1 |
William Hubbs schrieb: |
2 |
> On Tue, Jan 03, 2012 at 10:22:15PM +0100, Fabian Groffen wrote: |
3 |
>>> I'll have to go through on my system at |
4 |
>>> least and find all of the ebuilds that install things in |
5 |
>>> /{bin,sbin,lib}. I'll open a tracker bug as soon as udev-176 is |
6 |
>>> released; this will list all of the things we need to do to complete the |
7 |
>>> migration. |
8 |
>> |
9 |
>> I would suggest not to do this. It's more interesting to know what udev |
10 |
>> really requires to be in /usr/bin. |
11 |
> |
12 |
> The issues involve binaries in /{bin,sbin} that link to libraries in |
13 |
> /usr/lib as well as packages that install udev rules that run binaries. |
14 |
> |
15 |
>> |
16 |
>>> Basically I have these in my head: |
17 |
>>> |
18 |
>>> * mask udev-176 in the tree. |
19 |
>>> * figure out and document how to make a simple initramfs with dracut. |
20 |
>>> * unmask udev 176 making sure to point users with a separate /usr |
21 |
>>> partition to how to make an initramfs (I could probably do this with |
22 |
>>> ewarns in the ebuild and maybe a news item before we go stable). |
23 |
>>> * stabilize a version of dracut. |
24 |
>>> * stabilize >=udev-176 and kmod. |
25 |
> |
26 |
> The part of the process above is the part I am the most concerned about. |
27 |
> I think we need to get everyone who is using separate /usr switched over |
28 |
> to an initramfs with udev 176, and this needs to happen sooner than |
29 |
> later, without using things like wrapper scripts or ways to avoid the |
30 |
> initramfs. Those are just stop-gap options that will only work until |
31 |
> some package they are depending on migrates to /usr. |
32 |
> |
33 |
> Once we get to this point in the process, I think we could take each |
34 |
> package that installs things in / individually and migrate it. But, I |
35 |
> think the part of the process listed above needs to happen sooner than |
36 |
> later. |
37 |
> |
38 |
> What are your thoughts? |
39 |
> |
40 |
> William |
41 |
> |
42 |
|
43 |
If i did not miss something in this long thread, we are currently mostly |
44 |
talking about udev needing /usr to be mounted when it starts. Systemd is |
45 |
not our default init system and other packages are currently not |
46 |
changing their default. Since udev is our default and that requires a |
47 |
mounted /usr, we should show our users their options with a suggested |
48 |
default: |
49 |
|
50 |
1. minimal initramfs (default suggestion) |
51 |
2. switching from udev to mdev (avoids required /usr of udev) |
52 |
3. some wrapper script to mount /usr before udev starts |
53 |
|
54 |
for 1: I use myself a self-created initramfs, which has worked fine for |
55 |
many kernel releases, the only adjustment needed was back with kernel |
56 |
2.6.27. With this in mind, some default minimal initramfs should be |
57 |
doable without much maintaincence work. |
58 |
|
59 |
for 2: from what i did read up to now, this option looks interesting for |
60 |
those people, who dont want to use an initramfs or other mounting script |
61 |
and dont use advanced features of udev, so especially server setups or |
62 |
minimal desktop systems |
63 |
|
64 |
|
65 |
|
66 |
For the idea of complete migration to /usr, i see no reason to go this |
67 |
route in advance. Just keep with our default install locations and |
68 |
follow upstream, if and where needed. |
69 |
If a package switches to /usr and depending packages follow, fine, let |
70 |
them do that and we can follow without much additional work. |
71 |
There is no need to do such migration beforehand or changing our install |
72 |
location in advance. |
73 |
|
74 |
|
75 |
So in short: |
76 |
|
77 |
1. Tell our users about the change in udev and show them their options |
78 |
with a suggested default one. |
79 |
2. for the rest, just calm down, follow upstreams and only change the |
80 |
install location where needed. No need for mental pressure without an |
81 |
actual pressure. ;-) |
82 |
|
83 |
-- |
84 |
|
85 |
Thomas Sachau |
86 |
Gentoo Linux Developer |