1 |
On Sun, Jul 25, 2021 at 2:05 PM Alan Mackenzie <acm@×××.de> wrote: |
2 |
> |
3 |
> OK, so you're clever and you know this. You know to do the |
4 |
> couter-intuitive thing of putting @system packages into @world. Less |
5 |
> clever people like me follow the handbook, and assume that packages in |
6 |
> @system are protected. Putting init-systems into @world is an unnatural |
7 |
> thing to do, and must be construed as a workaround for deficiencies in |
8 |
> portage. |
9 |
|
10 |
I think you're really conflating the package manager with how some |
11 |
virtuals/ebuilds are configured. Portage shouldn't have |
12 |
service-manager-specific functionality. Nor should it do things like |
13 |
never uninstall things that packages say aren't needed just in case |
14 |
you might still be using them, when you run it with --depclean which |
15 |
basically is supposed to have it remove the stuff that isn't needed. |
16 |
|
17 |
All protage does is follow the dependencies. |
18 |
|
19 |
I think there is room for discussing whether daemontools should be |
20 |
treated as a service manager by default. I've never used it and can't |
21 |
speak to how it is typically used. You might want to talk to the |
22 |
maintainers of it to get their thoughts. |
23 |
|
24 |
> No, I did not make that mistake. I simply assumed, not entirely |
25 |
> consciously, that Gentoo would not destroy my system without me |
26 |
> specifically asking it to. |
27 |
|
28 |
It isn't like uninstalling openrc is going to "destroy your system". |
29 |
Worst case you could always just pass init=/bin/bash or whatever to |
30 |
the kernel and just reinstall it. Granted, it wouldn't really be |
31 |
welcome behavior. |
32 |
|
33 |
> It would be saner still to be kept in the system file (whatever that |
34 |
> might be). |
35 |
|
36 |
I suppose you might not care to hear that I've advocated a few times |
37 |
that the "system file" should disappear entirely and no packages |
38 |
should get special treatment. :) Granted, that has more to do with |
39 |
how assumed dependencies work in the repository. I don't really have |
40 |
a problem with confirmation before removing certain packages because |
41 |
reinstalling them can be quite painful. The service manager actually |
42 |
is one of the easier ones to fix. If you break the ability to |
43 |
bootstrap gcc/etc that can be a real bugger to fix. |
44 |
|
45 |
Really though posting on this list and successfully winning every |
46 |
argument with everybody who replies is going to do zero to change this |
47 |
behavior, because it is unlikely that anybody involved in this |
48 |
particular issue reads this list. It would make more sense to chat |
49 |
with the daemontools maintainer and get their thoughts on how the |
50 |
virtual ought to be configured as a starting point. You could always |
51 |
try for a second opinion/etc if that doesn't work. Plus the |
52 |
conversation would probably help you understand what their thinking |
53 |
was. |
54 |
|
55 |
-- |
56 |
Rich |