[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#1039102: debian-policy: make systemd units mandatory for packages shipping system services



On Sun, 30 Jul 2023 23:12:21 +0200 Bill Allombert <ballombe@debian.org>
wrote:
> On Sun, Jul 30, 2023 at 08:22:54PM +0100, Luca Boccassi wrote:
> > On Fri, 30 Jun 2023 00:04:29 +0100 Luca Boccassi <bluca@debian.org>
> > wrote:
> > > This happened a few days ago and nobody complained (if we ignore
> > > grumblings because of the fact that I used lintian.debian.org
queries
> > > which are hopelessly and silently out of date, sigh), and bugs
are
> > > filed, there have been a couple of uploads too already.
> > > 
> > > Can we go ahead, or do you want to wait a specified amount of
time?
> > > If
> > > so, how long (just so that I know when to come back)?
> > 
> > Hi, monthly ping. Anything I can do to move this forward?
> 
> I consider this proposal to be premature. Policy should document
current
> practice, and I do not think this proposal does that.

Not really - apart from the fact that it's been 10 years or so, and if
after a decade something can still be 'premature' then we'll all be
long dead before anything becomes 'mature'. More importantly, the clock
is ticking, and anything not shipping a unit file but still expecting
to work will break in the near future. So the policy change right now
would be correct - current practice is to ship at least a unit file for
anything shipping a service, and not doing that is a bug, of which the
severity is going to increase shortly, as the affected package will
stop working in the default scenario.

> It would it more useful to help maintainers adapt their script to
conform
> to this first, and change policy later.

The help already arrived - bugs have been filed notifying of the
required changes. If anybody has the time and interests in doing
anything more than that, that's great, the bug list is linked earlier
in the thread.

-- 
Kind regards,
Luca Boccassi

Attachment: signature.asc
Description: This is a digitally signed message part


Reply to: