1 |
On 09/29/2017 04:53 AM, Rich Freeman wrote: |
2 |
> On Thu, Sep 28, 2017 at 11:32 PM, Ulrich Mueller <ulm@g.o> wrote: |
3 |
>>>>>>> On Thu, 28 Sep 2017, Austin English wrote: |
4 |
>> |
5 |
>>> Talking with Whubbs about it, I found that our service script only |
6 |
>>> supports OpenRC, via rc-service. I looked around, and from what I |
7 |
>>> can tell, most distros ship a service tool for all supported init |
8 |
>>> systems. I.e., Debian/Ubuntu: supports sysvinit and systemd via |
9 |
>>> init-system-helpers CentOS/Fedora: provides support for systemd via |
10 |
>>> initscripts OpenSUSE: has a working service binary for systemd |
11 |
>>> (according to #suse) |
12 |
>> |
13 |
>> There is "eselect rc" which could be easily extended to support |
14 |
>> systemd. Patches are welcome. :) |
15 |
>> |
16 |
> |
17 |
> ++ |
18 |
> |
19 |
> Honestly, I could see the argument for having a generic "service" |
20 |
> command because that is what everybody else does, but there is little |
21 |
> point in arguing about the name of the file when nobody has bothered |
22 |
> to write it yet. |
23 |
> |
24 |
> If somebody writes such a tool and it proves useful, we can always |
25 |
> have the discussion about refactoring. |
26 |
> |
27 |
> To minimize list replies I'll tackle one of Duncan's points - he was |
28 |
> debating whether you really need this vs just using systemctl. The |
29 |
> obvious use case is scripts that are intended to support multiple init |
30 |
> systems - it makes far more sense to put the logic to figure out which |
31 |
> one to run in one place than many. If the runit users want to add |
32 |
> their own logic they could. IMO it would be potentially useful, even |
33 |
> if you and I don't personally have much use for it. |
34 |
> |
35 |
> That said, the sorts of people most likely to benefit probably don't |
36 |
> use Gentoo in the first place. |
37 |
> |
38 |
> In any case, arguing over whether it is useful is putting the cart |
39 |
> before the horse. It doesn't matter if it is useful if nobody bothers |
40 |
> to build it. If nobody has that much of an itch to scratch then how |
41 |
> useful could it be? |
42 |
> |
43 |
|
44 |
Great points. It'll be much easier to decide on something when/if there |
45 |
is something concrete to work with. There isn't much stopping a package |
46 |
from making it into Gentoo. If there is demand, it'll be written. |
47 |
|
48 |
-- |
49 |
Daniel Campbell - Gentoo Developer, Trustee, Treasurer |
50 |
OpenPGP Key: 0x1EA055D6 @ hkp://keys.gnupg.net |
51 |
fpr: AE03 9064 AE00 053C 270C 1DE4 6F7A 9091 1EA0 55D6 |