| 1 |
On Sunday 30 September 2007, Siong-Ui Te wrote: |
| 2 |
> On 9/30/07, Mike Frysinger <vapier@g.o> wrote: |
| 3 |
> > which version of baselayout ? |
| 4 |
> |
| 5 |
> sys-apps/baselayout-1.12.9-r2 |
| 6 |
|
| 7 |
probably wont be backporting a fix to this, but it probably isnt fixed |
| 8 |
already ... |
| 9 |
|
| 10 |
> > can you post the actual output please ? |
| 11 |
> |
| 12 |
> Remounting root filesystem read/write |
| 13 |
> Root filesystem could not be mounted read/write :( |
| 14 |
> |
| 15 |
> I already found the related code snippet in /etc/init.d/checkroot, |
| 16 |
> still trying to figure out how to avoid this. |
| 17 |
|
| 18 |
checkroot should not attempt to remount rw a filesystem that is already rw ... |
| 19 |
but it probably moved it to ro upon boot ... |
| 20 |
|
| 21 |
> > right ... we should properly account for this ... |
| 22 |
> |
| 23 |
> What I am trying to do is to build entire Gentoo base system in the |
| 24 |
> initramfs. |
| 25 |
> So far the most difficult part is to adapt the baselayout design to "no |
| 26 |
> real root" |
| 27 |
> nature of initramfs. Any suggestions? |
| 28 |
|
| 29 |
we already have logic in there for not doing fsck if the root is net based |
| 30 |
(like nfs) ... perhaps we should expand it slightly for the initrd/initramfs |
| 31 |
case ... |
| 32 |
-mike |
Archives