| 1 |
On Wed, 18 Jan 2006, Paul Heinlein wrote: |
| 2 |
|
| 3 |
> ----- %< ----- |
| 4 |
> * Checking root filesystem ...fsck.ext3: No such file or directory while |
| 5 |
> trying to open /dev/hda5 |
| 6 |
> /dev/hda5: |
| 7 |
> The superblock could not be read or does not describe a correct ext2 |
| 8 |
> filesystem. If the device is valid and it really contains an ext2 |
| 9 |
> filesystem (and not swap or ufs or something else), then the |
| 10 |
> superblock is corrupt, and you might try running e2fsck with an |
| 11 |
> alternate superblock: |
| 12 |
> |
| 13 |
> e2fsck -b 8193 <device> |
| 14 |
> * Filesystem couldn't be fixed :( |
| 15 |
> ----- %< ------ |
| 16 |
> |
| 17 |
> Sure enough, there was no /dev/hd* in the rootfs -- though the long |
| 18 |
> devfs representations were there: /dev/ide/host0/... I edited |
| 19 |
> /etc/fstab to point to them instead of /dev/hda*, and everything was |
| 20 |
> OK. |
| 21 |
|
| 22 |
Follow on... |
| 23 |
|
| 24 |
I was uncomfortable with hard-coding /dev/ide/... in fstab, so I ended |
| 25 |
up adding labels to all filesystems and swap partitions and editing |
| 26 |
/etc/fstab to use LABEL= for device specification instead. That should |
| 27 |
be a little more portable should I find a stable 2.6 kernel and move |
| 28 |
to udev. |
| 29 |
|
| 30 |
I'm still stumped why the /dev/hd* devices are no longer available to |
| 31 |
/etc/init.d/checkroot. |
| 32 |
|
| 33 |
-- |
| 34 |
Paul Heinlein <> heinlein@××××××.com <> www.madboa.com |
| 35 |
-- |
| 36 |
gentoo-sparc@g.o mailing list |
Archives