1 |
-----BEGIN PGP SIGNED MESSAGE----- |
2 |
Hash: SHA256 |
3 |
|
4 |
On 04-08-2011 07:55, Samuli Suominen wrote: |
5 |
> On 08/04/2011 05:30 AM, Michał Górny wrote: |
6 |
>> On Sat, 30 Jul 2011 10:27:27 +0300 |
7 |
<snip> |
8 |
>> So, let's sum up a little. |
9 |
>> |
10 |
>> The most common argument against separate /usr requiring a proper |
11 |
>> initramfs is 'it works now, thus it's great'. That is practically |
12 |
>> understandable that people don't like to switch things upside down |
13 |
>> like that, especially when machines are not locally reachable. |
14 |
>> |
15 |
>> What's the exact differences between an initramfs and an early |
16 |
>> bootup setup in rootfs? As I see it: - initramfs is a small fs |
17 |
>> which is used for a short while on boot, to setup the system |
18 |
>> necessarily for the early bootup sequence, - while initial rootfs |
19 |
>> is a rather large piece of fs which is supposed to contain random |
20 |
>> stuff necessary for the early bootup to be able to proceed and |
21 |
>> mount the necessary remaining stuff before the actual bootup |
22 |
>> begins. And we're mostly stuck with it for the whole runtime. |
23 |
>> |
24 |
>> As I see it, I see no reason to keep forcing things like complete |
25 |
>> glibc, ncurses and the whole other lot of libraries for the early |
26 |
>> bootup if all needed is some kind of minimal 'mount' program (for |
27 |
>> instance). |
28 |
>> |
29 |
>> In the ol' days I tried building a NFS-shared system and the main |
30 |
>> problem was that some of early run tools relied heavily on the |
31 |
>> local system libs and files before they were replaced by NFS |
32 |
>> mounts. And I had to keep them in sync manually which is not the |
33 |
>> most comfortable thing. |
34 |
>> |
35 |
>> I don't see how trying to fit the best set of libs and files into |
36 |
>> rootfs can solve it. You either want for the system to be clean or |
37 |
>> weirdly split to support various possible configurations. And |
38 |
>> decide which are not 'weird enough' not to support. |
39 |
>> |
40 |
>> And really, most of the things about separate /usr are hacks which |
41 |
>> were introduced because the system was incapable of a proper |
42 |
>> rootfs. Read-only /usr should be read-only rootfs with writable |
43 |
>> mounts on top of it. NFS-mounted /usr should be the whole system |
44 |
>> part network-mounted (which would be easier if everything went into |
45 |
>> /usr rather than being split). |
46 |
|
47 |
> It seems what we need is an migration plan. Sending out a Portage |
48 |
> News item, and correcting documentation as first step. |
49 |
> |
50 |
> Then giving people enough time to migrate. This would give us plenty |
51 |
> of time to work on the details for moving the files over from / to |
52 |
> /usr. |
53 |
|
54 |
Again, not all of us are willing to migrate away from a separate /usr |
55 |
partition, least of all when that is being "imposed" by some people |
56 |
trying to "shove" their pet projects to others and when we don't agree |
57 |
with or acknowledge the arguments. |
58 |
|
59 |
> It seems non-problematic for new installs, as stages could ship the |
60 |
> symlinks and files get installed to /usr through them, even before |
61 |
> the packages are changed. |
62 |
|
63 |
The symlinks will have to be part of baselayout as files get into stages |
64 |
through packages and not through catalyst. |
65 |
|
66 |
- -- |
67 |
Regards, |
68 |
|
69 |
Jorge Vicetto (jmbsvicetto) - jmbsvicetto at gentoo dot org |
70 |
Gentoo- forums / Userrel / Devrel / KDE / Elections / RelEng |
71 |
-----BEGIN PGP SIGNATURE----- |
72 |
Version: GnuPG v2.0.17 (GNU/Linux) |
73 |
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ |
74 |
|
75 |
iQIcBAEBCAAGBQJOOn/LAAoJEC8ZTXQF1qEPOQsP+we6tifTVnCXqr46ajXa2Xft |
76 |
NqXhJfxmONGbbhfDYPhoNiGK5ovojpoDncKEE0t158X35QfLRjqFqrudbPDUNzrh |
77 |
/zEJmQYacZckyMT866PE2iJBovEA5ZBnXB8y6RBHJLH3ky5/dO8R92jHSnNihi1y |
78 |
u639+dpRHP6cRQIk9i2sEHaph+bZo6e3X+GCT6FL63m4sNDSBfJGo4wtMewp/aDD |
79 |
HS2Ya41WAt+SYA131QLcVwLhyDz7sRdQm1iR7W06iScMxgE/mKHF9S25NKMYf+H+ |
80 |
Qtd+PF1SLcxC1lKztPsmNTr1lpDLlAoO5OQzpOnXoPmCWvuzBVyrHfSPo+cxQOFM |
81 |
6VA0mjdNODS4gbEL5Fu8Q/Asf3/byJ7gBOfLNuHkMksMfLSy/O0KXjx3fnmpj1a0 |
82 |
yXlt+iuer7z5rwuz7ZfXNCmw0DWzuMOUimz1jz0pUwTzXDD9zZJXKHOt/RR4oQb8 |
83 |
NLldmh8YBcl17r6l60H49GWyL8YiIhQetBZuNi9+Pm72o3vVsKmCnyXHP1Cf0CsQ |
84 |
ziVy4+Lub2qSSQfndrTHnJ6rDIDFSLT4iZYRDJmlf6Mhrk7abogze/s0Vgfkfrfl |
85 |
yJVNVPG3Evk4d1qIFROSmQhhu44EOkufhijYvytpCHeNLvWUupeaMZOchX6QUXp4 |
86 |
4FhE/udxLI1zpQtTHLbJ |
87 |
=DhRY |
88 |
-----END PGP SIGNATURE----- |