># > Actually, I like Bill Studenmund's suggestion that the fork can
># > (optionally, and interruptably!)  block until a slot opens.
>#
># That could easily wedge completly, after all the process slot
># doesn't become free until after the parent has reaped the
># child (by calling wait).
>
>...which is why it would be optional and interruptable.
>
>It could also have its own timeout after which to terminate, but that's
>probably bloat we don't need.
>
>It seemed a good thought to me, that's all.

if it was unconditional, no one would like it, and it would get
removed.

if it was optional, it might be accepted, but...

...if it was the default (and you could opt out), i expect everyone
would opt out.  additionally, fork-bombs would only need to get a tiny
bit smarter to work around it.

...if it was not the default (and you could opt in), then fork-bombs
would not be abated by it, and the code would be mostly worthless.

-- 
|-----< "CODE WARRIOR" >-----|
codewarrior@daemon.org             * "ah!  i see you have the internet
twofsonet@graffiti.com (Andrew Brown)                that goes *ping*!"
werdna@squooshy.com       * "information is power -- share the wealth."