Re: kill -9 <pid of X>

Rob Hagopian (hagopiar@vuser.vu.union.edu)
Fri, 14 Aug 1998 00:31:27 -0400 (EDT)

I can think of many more ways than that to make the system FUBAR...
It seems to me that the death of init should be dealt with by the kernel
in a sane way for very good reasons other than someone accidentally typing
'kill -9 1' (ie. when it gets that very rare oops). But I'm not going to
code it anytime soon so any more disussion on that front isn't going to
come from me. :-)
-Rob

On Fri, 14 Aug 1998, MOLNAR Ingo wrote:

> On Thu, 13 Aug 1998, Rob Hagopian wrote:
>
> > I missed that message along the line... For pure cleanliness, I think that
> > 'kill -9 1' should be blocked in userspace tools. It does make much more
> > sense there.
>
> the kernel assumes on some places that a process _always_ has a parent. So
> it can always blindly dereference p->p_pptr. So if we kill init, we will
> immediatelly have a nasty set of oopses. It's also a matter of how mildly
> Linux/hardware bugs escalate when we get an oops. (if a very rare oops
> happens to hit init, the system dies horribly with a storm of oopses,
> possibly making the identification of the former oops harder)
>
> -- mingo

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.altern.org/andrebalsa/doc/lkml-faq.html