Gentoo Archives: gentoo-dev

From: Rich Freeman <rich0@g.o>
To: gentoo-dev <gentoo-dev@l.g.o>
Subject: Re: [gentoo-dev] Possibility of overriding user defined INSTALL_MASK from an ebuild?
Date: Sat, 01 Mar 2014 01:29:31
Message-Id: CAGfcS_kCA35e62etBOyCz0khLre9xraz+1fdSMZfc6KJRNC9fQ@mail.gmail.com
In Reply to: Re: [gentoo-dev] Possibility of overriding user defined INSTALL_MASK from an ebuild? by Steev Klimaszewski
1 On Fri, Feb 28, 2014 at 8:09 PM, Steev Klimaszewski <steev@g.o> wrote:
2 > Please keep in mind that not every device that runs Gentoo has the
3 > ability to just pop new storage in with more space. The Beaglebone
4 > Black has 2GB eMMC.
5
6 Hence the reason I suggested that embedded systems are a perfect case
7 where INSTALL_MASK makes sense. I'm all for having the feature there
8 for users to use.
9
10 > I'm not exactly a fan of systemd, though I know it has some uses, and
11 > I'm still curious as to why it installs/stores *configuration* data
12 > in /lib - if only from an upgrade point of view, we back up /etc, we
13 > back up /home - now we need to back up /lib, /usr/lib, /var, or whatever
14 > some random upstream decides is a good place to store configuration
15 > information!?
16
17 Why would you back up default config settings, except as part of a
18 full system backup including all the binaries/libs/etc? If you lose a
19 file in /usr you just re-install the package and it will be there.
20
21 Users aren't supposed to edit any of the config files in /usr, just as
22 they're not supposed to edit make.defaults in /usr. If they don't
23 like what a unit file or udev rule in /usr does, they just have to
24 stick a corresponding file in /etc to override it, and that of course
25 should be backed up.
26
27 If they just hard-coded the defaults in the source code, would that
28 make it better (in which case it would be in /usr, and far more opaque
29 - and this is traditionally how these things were done)?
30
31 >
32 > I get it, we all are busy, we have better things to do than patch things
33 > from upstream, but sometimes, it's a requirement. We wouldn't install a
34 > script that does rm -rf / with root permissions (assuming a non-hardened
35 > box where it would cause quite a bit of damage) - but we're okay with
36 > just randomly storing configuration data all over the system? Please
37 > consider these things before saying we're going to blindly follow
38 > upstream - if something is too hard for you to maintain, ask for some
39 > assistance in doing so.
40
41 Nobody is saying that we should follow upstream simply for its own
42 sake. However, we shouldn't change things when there is no real
43 benefit to doing so. How is sticking a bunch of template config files
44 that users aren't supposed to touch in /etc helpful? How is removing
45 "systemd" from filenames helpful? How is sticking a few files in /lib
46 instead of /usr/lib more helpful than sticking copies of them in an
47 initramfs?
48
49 And I don't see the value in the more traditional config file style
50 where you just give the user one huge file and allow them to use
51 etc-update/etc to constantly re-merge changes into it. I greatly
52 appreciate having a 8 line file in /etc/X11/xorg.conf.d vs what my
53 config files used to look like. I still have a backup of XF86Config
54 in /etc from 2004 and it is 464 lines long and was no more functional,
55 beyond being more likely to break during upgrades.
56
57 I really don't see anything that can't be done with Gentoo today that
58 could be done if we just moved a bunch of files around.
59
60 However, if there really are things that are causing genuine problems
61 I'm all for making little changes that fix them when they're
62 practical. My main concern with this whole area is that it really
63 strikes me as fighting over "the one true way." Is there a certain
64 elegance to separating partition management, raid, and file storage?
65 Absolutely, but that doesn't mean that there aren't real benefits to
66 something like btrfs.
67
68 Rich