Hi we completed the lfs but after rebooting it showed the error

VFS:cannot open device "hda6" or unknown block(3,6)
please append a correct "root=" boot option
kernel panic:VFS:unable to mount root fs on unknown block(3,6)

Our grub file menu.lst is given below

# Begin /boot/grub/menu.lst
# By default boot the first menu entry.
default 0
# Allow 30 seconds before booting the default.
timeout 30
# Use prettier colors.
color green/black light-green/black
# The first entry is for LFS.
title LFS 6.0
root (hd0,5)
kernel /boot/lfskernel-2.6.8.1 root=/dev/hda6
title Red Hat
root (hd0,0)
kernel /boot/kernel-2.6.1-358 root=/dev/hda2
initrd /boot/initrd-2.6.1-358

And details of our system partions are
device boot start end blocks id system
/dev/hda1 * 1 203 102280+ 83 linux
/dev/hda2 204 30679 15359904 83 linux
/dev/hda3 30680 31199 262080 82 linux swap
/dev/hda4 31200 54372 11679192 5 extended
/dev/hda5 31200 31700 252472+ 83 linux swap
/dev/hda6 31701 45000 6703168+ 83 linux

Our lfs is mounted on partition hda6 which is in hda4(this is an extended partition).

So my question is in the command "kernel /boot/lfskernel-2.6.8.1 root=/dev/hda6"
Am i supposed to write root=/dev/hda6 or /dev/hda4?

And are the entries in my grub file correct???
also in the fstab file
# file system mount-point type options dump fsck order


/dev/[xxx] / [fff] defaults 1 1
/dev/[yyy] swap swap pri=1 0 0

in place of [xxx] shall i write hda4 or hda6?in our fstab file i wrote hda6.

Please help!!!

the grub entries look correct. in your kernel, did you make your file system a module instead of inside the kernel? not too sure about lfs, but in other distro's, you need the incomplete driver thing option, the first one, checked as well.

A good place to begin would be to reboot in your host system and see if you can mount /dev/hda6 from there. If so, check what filesystem-type it is using, and see if any modules are required to use that particular filesystem. For obvious reasons, support for your root-filesystem (and "/boot") must be compiled into the kernel.

If your host system can mount the device, dismount the disk and run fsck against it from the host. This will rule out the possibility that the filesystem has errors.

Does a device-node exist for this block device?

The code is in /init/do_mounts.c "Use the source, Luke!"