1 |
Luca Barbato posted on Sat, 09 Apr 2016 15:03:15 +0200 as excerpted: |
2 |
|
3 |
> On 09/04/16 14:37, Rich Freeman wrote: |
4 |
>> I've certainly haven't had many problems with dracut. When it fails it |
5 |
>> is usually because I'm doing something ELSE that is off-the-wall and it |
6 |
>> just doesn't have a plugin for it yet. (And in those cases it isn't |
7 |
>> like the kernel tends to get it right without an initramfs.) |
8 |
>> |
9 |
>> I'd certainly want to test it on a merged /usr, but I'd be surprised if |
10 |
>> it doesn't work, since it was designed to run on distros that are using |
11 |
>> a merged /usr. |
12 |
> |
13 |
> I think that should be the first thing to do not the last one =) |
14 |
|
15 |
FWIW, dracut works just fine with a "reverse-merged" /usr (usr -> .), as |
16 |
well as the bin/sbin merge. And if it works with that, it'll certainly |
17 |
work with (normal) merged usr, as AFAIK upstream's fedora/rh sponsored. |
18 |
|
19 |
>> In an ideal world, you might argue that / should just be a tmpfs or |
20 |
>> something almost as ephemeral. It is just a place you hang everything |
21 |
>> else off of. |
22 |
|
23 |
> That would be the core concept, but then you can just not have /bin |
24 |
> /sbin /lib |
25 |
|
26 |
That's in the context of (forward) /usr merge, which would make all those |
27 |
symlinks to the appropriate /usr location anyway. Those symlinks could |
28 |
of course be created dynamically, as could the various mountpoint |
29 |
directories. |
30 |
|
31 |
Of course /etc would have to be dynamically mounted in that scenario as |
32 |
well, but that's not a big issue as long as there's an initr* |
33 |
|
34 |
I actually thought about doing a tmpfs-based / here, or effectively just |
35 |
never doing the pivotroot off the initramfs, with everything else |
36 |
dynamically mounted over top the initramfs dirs, but decided to go a |
37 |
different way instead, putting (almost) everything installed by the PM |
38 |
on /, and doing a reverse-/usr-merge, with /usr -> . , with / then ro- |
39 |
mounted by default. Seemed simpler for what I wanted to do than the |
40 |
tmpfs or stay-on-initramfs / route. |
41 |
|
42 |
>> The thing I like about the merge is that it basically puts all your |
43 |
>> distro-supplied stuff in one place. /usr basically becomes the OS |
44 |
>> minus state. If things started out that way and you just had a short |
45 |
>> stub loader that gets things initialized, and I were arguing that |
46 |
>> instead of that little initialization stub you should break up /usr so |
47 |
>> that the root count mount /usr, would that sound all that compelling? I |
48 |
>> think having it all in one mountpoint seems a lot more compelling. |
49 |
> |
50 |
> you cannot ever have everything in 1 mount point, you just move the |
51 |
> problem somewhere else you notice less (initramfs), but the problem |
52 |
> remains and either is solved or not. |
53 |
> |
54 |
> having everything in /usr and then copy it over ${somewhere} is there, |
55 |
> it can be debated if /bin or initramfs is the best place to put it. |
56 |
|
57 |
I suppose many of us have made that point at least to ourselves, at some |
58 |
point. =:^) |
59 |
|
60 |
-- |
61 |
Duncan - List replies preferred. No HTML msgs. |
62 |
"Every nonfree program has a lord, a master -- |
63 |
and if you use the program, he is your master." Richard Stallman |