:: Re: [DNG] powerdns upstream has dro…
Góra strony
Delete this message
Reply to this message
Autor: wirelessduck
Data:  
Dla: dng
Temat: Re: [DNG] powerdns upstream has dropped sysvinit support


> On 13 Oct 2023, at 07:54, tito via Dng <dng@???> wrote:
>
> On Thu, 12 Oct 2023 17:16:38 -0300
> altoid via Dng <dng@???> wrote:
>
>>> On 12 Oct 2023 at 16:00, tito via Dng wrote:
>>>
>>> it just needs the init system that pleases you?
>> I don't care which, I really don't have a preference.
>>
>> Just as long as it is *not* systemd or a *systemd_like* init and
>> adheres to the basic Linux principles/philosophy.
>>
>>> We are already there it is very imperfect and it is everywhere.
>> Not everywhere, at least not yet.
>>
>> My Devuan box uses sysvinit, the default in a most (?) non-systemd
>> distributions.
>>
>>> ... you are right we do not need to discuss it is already all decided ...
>>> ... should still be some minimal details to decide you are here.
>> You have obviously missed the point I am trying to make.
>
> No, I fully understand the point you try to make, but there are
> other points here that are quite interesting:
> a) if we will not be able to cut  the umbilical cord  that ties us to systemd
>     they will be able to grind us down by attrition in medium term
>     by changing every release that little bit that put exponential extra work
>     on the shoulders of init-freedom developers or "give me  just another init"
>     developers.
> b) how can the cut be achieved:
>      1) using systemd service files: no, we are in the same trap as before.
>      2) collecting init scripts: maybe, but it restricts the range of developers
>           because some distros don't use sysvinit but a different init.
>      3) creating a collection of service files that could be sourced by every
>            init system out there (even systemd)  can give you the critical mass
>            of developers you would need  to port the over 1000 (if I recall correctly)
>            init scripts to orphan-init-scripts and init freedom choice would be
>            broadened  to other init systems.

>
> Is this an optimal solution? No it is not, because in this struggle there is none.
>
> Ciao,
> Tito


Have you read the s6-frontend website pages? It says that s6-frontend will get its own declarative service file format.

https://skarnet.com/projects/s6/frontend/index.html

https://skarnet.com/projects/s6/frontend/servicefiles.html

All I can do here is wait and hope it gets finished sooner rather than later.

Tom