In some email I received from Luke Mewburn, sie wrote:
[...]
> A new system should have the following attributes:
> 	* Separate scripts for each function, which support
> 	  `start' and `stop' type functionality.

How about `status' too ?
0 = started successfully
1 = failed to start
2 = starting (not completed)
3 = not configured to be started (/etc/rc.conf turned this script off
    but it's still in /etc/rc3.d because mkrc hasn't been rerun)

"reload/restart" would be another useful one.

I'd consider both of those as "must have" before adding this to
NetBSD, based on what's in the marketplace today.

> 	* The ability to generate /etc/rc and /etc/rc.shutdown
> 	  from `/etc/init.d/*'.

Good.

> 	* Implementers should be able to add extra actions to a
> 	  script (e.g, `reload', `debugon', etc) fairly easily.

Any `additions' should be standard across all scripts which are
NetBSD derived (vs 3rd party) - even if they're null ops so that
invalid actions always return errors.

[...]
> 	NOTE:	During development this directory was called `/etc/rc.d'.
> 		It was changed to `/etc/init.d' because:
> 		    a) prior art (cf. SYSV), and people are used to running
> 			/etc/init.d/foo start

Or /etc/rc2.d/foo start
And what about /etc/rcS.d ?

> c) /etc/rc.sh
> 
> 	A script which effectively does:
> 		for i in `rcorder /etc/init.d/*`; do
> 			$i faststart
> 		done
> 	(`faststart' is like start but without the check for
> 	running processes).
> 
> 	If a user wants to use this method (instead of having to
> 	rebuild /etc/rc when necessary), just symlink /etc/rc.sh
> 	to /etc/rc.

"faststart" seems like a bad idea to me.  It seems to me something
like what you'd use in testing, but that's about it.  Or maybe you'd
use it when booting a vax so that it boots `today' ;)  Granted this
is pretty much what we do today, but who's to say that's something
to brag about ? :->

[...]
>     ====
> 	# PROVIDE: mountd
> 	# REQUIRE: beforemountlkm network portmap

Is that meant to mean "before mountlkm" or "before mount lkm" ? Or is
"beforemountlkm" really the name of a single script ?  If it's the latter
then the name just sucks.  If either of the former, split it up.

On the whole good, but I've got one major criticism:

in an effort to not piss anyone off (too much) there's too much fence
sitting being done.  I don't mind the /etc/rc still being supported
(the everything in one script version) but the rest is of concern.
There's an /etc/rcX.d which gets used sometimes and an /etc/init.d
which also gets used sometimes.  It's possible to have a /etc/rcX.d
populated but then ignored by a script which independantly interrogates
scripts in /etc/init.d.  From a featurism point of view, sure, it's
great that we've got all these features, but I think these two are
competing with each other and offer no real long term benefit.  We
need to pick one and go with it.  I think this is trying to be everything
to everyone and ultimately we'll lose.

Darren