:: [DNG] Debian support for starting s…
Top Pagina
Delete this message
Reply to this message
Auteur: Olaf Meeuwissen
Datum:  
Aan: sawbona
CC: dng
Oude Onderwerpen: Re: [DNG] bookworm without systemd
Onderwerp: [DNG] Debian support for starting system services (was Re: bookworm without systemd)
Hi,

altoid via Dng <dng@???> writes:

> [...]
> Beside the fact that (as far as I know) Debian is a) supposed to
> support sysvinit on equal terms to systemd and that b) this being
> possible *today* does not in any way mean that it *will* be possible
> tomorrow.


Per Section 9.3.1. Introduction[1] of the Debian Policy's 9.3. Starting
system services (emphasis added)

Packages that include system services should include systemd service
units to start or stop those services. [...]

If the package does not include a service unit [...] including an init
script [...] to start the service is *encouraged*.

Packages including a service unit may *optionally* include an init
script to support other init systems.

[1]: https://www.debian.org/doc/debian-policy/ch-opersys.html#s-services-intro

So, no, Debian does not support sysvinit on terms equal to those for
systemd. And, yes, that sucks ... unless there is support for other
init systems and you happen to use one that's supported.
Think runit-init or openrc.

> ie: when (not if) Debian devs decide that instead of half-supporting
> sysvinit as they do now, they simply stop supporting it.
>
> Even *half-upporting* is being loosely used by me here, the fact is
> that Debian devs do not consider Debian+sysvinit to be supported.


Some devs are "hell-bent" on dropping everything except systemd support,
others are willing to keep init scripts in /usr/share/doc/$pkg/examples/
and yet others welcome init scripts and/or fixes to them. Your distance
may vary between devs ... and how you phrase your bug report :-)

> This is a snip from a reply to a bug report filed to Debian this
> year:
>
> ---
> "Since sysvinit is not enabled by default in Debian, I do not
> consider this bug as release-critical."
> ---


If the package this bug was reported against has a service unit, then
including an init script is, like it or not, optional. And even if it
doesn't have a service unit, providing an init script is still only
*encouraged*, not required, so, yes, it's not release-critical :-/

Hope this helps,
--
Olaf Meeuwissen