Re: RFC: Alternate patch to have true new-style rc.d scripts in ports(without touching localpkg)

> From: sthaug_at_nethelp.no
> Date: Tue, 17 Aug 2004 15:11:39 +0200
> Sender: owner-freebsd-current_at_freebsd.org
> 
> > I think that a better way would be to find an elegant method of 
> > allowing /usr/local/etc/rc.d to participate in rcorder.  I've got plenty of 
> > ideas about how to do this without breaking the filesystem dependency, but 
> > I'll wait to see what -current and -hackers come up with.  I am sure that 
> > their method will be cleaner.
> 
> I would much prefer to keep ports out of /etc (or out of the root file
> system in general). I agree with the point made by several others that
> the clean separation of base system and local mods is one of the great
> strengths of FreeBSD.
> 
> Since /etc/rc.d/local (or similar) has been proposed:
> 
> - Why cannot /usr/local/etc/rc.d be used with rcorder if /etc/rc.d/local
> is okay?

What if a startup script need to do something BEFORE /usr is mounted? My
case in point is Tobias Roth's profile.sh script. Since this is a script
only with no programs in /usr/local, the only way to put it into a port
is to allow it into the root filesystem in some place where it can be
run before any filesystem is mounted. (It is dependent on ly on fsck.)

> - If the argument is that /usr/local is not available: Okay, but in that
> case you won't be able to start the ports anyway, since they are located
> somewhere under /usr/local.

Nope. The man page could be in /usr/local/man, but the startup script is
the entire port, so there is nothing in /usr/local needed to run it. I
can conceive of other ports that DO have executables in /usr/local that
need to do some type of initialization before /usr is mounted.
-- 
R. Kevin Oberman, Network Engineer
Energy Sciences Network (ESnet)
Ernest O. Lawrence Berkeley National Laboratory (Berkeley Lab)
E-mail: oberman_at_es.net			Phone: +1 510 486-8634