FWIW, I fail to see where anyone explained anything like what you are
saying. So far, I've seen bad example with thttpd. Generalizing on
basis of thttpd behaviour isn't correct, IMHO.

For a while, assume misconfigured system on edge of collapse due
to filled process slots. Do you think it's better to have huge
number of processes busy looping calling fork(), or have the
processes frequently napping and trying fork() just occassionally?
Which case is likely to thrash less and recover more quickly? In
which case would admin have easier job to fix the situation?

Jaromir

Perry E. Metzger wrote:
> Michael Graff <explorer@flame.org> writes:
> > Jaromir Dolecek <jdolecek@netbsd.org> writes:
> > 
> > > The sleep thing is smart & simple solution to tough problem. I don't
> > > see how it could cause any new problem. So I currently don't see why
> > > to not have it there.
> > 
> > It also only occurs in the case where a process is trying to exceed
> > its limit.
> 
> You mean like, perhaps, a system already on the edge of collapse where
> the delays will make the collapse worse, as has already been explained
> by others?

-- 
Jaromir Dolecek <jdolecek@NetBSD.org>            http://www.NetBSD.org/
-=- We should be mindful of the potential goal, but as the tantric    -=-
-=- Buddhist masters say, ``You may notice during meditation that you -=-
-=- sometimes levitate or glow.   Do not let this distract you.''     -=-