mkinitrd & initial RAM disk issues

OK so I am trying to restore a, identical in all ways except GPU, workstation (HP4400) from a "dd" image of an identical workstation. The image is stored on a SANS device and mounted in knoppix. I did notice that the Nvidia GeForce 7600 GPU isn't allowing knoppix to boot into runlevel 5 by default, I need to use

# knoppix 3

Anyway, he is how the image is created and restored:

[1] CREATED

# dd if=/dev/sda bs=1k conv=sync,noerror | gzip -c | ssh dx1 "dd \ of=/awips/GFESuite/xtHP.dd.gz bs=1k"

[2] RESTORED

# dd if=/awips/GFESuite/xtHP.dd.gz | ssh xt1 "gunzip -c | \
dd of=/dev/sda bs=1k"

see:

http://www.inference.phy.cam.ac.uk/saw27/notes/backup-hard-disk-partitions.html

for more information

=====

So the problem arises once the restore is done and the restored box comes up -- it passes thru GrUB just fine
then onto the Decompressing of the kernel and loading of the initrd, this is where it bombs-out:


echo Scanning logical volumes
lvm vgscan --ignorelockingfailure
echo Activating logical volumes
lvm vgchange -ay --ifnorelockingfailure vg00
echo Creating root device
mkrootdev /dev/root
umount /sys
echo Mounting root filesystem
mount -o defaults -ro -t ext3 /dev/root/sysroot
mount -t tmpfs --bind /dev /sysroot/dev
echo Switching to new root
switchroot /sysroot
umount /initrd/dev

So you can see, root=/dev/vg00/lvol01 and it has issues finding the volume group -- however after
uncompressing the initrd, all looks fine, dm-mod and various dm modules needed are loaded just
fine, and you see this just before the above error, also, if you boot into knoppix and load the modules
that are part of the initrd image, and use them you can vgchange and mount just fine.

So I did a lot of things, all to no avail:

# mkinitrd -v --preload=ata_piix initrd-2.6.9-42.19.ELsmp.img   \
2..6.9-42.19.ELsmp

That didn't work either so I decided to heck with it and edited the init
script inside initrd to do the below, and I copied in the
/sbin/lvm.static to the bin directory of the image:

mkdmnod
/bin/lvm.static vgscan --mknodes --ignoreloackingfailure
/bin/lvm.static vgchange -a y --ignorelockingfailure

no go with those, so tried in rescue mode to make an initrd and preload the hdd modules thinking it maybe can't
read it, as usually is the case, but no go with this either:

# chroot /mnt/sysimage
# mount /boot
# mv [existing initrd image]
# mkinitrd --preload mptbase --preload mptscsih --preload ata_piix \
  /boot/[initrd image name]
# unmount /boot
# exit


ANYONE? ;-)