:: Re: [DNG] There seems to be some st…
トップ ページ
このメッセージを削除
このメッセージに返信
著者: Joel Roth
日付:  
To: dng
題目: Re: [DNG] There seems to be some strong disagreement in Debian regarding usrmerge
On Sun, Dec 31, 2023 at 02:00:42PM -0500, Steve Litt wrote:
> altoid via Dng said on Sun, 31 Dec 2023 05:59:08 -0300
>
> >Hello:
> >
> >On 12/30/23 12:25, altoid via Dng wrote:
> >
> >>> Linux has tens of thousands of very capable users ...
> >
> >Lars Noodén via Dng wrote:
> >
> >> ... work flow be modified to let them in?
> >> ... the sea change needed in how work is split up ...
> >
> >I don't know exactly how the maintainer work flow is designed, I
> >expect that each maintainer or group of maintainers has their own way
> >to get their work done, but I.may be missing something.
> >
> >I think that the best way (?) to get this done would be as was said
> >earlier by tito, to leverage/use what we *already* have at hand.
> >
> >ie: find a way to leverage systemd .service files to convert them to
> >initscripts.
> >
> >Both .service files and initscripts are, by design, standardised and
> >by comparison to human language, very simple so it should be (?)
> >straightforward to translate one to the other.
> >
> >I may be mistaken, but I really don't think Devuan will be able to
> >count on package maintainers to keep initscripts up to date for *us*
> >to be able to use their packages in systemd-less distributions.
> >
> >That task cannot be left to package maintainers basically because it
> >will *not* get done, it has to be done at distribution maintainer
> >level.
> >
> >On one hand because, maintenance wise, it is more work for them.
> >At the most, some may leave them around for historical purposes but
> >the scripts will not be maintained.
> >
> >And on the other, because systemd is *the* tool with which MS/IBM/RH
> >has been infiltrating the Linux ecosystem.
> >
> >So we can only expect for things to get even murkier.
> >eg: unreadable .service blobs anyone?
> >
> >Best,
> >
> >A.
>
> So what is your solution?


A bunch of folks on this list earlier posted
the results of ls /etc/init.d.

A next step would be to track the appropriate debian
git respositories for those packages. At some time
in future, if the debian packager removes the init
script, you revert that commit in your repo,
and using the debian toolchain, generate your
version of the package *with* the sysvinit script,
and continue to track the package.

Debian publishes package popularity information, which
could also be used to help select packages to follow.

Even if you just copied the sysvinit scripts out of the
latest debian packages, you would always have the most
recently available version. That could be a quick hack.

I think the majority of important, mature packages won't
often have changes in their init system.

Wish everyone on this list--and the whole world--good health
and happiness in the New Year.


--
Joel Roth