Autor: zeitgeisteater Data: A: devuan-dev Assumpte: Re: [devuan-dev] [DNG] Debian Buster release to partially drop
non-systemd support
On 2018-10-17 15:03, Daniel Taylor wrote: > On 10/17/18 9:58 AM, KatolaZ wrote:
> > On Wed, Oct 17, 2018 at 09:30:50AM -0500, Daniel Taylor wrote:
> >
> > [cut]
> >
> > user-space (please remember that the only way to reliably know that a
> > process is dead under unix is to be the parent of that process....).
> You can be the parent process of a userspace without being PID 1.
> That's the beauty of it.
> > I know the issue looks easy and straightforward on the surface. But
> > when you start looking into it seriously, you quickly realise that
> > things are not as straightforward as you thought ;)
> >
> The problem description is very straightforward.
>
> I don't know how hard implementing it will be.
>
> I guess as the person who suggested it, it's my responsibility to at
> least scope it out properly.
>
> Me and my big mouth.
>
> --
> Daniel Taylor
>
>
I'm actually interested in tracking with this.
I've been working on a Tails remastering pipeline that strips systemd from the
image. Part of that pipeline was slated to include an automatic unit file
converter to sysvinit scripts.
One of the things that slowed me down is they have scripts that are highly
dependent on systemd NetworkManager calls. From memory I think they were sysctl
calls with extended parameters to invoke the network manager. I'd need to look
again but my idea here was to do a "drop in replacement" for the "ctl" commands.
I was able to strip their calls to localectl to make things go but in theory
simple versions of these could be reimplemented as bash scripts or something.
In my case the scope is fairly narrow (i.e. only what Tails needs is what I need
to reimplement) but with the sysvinit hit coming for Buster I've been
contemplating if the scope should be expanded to automatic package re-packaging
(where unit files are automatically converted to sysvinit scripts). At that
scale it would have to be really well implemented or at the very least provide a
list of things it can't convert automatically.
I have my hands on a python based converter from github dated 2012. I was
planning on starting there and seeing how bad it munged the unit files that
Tails depends on.
I've been wondering if a project would like this would gain traction from others
that are trying to "un-systemd" their favorite distros.