1 |
On Friday 28 February 2014 17:56:06 Samuli Suominen wrote: |
2 |
> On 28/02/14 17:34, Anthony G. Basile wrote: |
3 |
> > On 02/28/2014 06:15 AM, Patrick Lauer wrote: |
4 |
> >> On 02/27/2014 09:08 PM, Anthony G. Basile wrote: |
5 |
> >>> Hi everyone, |
6 |
> >>> |
7 |
> >>> I'm putting the call out there for any agenda items for the next |
8 |
> >>> Council |
9 |
> >>> meeting, which will be held on March 11, 2014 at 1900 UTC. This is |
10 |
> >>> short notice but we got off track because of FOSDEM and we're going to |
11 |
> >>> try to get back on track. |
12 |
> >>> |
13 |
> >>> So far, the only item is final ratification of glep 63 [1]. |
14 |
> >> |
15 |
> >> Since it's still a bit cold I'd like to start a nice fire to warm us up: |
16 |
> >> |
17 |
> >> I'd like QA and Council to figure out how much we care about FHS. |
18 |
> >> |
19 |
> >> My main complaint is some projects (including e.g. systemd and |
20 |
> >> apparently now also udev) storing config files in /lib and/or /usr/lib. |
21 |
> >> |
22 |
> >> From FHS' point of view this is totally wrong, config files go to /etc |
23 |
> >> |
24 |
> >> Only libraries should be in /lib. |
25 |
> >> Moving things to /usr/lib adds the extra fun that /usr needs to be |
26 |
> >> mounted to acces *config files*. This is bad for our collective blood |
27 |
> >> pressure. |
28 |
> >> |
29 |
> >> So I'd like to see config files stored in /etc again. Where they can be |
30 |
> >> properly tracked and versioned ... |
31 |
> >> |
32 |
> >> (iow, storing config files in any other location than /etc is wrong; |
33 |
> >> storing example configs in e.g. /usr/share is fine too; storing config |
34 |
> >> in any other place is a valid bug that needs to be fixed) |
35 |
> >> |
36 |
> >> For upstreams that insist on splitting configs in "system default" and |
37 |
> >> "local override" (which is rather nonsensical, but let them have some |
38 |
> >> fun) I would suggest a subfolder of /etc, maybe /etc/defaults or |
39 |
> >> /etc/systemdefaults or maybe /etc/lib/etc/usr/static if that's what |
40 |
> >> makes people happy |
41 |
> >> |
42 |
> >> |
43 |
> >> Enjoy the exothermic oxidation, |
44 |
> >> |
45 |
> >> Patrick |
46 |
> > |
47 |
> > Speaking as a council member and the next chair: Patrick, how would |
48 |
> > you pose this as a motion? As stated, the council should "discuss |
49 |
> > FHS" but how would you word this as a policy that we can rule on? I |
50 |
> > have an idea but would like to hear what you want. |
51 |
> > |
52 |
> > Speaking as a gentoo dev: This is one of my objections with systemd |
53 |
> > and the whole / + /usr merge. It violates a standard which is assumed |
54 |
> > in many setups, namely FHS. Another is that systemd violates the "one |
55 |
> > thing well" principle. |
56 |
> |
57 |
> That isn't true; systemd has dozens of sep. executables and each of them |
58 |
> do one'ish task, and do it well. The one running in PID 1 has very |
59 |
> little to do with them. |
60 |
|
61 |
Bad troll is bad - half the bits rely on PID1, e.g. logind is basically a |
62 |
convoluted dbus proxy into PID1. |
63 |
|
64 |
If you can show logind or journald running with sysvinit as PID1 your argument |
65 |
has a chance ... |