| 1 |
Rich Freeman posted on Sat, 04 May 2013 08:54:16 -0400 as excerpted: |
| 2 |
|
| 3 |
> On Sat, May 4, 2013 at 6:42 AM, Luca Barbato <lu_zero@g.o> wrote: |
| 4 |
>> Hopefully we might have a gsoc student volunteering to make a |
| 5 |
>> runscript/lsb-script/systemd-unit compiler and a small abstraction so |
| 6 |
>> we write a single init.d script and generate what's needed. |
| 7 |
>> Probably we might even support pure-runit that way with minimal effort. |
| 8 |
>> |
| 9 |
> I'm skeptical that this will ever make sense - both init systems have |
| 10 |
> features that it would make sense for units/scripts to make use of in a |
| 11 |
> more tailored fashion. |
| 12 |
|
| 13 |
Same here. Sure, an automated conversion is possible and arguably might |
| 14 |
serve as a starting point, but you're losing the best of both initsystems |
| 15 |
in the process. |
| 16 |
|
| 17 |
> That said, if you really wanted to inter-convert, my gut feeling is that |
| 18 |
> it would be easier to go from a systemd unit to an init.d script, and |
| 19 |
> not the other way around. A systemd unit is more like a specification - |
| 20 |
> it describes the end result of what systemd should do. |
| 21 |
> An init.d script is an executable program |
| 22 |
|
| 23 |
You're a bit behind on openrc features, I think. =:^) It's actually |
| 24 |
quite possible for openrc/runscript "scripts" to be written in a "spec- |
| 25 |
style" format similar to systemd's unit files, just as it's possible for |
| 26 |
systemd to run "legacy" shell-style scripts with little or no |
| 27 |
modification, as some distros did with their initial conversion, |
| 28 |
according to what I've read. |
| 29 |
|
| 30 |
I think there's even some in-tree examples, tho I'm too lazy to go |
| 31 |
looking ATM and their package maintainers and/or williamh would be more |
| 32 |
familiar with them and could probably point them out off the top of their |
| 33 |
head without looking. |
| 34 |
|
| 35 |
> The reality is that systemd units are floating around all over the place |
| 36 |
> - when I installed it on a Gentoo box I ended up just Googling for |
| 37 |
> already-written units for daemons that lacked them in Gentoo and tweaked |
| 38 |
> them. |
| 39 |
|
| 40 |
That's what I have always figured I'd do, if I were to decide to convert |
| 41 |
here before all the packages I init here had in-tree unit-files. |
| 42 |
|
| 43 |
> Systemd units are much easier to write (typically) than init.d scripts |
| 44 |
> so this could be an area where end-users could contribute. |
| 45 |
|
| 46 |
See above. In theory it should be about even either way, since both |
| 47 |
systemd and openrc can do either scripted or spec-style "units". |
| 48 |
However, I expect systemd's "google resource" to be deeper in this |
| 49 |
regard, both with regard to the units themselves and to documentation |
| 50 |
about them, and the experience quotient probably favors systemd as well, |
| 51 |
so in practice you're almost certainly right, if only from the previous |
| 52 |
experience and googlable documentation and samples perspective. |
| 53 |
|
| 54 |
-- |
| 55 |
Duncan - List replies preferred. No HTML msgs. |
| 56 |
"Every nonfree program has a lord, a master -- |
| 57 |
and if you use the program, he is your master." Richard Stallman |
Archives