Bug#727708: systemd and upstart, a view from a daemon Debian maintainer
Steve Langasek <vorlon@debian.org> writes:
> On Sun, Dec 29, 2013 at 12:42:24AM +0100, Vincent Bernat wrote:
>> Ian Jackson <ijackson@chiark.greenend.org.uk> writes:
>>> The package maintainer scripts exposed more complexity too. It was
>>> necessary to add new systemd-specific calls to "deb-systemd-helper".
>>> The boilerplate required here was too much to simply include in my
>>> existing scripts, so I had to switch the package to using
>>> debhelper-generated maintainer scripts. (This is of course a good
>>> idea in the long run, but it would be better to require less
>>> yak-shaving.)
>> Most of this complexity is because systemd's maintainers are also
>> trying to fix the problem with daemon automatically starting after
>> install. They would have used triggers otherwise.
> What "problem" do you refer to here? Starting daemons automatically on
> install is a policy-driven expectation, not a problem.
The purpose of deb-systemd-helper is to set up the proper links for
systemd unit files on a system where systemd is not installed, so that if
you later install systemd, the right configuration (echoing the sysvinit
configuration) is already in place.
I think that's the installation that Vincent is referring to, not the
installation of the package on a system with systemd already running.
You can't use triggers in the systemd package to handle this because the
whole point is that the systemd package is not necessarily installed. If
we standardized on systemd as the required init system, much of that
complexity could be removed; it's there to be a good Debian citizen and
interoperate with the rest of the archive, including switching back and
forth between different init systems.
--
Russ Allbery (rra@debian.org) <http://www.eyrie.org/~eagle/>
Reply to: