[ On Wednesday, March 15, 2000 at 14:27:16 (-0500), der Mouse wrote: ]
> Subject: Re: rc.d
>
> A disaster when, say, I'm trying to run an experimental replacement for
> one of those services and the start script notices the stock one isn't
> running and tries to start it, not able to tell that my replacement is
> running instead.
> 
> Remember what I said about the OS-hacker niche?  Guess what perspective
> this message is being written from. :-)

I think you're over-reacting -- a lot.  Just because scripts exist to
manage the startup and shutdown of daemons doesn't mean you have to use
them.

Your example of wanting to start a new test process that works in
concert with a number of other dependent processes is certainly not a
scenario where I would use the rc.d script (unless of course the entire
subsystem did have to be shut down and restarted to effect some
initialisation required).  Just kill the damn process and start your new
one!

Where such problems would require different solutions is if the program
was running under a SysV-style "init" or SAC monitor.  Obviously in such
a scenario you'd probably want to either shuffle around some binaries or
edit the monitor config file before manually killing the desired process
and thus allowing the monitor system to restart the appropriate test
program.

Note that inetd isn't much different than a SysV init in this respect....

-- 
							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>