Florian Philipp wrote:
Am Dienstag 14 August 2007 20:37:59 schrieb Neil Bothwick:
Hello Florian Philipp,
3. mount root and boot
4. mount their mount point on my desktop via sshfs
5. create a tar ball
6. unmount everything, create reiserfs, remount everything
7. extract tar ball, edit fstab
8. reboot
Now I have the following problem: If I boot, the kernel starts normally
(from what I see) but then tells me that it is "unable to start initial
console" (if I remember correctly) and reboots after a few seconds.
When you created a tarball of the root partition, you didn't include the
contents of the dev directory (not the dev filesystem mounted on it). The
safest way to do this is mount root on another mount point, with "mount
--bind / /mnt/root" then tar up /mnt/root.
Alternatively, mknod /dev/console and /dev/null on the new filesystem.
I've solved this problem and ran into number two while booting:
"Checking root filesystem ...
Reiserfs super block in block 32 on 0x807 of format 3.6 with non-standard
journal
Blocks (total/free): 7863776/5293202 by 2048 bytes
Filesystem is clean
Filesystem seems mounted read-only. Skipping journal replay.
Checking internal tree .. finished
Remounting root filesystem read/write ...
Root filesystem could not be mounted read/write :(
Give root password for maintenance
(or type Control-D to continue):"
I think it's due to the fact that I increased the journal size by the factor
1.5. I'll check this but would like to know if there might be another reason.
It's like Florian said: the new root doesn't have the /dev/null and
/dev/console device entries. These aren't managed by udev and have to be
created manually. When you install Gentoo, they're in the tarball for
your stage; however, just tarring a root won't keep them by default. You
can solve this problem (again, like Florian said) by doing:
# mknod /dev/null c 1 3
# mknod /dev/console c 5 1
--
[EMAIL PROTECTED] mailing list
