Il 31/03/2016 13:43, Emmanuel Kasper ha scritto:
Hi again
As Alessandro said this is probably a mismatch of the root device.
Do you have your old VM running on VmWare ?
If yes please note there the value of the root device in /proc/cmdline
it should be something like root=/dev/mapper/susa--vg-root
this/dev/ something should be a path to your root file system and
should appear when you do
ls -l /dev/mapper/susa--vg-root
lrwxrwxrwx 1 root root 7 Mar 31 09:31 /dev/mapper/susa--vg-root -> ../dm-0
Now when running the VM in Proxmox, after the error you see, you should
be dropped to a rescue shell
in this rescue shell, check if the same 'ls -l' command return something
if 'ls' returns nothing,
tries the command vgchange -ay
and try 'ls -l ' again
if ls still fails, please run ls -l/dev/*da in the rescue shell and
post the output here
I think that the rescue wont come up as the kernel is panicing (cannot
find root device)
This is what I'd do:
Set the disk controller as IDE
EITHER
Boot the VM with a CentOS / RHEL 4 iso in rescue (linux rescue)
and Fix the /etc/fstab (if needed) and the /boot/grub/grub.conf
to point to the correct device
OR
if you know where the root partition is you might try with the "e"
option in grub when booting and change the root=/dev/xxx
With the ide controller disk 1 should be /dev/hda
but that depends if you are using LVM or direct device, and how many
partitions there are.
I think that the initrd/kernel should already contain the drivers for
the IDE controller, so no need to mess with modprobe.conf and initrd for
now. (sata would need the ahci driver)
Maybe if you post what the root=/dev/xxx is (with the e option when grub
appears) this might give us some more info.
Hope this helps,
Alessandro
_______________________________________________
pve-user mailing list
[email protected]
http://pve.proxmox.com/cgi-bin/mailman/listinfo/pve-user
