an
> >   argument of some number of microseconds, and will require that the
> >   following return from a libpam pam_XXX() function, that is
> >   unsuccessful, will sleep for this much time before passing control
> >   to the application.

> Again, I think that Ted's suggestion for an optional *random*
> delay with the standard deviation specified makes more sense
> than violating the standard.  It's *sufficient*, and pam_fail_delay
> is not *necessary*, so I vote that we stick to the standard.

I think that the suggestion was to allow implementation of the "delay
from start of module" feature [as opposed to the more common end of
module], rather than specifically disallowing a randomly distributed
delay time.  I think that both together would be good.  Neither would
violate anything in the standard.  The documentation should suggest
that overly random distributions might cause time-out failures.

Correct me if I am misunderstanding, please.

Joe Yao				jsdy@cais.com - Joseph S. D. Yao



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