On Thu, 11 Feb 2010, Bruce Ward wrote:
Greetings.
More data. I have managed to get an initramfs debug out of the
system and have also provided a debug from a 2.6.26-1-686 kernel
which boots OK. I'm damned if I can see any reason from the debug
for the recent kernel (2.6.26-2-686) to fail to boot!
Interesting (but possibly not relevant) observations from my playing
around (exploration). Remember there are only a few modules (thermal
ones) loaded when the system drops out.
1. Load USB drivers - they find the pendrive, but no entry created in /dev.
2. Use mknod to create /dev/sda3 for the ext3 partition on pendrive.
sda3 is the only SCSI entry in /dev.
3. Load ext3 module, and mount the sda3 partition and, behold, there
are entries in /dev/ for sda, sda1, and sda2, as well as sda3.
4. Load the SATA drivers and suddenly there are sdb entries in /dev
for all partitions on the drive I have failed to boot from.
that points a bit to udev is it running at this stage?