1 |
Steven J. Long posted on Fri, 02 Aug 2013 12:31:08 +0100 as excerpted: |
2 |
|
3 |
> As Rich said, lvm doesn't link outside rootfs so it's not an issue: you |
4 |
> only really need an initramfs if rootfs is on lvm/encrypted/raid, or you |
5 |
> need udev to get through localmount. |
6 |
|
7 |
Or, unfortunately, for root on mult-device btrfs[1], since the usual |
8 |
kernel commandline rootflags=device=/dev/whatever doesn't seem to work |
9 |
for some reason.[2] |
10 |
|
11 |
Tho hopefully that bug will be fixed before the "experimental" label is |
12 |
stripped from btrfs... |
13 |
|
14 |
--- |
15 |
[1] Yes, as appropriate for running on an experimental fs, I have |
16 |
backups, tho the dual-device raid1 mode I'm using is /reasonably/ stable, |
17 |
now. I switched to btrfs when I upgraded to ssd, both for the ssd |
18 |
support and for checksummed data/metadata with a second copy to retrieve |
19 |
from if the first fails the checksum. |
20 |
|
21 |
[2] I tried with a btrfs raid1 and could mount either device with root= |
22 |
and rootflags=degraded, so it was definitely parsing rootflags, but no |
23 |
way was it taking rootflags=device= to mount them undegraded without a |
24 |
userspace btrfs device scan first, and that requires an initr*. Maybe it |
25 |
was a problem with the kernel splitting at the second = instead of the |
26 |
first, I don't know, but it definitely wasn't working. So now I'm |
27 |
running dracut with several of its default modules stripped via |
28 |
INSTALL_MASK including the one pulling in kmod since I'm running |
29 |
monolithic kernel and have kmod package.provided, and USE=btrfs plus |
30 |
adding it to the module config. |
31 |
|
32 |
-- |
33 |
Duncan - List replies preferred. No HTML msgs. |
34 |
"Every nonfree program has a lord, a master -- |
35 |
and if you use the program, he is your master." Richard Stallman |