Bug#727708: systemd and upstart, a view from a daemon Debian maintainer
Ian Jackson <ijackson@chiark.greenend.org.uk> writes:
> I found configuring upstart to be utterly trivial. There was little
> opportunity for error. More guidance in debian-policy would be a good
> idea, including perhaps a reference to some example packages.
I have a much longer message that goes into detail about what I found
inadequate in the packaging side of the documentation and the upgrade
issues that I encountered, which unfortunately is being rejected by the
Debian email system as malware, probably due to some broken virus
definition. (I have a Debian RT ticket open about this.) I'm happy to
send this to anyone directly until we can figure out what's keeping it
from making its way through the BTS and mail system.
I had the opposite experience that you did: I found writing unit files for
systemd trivial and upstart considerably more complex. However, both can
benefit from some additional guidance, and in neither case do I think the
issues were particularly serious.
> 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.)
It's worth noting that if you're using dh (which I was), this is trivial,
and just involves a build-dependency on dh-systemd plus adding it to the
dh invocation.
> Also, the approach to the systemd integration introduces a new runtime
> package dependency on "init-system-helpers", which despite its
> generic-sounding name actually contains only helpers for systemd and is
> maintained in Debian by the systemd maintainers.
The maintainers of the package have openly offered any other useful
helpers for any other init systems a home in that package. I think it's
more due to an accident of history and existing usage that the bit of
necessary supporting glue for upstart ended up in lsb-base instead of
init-system-helpers.
It's worth noting that the systemd support in update-rc.d is substantially
better than the upstart support. (There's more about this in my other
message that's being temporarily blocked.)
--
Russ Allbery (rra@debian.org) <http://www.eyrie.org/~eagle/>
Reply to: