Date: Tue, 2 Feb 1999 09:52:55 -0800 (PST) From: David Wolfskill <dhw@whistle.com> To: freebsd-hackers@FreeBSD.ORG Subject: Re: more modular rc/init/uninit system... Message-ID: <199902021752.JAA16952@pau-amma.whistle.com> In-Reply-To: <XFMail.990202171342.doconnor@gsoft.com.au>
next in thread | previous in thread | raw e-mail | index | archive | help
>Date: Tue, 02 Feb 1999 17:13:42 +1030 (CST) >From: "Daniel O'Connor" <doconnor@gsoft.com.au> >> That was what drove my idea to use a makefile; you could write the >> start/stop scripts and express the dependencies in the makefile; the >> start/stop scripts would be useful on their own. The disadvantage is >> that you now have to edit the Makefile to add or remove something; >> which we were trying to avoid. >Yes, which could be problematic to do automatically :) That is one respect in which the SysV approach is relatively easy: by blacing a (link to the) script in an appropriate directory with an appropriate name, it gets started/stopped at a (relatively) well-defined point in the run-level change. So to avoid tinkering with the insides of a Makefile, have whatever implementation is used read information about what services are wanted from a file or a directory. (The directory is relatively appealing, in that with a suitable implementation, it could be adequate to merely "touch" a file with a name of an appropriate pattern, and thus minimize the probability of clobbering another service. On the other hand, if it were a file, I'd always use RCS whenever I made a change anyway, just as I do with everything else. And the file could have some explicit ordering specifications... without regard to naming conventions.) david -- David Wolfskill UNIX System Administrator dhw@whistle.com voice: (650) 577-7158 pager: (650) 371-4621 To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199902021752.JAA16952>