Bug#551992: general: stopping squeeze chroot shuts system
On Fri, Oct 23, 2009 at 12:04:02PM +0200, email@example.com wrote:
> >> In a lenny chroot, on the same system, rc6.d/S90reboot calls 'reboot -d
> >> -f
> >> i' and that works like I'd expect; the chroot is stopped and the system
> >> keeps on running.
> >> In the squeeze chroot we have rc6.d/K09reboot and the same 'reboot -d -f
> >> -i'.
> >> I will check if the -f does the trick but I still think the behaviour
> >> should be the same for both lenny and squeeze. I also think that the
> >> ability to reboot a system by rebooting a chrooted environment shouldn't
> >> be possible so the bug remains.
> >> According xen or kvm; eventually we will make that change of course, but
> >> at the moment that's not something we're focussing.
> > Your script only calls all K* scripts (not S* scripts) when stop is
> > invoked. Hence in the system where reboot is S90, it won't call it.
> > Where it's K09, it will be called. I suspect if you rm K09reboot in
> > the squeeze chroot, it will fix the behaviour to what you want.
> > As far as rebooting from inside a chroot rebooting the system I don;t
> > know whether that's a bug or not. As far as I'm aware it has been the
> > way it's behaved for a long while...not sure what you can do to
> > prevent it even...
> Ah, never occured to me that would be considered as normal behaviour. To
> be honest, I still think it shouldn't be possible but I don't want to
> start a religious fight about that.
There is no fight about this, you're just expecting too much from the
wrong tool. What you want is simply not chroot, but containers, such as
openvz, vserver or lxc.