Take the 2-minute tour ×
Server Fault is a question and answer site for professional system and network administrators. It's 100% free, no registration required.

I have a cloud server hosted at Rackspace, and they use Xen Server hypervisor. I was running Ubuntu 12.04, and really messed up by attempting to do a release upgrade manually.

The hypervisor apparently doesn't get along with the changes made to GRUB (I think; there could be more) and thus does not reboot; it gets stuck at the initramfs stage of booting:

http://i.imgur.com/Bzx6I36.png

I've been speaking with a tech for a long time. We're rather undecided on what to do; would it make sense to provision a new server, grab that kernel / GRUB config, and drop it on the current server? Or to try and roll back the changes / downgrade?

If you have any suggestions, they would be greatly appreciated.

share|improve this question

2 Answers 2

Not really a xen thing. The operative bit is /dev/disk-by-uuid/dev/xvda1 does not exist.

  1. Access grub.conf by standard means (e.g. boot rescue or mount the partition/file/whatever)
  2. see if thingies like root=UUID=/dev/xvda1 appear and change them to root=/dev/xvda1
  3. check that /etc/fstab also complies e.g it reads:

    /dev/xvda1 / ext3 defaults,errors=remount-ro 0 1

and not:

UUID=/dev/xvda1 /   ext3 defaults,errors=remount-ro 0       1

That should be it.

Hint: unless the "tech" you are talking with has already tried this, he/she is out of his/her depth. This is in Google's first page, and it's Rackspace related to boot ;-) Cheers, alf

EDIT

Having seen your (grub2) config I would try to replace it with something along the lines of:

timeout=5


menuentry 'Ubuntu, with Linux 3.2.0-24-generic' {
#not sure about the following line
root=xvda,1
linux   /boot/vmlinuz-3.2.0-24-generic root=/dev/xvda1 ro console=hvc0
initrd  /boot/initrd.img-3.2.0-24-generic
} 

And then, I'd get down to find out a way to make it permanent, because, on the next upgrade grub-mkconfig will hose your config again. Here you find a different strategy which might also work not to mention advice that should probably be heeded.

share|improve this answer
    
Do you mean /boot/grub/grub.cfg? –  tekknolagi May 10 '13 at 14:12
    
Also, there's no such line in the grub conf. I can upload mine / take a screenshot if you like. –  tekknolagi May 10 '13 at 14:14
    
Are these lines related? pastebin.com/my5KqSmG –  tekknolagi May 10 '13 at 14:18
    
If yours is called grub.cfg, then we're likely dealing with grub2 (which I do not know much, or at all). Seeing your cfg would help and ye, the lines you posted are probably involved. –  Alien Life Form May 10 '13 at 14:47
    
I shall post my config promptly, then. Do you think you can help? –  tekknolagi May 10 '13 at 21:01
up vote -1 down vote accepted

So the problem wasn't particularly correctly analyzed by anyone here, but the higher-tier Linux Admins at Rackspace were instrumental in solving this problem.

I replaced my GRUB2 conf with what Alien Life Form gave me, but changed the root line to:

set root=(hd0)

Then, in my menu.lst, I removed the by-uuid line, so that it could find my drive.

From there, it booted properly.

Thank you all!

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.