On Tue, May 23, 2000 at 01:33:01PM -0400, Andrew Gillham wrote:
> Manuel Bouyer writes:
> > On Wed, May 24, 2000 at 12:46:59AM +1000, Simon Burge wrote:
> > > That's the point - should there be an arbitary limit?
> > 
> > Ok. Does anyone agree that there shoudln't be any limit ?
> 
> Would it make sense to print a warning about "ridiculously large"
> numbers or something?  (e.g. greater than some number)
> I would hate to see a typo launch 1000+ nfsd processes.
> Not that a warning is necessary or anything, just an idea.

So maybe bump set the upper limit at 100 ?
Or, better, warn if the number given is larger than rlimit.maxproc.soft ?

--
Manuel Bouyer, LIP6, Universite Paris VI.           Manuel.Bouyer@lip6.fr
--