1 |
On Thu, Aug 4, 2011 at 7:49 AM, Dale <rdalek1967@×××××.com> wrote: |
2 |
> From a users perspective. Could it not be possible to have some USE flag, |
3 |
> or other setting, that would tell portage that a separate /usr partition is |
4 |
> being used then have the needed files placed elsewhere on / ? I'm not a dev |
5 |
> and I don't play one on TV but I do like options and being able to customize |
6 |
> some things. It is one of the things Gentoo is about. |
7 |
> |
8 |
|
9 |
I don't see what a USE flag gets us: |
10 |
|
11 |
1. If you have a separate /usr then either booting without an |
12 |
initramfs will work or it won't work - largely depending on how |
13 |
complex your environment is. Booting with an initramfs will work |
14 |
reliably (well, if we sort out the initramfs situation - having done |
15 |
some more tests I have one virtual machine which was pretty easy to |
16 |
get running, and one physical box that for whatever reason wouldn't |
17 |
detect/start the RAID). |
18 |
|
19 |
2. If you don't have a separate /usr than booting will always work |
20 |
regardless of where the files are, since the system will always find |
21 |
them. |
22 |
|
23 |
Unless what is being proposed is to actually do the Fedora thing and |
24 |
make /bin, /lib, etc a symlink into /usr/bin, /usr/lib, etc than there |
25 |
isn't anything at package-install time for the flag to affect. If we |
26 |
do want to do the Fedora thing would a flag even work, since those |
27 |
directories get created from the stage3? It seems to me that if you |
28 |
want the symlinks you just need to set them up when doing the install |
29 |
(or from a rescue disk), and then the package manager should follow |
30 |
the links when doing subsequent installs. Oh, and not all package |
31 |
managers like the top-level directories to be symlinks. |
32 |
|
33 |
I think that as was the case with the use of bash vs sh we may need to |
34 |
have a policy decision made here. Right now the general policy has |
35 |
been to conform to FHS, and the Fedora/etc proposal does not do this |
36 |
(and apparently we are already a bit out of compliance). I think that |
37 |
moving in a different direction is a big decision. |
38 |
|
39 |
And, if we do decide to move in that direction, I agree with Samuli |
40 |
that we need a transition plan. Packages can't just start breaking |
41 |
initrd-less setups left and right overnight. To start, we need to get |
42 |
dracut/etc configurable to mount any necessary directories (I checked |
43 |
- it is fairly smart (though not 100% effective) at finding root, but |
44 |
does not try to mount anything else). Then we need to update our |
45 |
documentation. Then we need to communicate the change to users, and |
46 |
give them time to migrate. Only then can packages have the freedom to |
47 |
require usr to be available at boot. |
48 |
|
49 |
I don't propose that if we move in this direction that we "fix" |
50 |
anything that isn't currently FHS-compliant - the damage is already |
51 |
done. We just should avoid propagating the situation until users are |
52 |
ready. |
53 |
|
54 |
Rich |