| 1 |
On Sat, Sep 25, 2010 at 8:12 AM, Joseph Jezak <josejx@g.o> wrote: |
| 2 |
<SNIP> |
| 3 |
> None of these things are weird, they're expected! |
| 4 |
> |
| 5 |
|
| 6 |
Really? OK - cool then. |
| 7 |
|
| 8 |
> When you start the system with "init=/bin/bash", the only thing your |
| 9 |
> system does is start bash as the only process. Things like mounting |
| 10 |
> proc, remounting the root file system r/w (and setting up mtab) aren't |
| 11 |
> done because they're part of the normal startup process. |
| 12 |
> |
| 13 |
|
| 14 |
I need to find a good _SIMPLE_ book on how Linux boots. I'm just a |
| 15 |
long time use with no real sys admin background so for a decade or |
| 16 |
longer I just want the things go by at boot and then use my system. |
| 17 |
|
| 18 |
> Can you run fsck.ext3 from the shell that starts when you replace init? |
| 19 |
> |
| 20 |
|
| 21 |
I cannot due to the same sort of message I get when trying to boot: |
| 22 |
|
| 23 |
e2fsck_check_if_mount: Can't check if filesystem is mounted due to |
| 24 |
missing mtab file |
| 25 |
fsck.ext3: Device or resource busy while trying to open /dev/hda4 |
| 26 |
Filesystem mounted or opened exclusively by another program? |
| 27 |
|
| 28 |
For kicks I tried umount /dev/hda4 but that complains about the |
| 29 |
missing mtab file also. It seems from the man page I could force it |
| 30 |
but I didn't want to do that, and again if I did and it even worked |
| 31 |
then where would the system read fsck.ext3 from to do the check since |
| 32 |
it's on /dev/hda4 itself? |
| 33 |
|
| 34 |
> Also, which version of udev are you using? I forgot to ask if you're |
| 35 |
> running ~ppc instead of ppc, you may want to try switching to the |
| 36 |
> PATA_MACIO driver in order for the system to work properly. |
| 37 |
|
| 38 |
udev would be whatever comes with a stable ppc system these days. |
| 39 |
|
| 40 |
Other than running ~ppc portage and sandbox the machine is currently |
| 41 |
all stable with no USE flags. The make.conf file is pretty much empty |
| 42 |
other than the stuff the install has me put in. (Which mirrors to sync |
| 43 |
from, etc.) |
| 44 |
|
| 45 |
I've gone back and done a couple more experiments: |
| 46 |
|
| 47 |
1) So far anything I try that refers to the drive as /dev/sda4 results |
| 48 |
in a VFS not syncing type error so at least as far as the boot process |
| 49 |
goes hda4 seems to be the right device. |
| 50 |
|
| 51 |
2) I tried building in the PATA_MACIO driver but that by itself didn't |
| 52 |
change anything: |
| 53 |
|
| 54 |
(chroot) livecd / # cat /usr/src/linux/.config | grep MACIO |
| 55 |
CONFIG_PATA_MACIO=y |
| 56 |
CONFIG_ADB_MACIO=y |
| 57 |
(chroot) livecd / # |
| 58 |
|
| 59 |
The boot still fails at the same point with the same message. Possibly |
| 60 |
some other ATA driver is being selected ahead of this one and maybe it |
| 61 |
needs to be removed from the kernel? I don't know what to remove |
| 62 |
though so I'm sort of stuck on that account. |
| 63 |
|
| 64 |
3) If this is a kernel config issue - and it seem plausible that it is |
| 65 |
- then I'll point out that the kernel config page of the Gentoo PPC |
| 66 |
Istall Guide had a possible command that didn't work for me: |
| 67 |
|
| 68 |
http://www.gentoo.org/doc/en/handbook/handbook-ppc.xml?part=1&chap=7 |
| 69 |
|
| 70 |
Specifically: |
| 71 |
|
| 72 |
make pmac32_defconfig |
| 73 |
|
| 74 |
which supposedly will make a config file that would boot most 32-bit |
| 75 |
machines. Unfortunately the command doesn't work for me on current |
| 76 |
kernels. The install guide should be updated to either remove this or |
| 77 |
make it more clear if I need to do anything more than that command in |
| 78 |
the /usr/src/linux directory. |
| 79 |
|
| 80 |
I've looked around at lots more stuff in my kernel config and compared |
| 81 |
it to what the Install guide says. I don't see any differences but I |
| 82 |
suspect there probably are. |
| 83 |
|
| 84 |
Thanks! |
| 85 |
|
| 86 |
- Mark |
Archives