[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Runlevels vs. facility names (Re: Facility names for init scripts)



On Thu, 12 Jul 2001, Thorsten Kukuk wrote:

> On Mon, Jul 09, Bodo Meissner wrote:
>
> > I agree with aj. As a software manufacturer I'm not primarily interested
> > in runlevel numbers, but I want to run my daemons (facilities) when
> > specific preconditions are met.
>
> The runlevel numbers are also defined in the LSB. So, $network,$multiuser
> should give always identical numbers on a LSB conform system. I don't see
> a reason to add this defines, because the result should always be the
> same.

True. The runlevel numbers are currently defined in the LSB. Maybe there
has been some discussion about that which I missed because I joined the
list too late. In this case, please, point me to an archive of this older
discussion.

Why do we require all compliant distributions to use runlevels in the same
way? There might be reasons for a different runlevel numbering scheme.
Or there might be other ways to do the system initialization without
runlevels at all. I think the definition of symbolic names (facilities)
would be more flexible.


Bodo



Reply to: