1 |
On Sat, Apr 09, 2016 at 12:06:47AM -0400, Anthony G. Basile wrote: |
2 |
> On 4/8/16 11:03 PM, Rich Freeman wrote: |
3 |
> > On Fri, Apr 8, 2016 at 9:51 PM, Anthony G. Basile <blueness@g.o> wrote: |
4 |
> >> |
5 |
> >> Alternatively, this may introduce problems. So it seems like we're |
6 |
> >> fixing something that isn't broken. |
7 |
> >> |
8 |
> > |
9 |
> > What problems are you anticipating, especially in light of the fact |
10 |
> > that many distros actually do it this way already? |
11 |
> |
12 |
> RBAC policy files for one. You'll probably break every single hardened |
13 |
> gentoo server out there. |
14 |
|
15 |
Tell me more about this; I don't know a lot about what would break. |
16 |
Also, are you sure it would break, or are you just thinking that it |
17 |
would? |
18 |
|
19 |
> scripts and programs that assume different executables with the same |
20 |
> name at different points along the path, eg I know a company where |
21 |
> they've set up an ssh wrapper at /usr/local/bin/ssh which wrap /usr/bin/ssh. |
22 |
|
23 |
This won't break, because /usr/bin/ssh would still exist as it does and |
24 |
/usr/local is not touched. |
25 |
|
26 |
File colissions between the directories that are being merged would |
27 |
definitely be an issue that needs to be worked out before this could |
28 |
happen, and I understand that. I know of at least one, and we would need |
29 |
to find out if there are others. |
30 |
|
31 |
Forgetting about /usr/local since we don't control that, this type of |
32 |
file name collision across the bin directories is not good whether or |
33 |
not you merge /usr. It would cause issues in path name resolution. |
34 |
|
35 |
> security measures where you don't dereference sym links along $PATH |
36 |
> because sym links can be used in various types of exploits. |
37 |
|
38 |
Every amd64 gentoo system already has one of these, the "lib" symlink, |
39 |
both in / and /usr, so if you aren't dereferencing symlinks, aren't you |
40 |
already broken? |
41 |
|
42 |
> > |
43 |
> > I don't really have a problem with making it optional or the default. |
44 |
> |
45 |
> if we don't make it optional we're going to cause some serious headaches |
46 |
> for people who are invested in the current status quo. |
47 |
> |
48 |
> > |
49 |
> > It can also be left up to the maintainers, and of course somebody |
50 |
> > could even fork baselayout/etc if they wish and virtualize it in |
51 |
> > @system. Most things in Gentoo don't actually require a consensus to |
52 |
> > move forward, especially if they aren't defaults. |
53 |
> |
54 |
> if we deprecate the linker scripts in /usr/lib by stubbing out |
55 |
>gen_usr_ldscript, then its not as simple as "maintainer's choice". |
56 |
|
57 |
gen_usr_ldscript is only needed if you are using separate /usr without |
58 |
an initramfs. This is unsupported and orthogonal to the usr merge. |
59 |
|
60 |
> > |
61 |
> > In any case, what is the point of this thread? If somebody wants to |
62 |
> > implement a merged /usr what exactly is stopping them from doing so? |
63 |
> |
64 |
> i'm against something that doesn't maintain backwards compat. |
65 |
|
66 |
If there are ways that merging / into /usr does not maintain backward |
67 |
compatibility, I want to know what they are. |
68 |
|
69 |
This is not directed at anyone specifically, it is just a general |
70 |
comment. |
71 |
|
72 |
I've seen a lot of speculation on this thread about what might break, |
73 |
and a lot of comments about a perceived removal of choice. |
74 |
|
75 |
Can someone help get a summary together? let's get a single message |
76 |
summarizing everything. |
77 |
|
78 |
Thanks, |
79 |
|
80 |
William |