initramfs / UUID problems after cloning

Question

So I just cloned a Centos7 "everything install" VM to a physical disk using VMWare Workstation and Clonezilla using the instructions here:
https://www.howtoforge.com/converting-a-vmware-image-to-a-physical-machine-p2

Now, when the clone (physical disk) boots up, it drops into the command line with this:

dracut-initqueue[417]: Warning: Could not boot. dracut-initqueue[417]:
Warning: /dev/disk/by-uuid/4b8b5627-9027-46f7-b722-61d3ddd28e74 does
not exist Starting dracut emergency shell...

What I've tried:

I checked /etc/fstab and the proper UUIDs are there, including the one above.

Tried booting into USB rescue mode, chroot /mnt/sysimage , and:

mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r)-oh_crud.img

and then

dracut /boot/initramfs-$(uname -r).img $(uname -r)

...but I get this:

cat: write error: Broken pipe

Additional notes:

When I enter:
```
blkid
```
I see the HD in question showing up as both /dev/block and /dev/sdb`
I tried:
```
lshw -class disk
```
but it says "command not found".
The rescue shell is mounting my disk in /mnt/sysimage and all my files are there, so it appears it is seeing and mounting the HD.

mikeserv- I don't understand. The UUID for the disk shown by "blkid" matches what is in /etc/fstab. What do you suggest I change the UUID to? — Are Vee, Feb 09 '15 at 16:41
mikeserv- the source disk (the disk that was cloned) is a VM and does not exist on this machine, so why would there be a UUID conflict? — Are Vee, Feb 09 '15 at 16:46
Well, I'm out of ideas. Sorry. The best suggesttion I can make is rebuilding initramfs on the baremetal clone with `dracut`. — mikeserv, Feb 09 '15 at 17:21
I would guess you actually are missing a kernel module, so the hardware isn't showing. Do you have any disks showing up in your rescue shell? Are there any other error messages, further up? — derobert, Feb 09 '15 at 17:54
@derobert - ahh, yeah! don't `vmware` disks mount as some kind of `vmware` block dev with the kernel headers they insert? Or maybe thats `vbox`... still, I think `vmware` has those *guest modules* or whatever - and it would be strange to include much else but those in a kernel targeted for virtual machine install. I think you're probably right on. — mikeserv, Feb 09 '15 at 20:32
derobert- how do I determine if any other disks are showing up in my rescue shell (note I am booting into rescue shell from USB installer if that matters)? I see the HD in question showing up when I type "blkid" as both /dev/block... and /dev/sdb...I tried "lshw -class disk" but it says command not found. Finally, the rescue shell says it is mounting my disk in /mnt/sysimage and all my files appear to be there. — Are Vee, Feb 10 '15 at 06:15
EDIT: original post updated to indicate rescue shell is mounting my HD. — Are Vee, Feb 10 '15 at 19:39
Did you ever try just using /dev/sdb1 or whichever instead of the UUID? — Nathan V, Sep 09 '15 at 22:33

score 1 · Answer 1 · edited Apr 13 '16 at 14:32

I had a similar issue and found that running uname -r didn't return the correct kernel version under rescue mode that I was using normally.

I had to check what initramfs file grub was trying to boot from by pressing e at the grub boot screen and then once in rescue mode do something like this -

cd /boot
initramfs -f initramfs-3.10.0-327.4.4.el7.x86_64.img 3.10.0-327.4.4.el7.x86_64

uname -r in my case was just returning "3.10.0-327.el7.x86_64" which wasn't the kernel version I was trying to load from grub (3.10.0-327.4.4).

I wonder if your lack of the -f switch or a mismatch of the kernel versions used by rescue mode and normal mode may be your issue here

initramfs / UUID problems after cloning

1 Answers1