Bug#21585: PROPOSED] /etc/init.d/<script> need not be a conf
> On 30-May-99 Julian Gilbey wrote:
> > retitle 21585 [PROPOSED] /etc/init.d/<script> need not be a conffile if
> > actively maintained
> > thanks
> >
> > I second this proposal. I suggest that the paragraph in question (now
> > in section 3.3.5) be reworded as:
> >
> > _Do not_ include the `/etc/rc<n>.d/*' symbolic links in `dpkg''s
> > conffiles list! _This will cause problems!_ _Do_, however, either
> > include the `/etc/init.d' scripts in conffiles or handle them
> > fully in the maintainer scripts. (This is important since we want
> > to give the local system administrator the chance to adapt the scripts
> > to the local system--e.g., to disable a service without deinstalling
> > the package, or to specify some special command line options when
> > starting a service--while making sure her changes aren't lost during
> > the next package upgrade. Hence, if the maintainer scripts option is
> > taken, the scripts must be careful not interfere with any
> > user-modified scripts, and must take care to remove the file when
> > the package is purged.)
> >
>
> This would take away the admins ability to make modifications to the scripts.
> I would find this to be a bad thing.
Maybe the wording's slightly wrong then. Maybe I should say
"locally-modified scripts" rather than "user-modified scripts"?
I mean files such as /etc/init.d/xntp3 which are created in a
customised way by the postinst when xntp3 is first installed. Having
it as a conffile would not work, because then every time the package
is upgraded an unnecessary warning would be given. Rather, the
postinst checks to see whether the file already exists and creates it
if it doesn't. So no local modifications would be lost.
Does that make things clearer?
Julian
=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Julian Gilbey, Dept of Maths, QMW, Univ. of London. J.D.Gilbey@qmw.ac.uk
Debian GNU/Linux Developer, see http://www.debian.org/~jdg
Reply to: