[ On Thursday, May 11, 2000 at 20:52:55 (+0200), S.P.Zeidler wrote: ]
> Subject: Re: The new rc.d stuff... [now rc.conf]
>
> If you have to locate the source of and fix a problem on a "foreign"
> machine, i.e. one not usually administered by you, where will you have
> less trouble finding out what that machine actually boots with and with
> what options it gets started, rc.d or rc.conf?

I've worked on so many different kinds of machines that it *never* makes
any difference to me when I'm approaching a problem on a "foreign"
machine -- I treat each and every one as unique and start by always
verifying any assumptions I make about how it is set up.  Certainly
things can go faster if my assumptions are correct, but no matter what I
won't curse too much under my breath unless the system looks on the
surface to have one kind of setup but is in fact entirely different
under the hood (or unless it's so convoluted and undocumented as to
cause me to go in circles).

Note that if I hadn't been directly aware of this migration of NetBSD to
rc.d but instead had stumbled upon it in the wild I would curse it a bit
because it does look a bit like an antiquated SysVr2 setup but turns out
to be entirely different once you start looking inside the files.

-- 
							Greg A. Woods

+1 416 218-0098      VE3TCP      <gwoods@acm.org>      <robohack!woods>
Planix, Inc. <woods@planix.com>; Secrets of the Weird <woods@weird.com>