You got right to the point – the reason of kernel panic is that kernel can’t find RAID device node for mounting it’s root file system…
However, you shouldn’t create any “virtual RAID” for your guest VM – RAID is a “real hardware” business and the host should care about it. You should do at least two things before migrating the server to VM:
(1) edit /etc/fstab so all references to whatever RAID device you have will point to a /dev/hd* or /dev/sd* node, which you’ll create/attach in the VM;
(2) instruct kernel to look for the root file system in a correct place – that is virtual drive /dev/hd* or /dev/sd*, as in (1). The exact procedure depends on your bootloader – with GRUB, just edit /boot/grub/menu.lst (or /boot/grub/grub.conf, whichever you have) and add root=/dev/hd.. to the “kernel” line, and you are done; with LILO – edit /etc/lilo.conf to change the device on root= line , AND then run lilo to update the boot record (if your Red Hat is old enough, you may have LILO).
If you have any other references to the RAID device node, change them also (chances are that you have no other references to real devices)
Make sure NOT to reboot the real server with this changes made – you’ll get kernel panic, of course; now you can make backup of the drive(s), restore them in VM, and you have to be OK (hopefully).