Date: Mon, 26 Dec 1994 06:57:59 -0600 From: Peter da Silva <peter@bonkers.taronga.com> To: roberto@blaise.ibp.fr (Ollivier ROBERT) Cc: jgreco@brasil.moneng.mei.com (Joe Greco), hackers@freebsd.org Subject: Re: /etc/rc.shutdown (First shot) Message-ID: <199412261258.GAA25259@bonkers.taronga.com> In-Reply-To: Your message of "Mon, 26 Dec 94 13:10:32 %2B0100." <9412261210.AA09778@blaise.ibp.fr>
next in thread | previous in thread | raw e-mail | index | archive | help
> 1. nothing at all -- old BSD way > 2. the half-way now (only shutdown run the script) -- what *I* have > 3. init does all the work and halt/reboot/shutdown all make init run the script ... > > I need your opinion, folks. > > present: one for 3 (Peter) and one for 1 (Joe). I'm changing my vote to 2. Joe reminded me of "uadmin 2 0": you do need a backdoor. Though I'd rather vote for "4", which is kind of a cross between 2 and 3. That is, you have a program that does what halt does. You have init call it. The normal programs just tell init to shut down and let init worry about exactly how that works. That way you don't replicate this logic throughout the system (DEC OSF/1 is REALLY bad about that). If you want that program to be "halt" as badly as Joe does, so be it. I think a little bit more luser-proofing is in order, but so long as the logic's there and documented that's fine.
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199412261258.GAA25259>