Results 1 to 3 of 3
I have a virtual machine setup to dual boot two different Linux systems
I have a disk with the following partitions:
Primary 1 /dev/sda1 ext4 for /boot partition of multiple ...
Enjoy an ad free experience by logging in. Not a member yet? Register.
- 03-08-2012 #1Just Joined!
- Join Date
- Mar 2012
- Posts
- 3
mbr reference to grub issue
I have a virtual machine setup to dual boot two different Linux systems
I have a disk with the following partitions:
Primary 1 /dev/sda1 ext4 for /boot partition of multiple OSs
Primary 2 /dev/sda3 ext for data transfer purposes--not mounted at boot
Extended 1 /dev/sda5 ext4 for / partition of unbuntu 32bit variant OS
Extended 1 /dev/sda6 ext4 for / partition of debian 64bit variant OS
Extended 1 /dev/sda7 ext4 for /data partition to be mounted on both OS
Extended 1 /dev/sda8 linux-swap
I'm able to repeatedly boot both OSs from grub on /dev/sda1 with no issues.
My objective is to get this working environment redeployed to physical hardware that will have the same partition setup, but the partitions will be of a larger size.
To prepare for the re-provisioning I have booted from a rescue CD (w w w.sysresccd.org), have mounted the three partitions and tared up the contents with tar -cvpzf
On the physical hardware I've done the following:
1. Boot from the aforementioned rescue CD
2. Used gparted to replicate the partition setup and format the partitions with /ext4 like the original installation.
3. Mounted the partitions to restore files
4. copied the tarfiles over, gunzipped them and untared them using -xpvf
5. Verified that the files and directories appear to be restored correctly
This is the point that I'm kinda stuck at. I know that I need to prepare the MBR and get it to load grub from /dev/sda1 that has the linux kernals and grub on them.
I've looked at serveral articles about reinstalling grub, and although I don't think I need to actually reinstall it, I thought that this article (w w w.sysresccd.org/Sysresccd-Partitioning-EN-Repairing-a-damaged-Grub) would be nearly exactly what I'm looking to do, but I'm not successful.
In my case I sda5 is equivalent to sda2 from the article, but when I boot with rescuecd root=/dev/sda5, I get a kernel panic. Screenshot at harm.net/tempimages/kernelpanic.jpg
Seems like there should be a simpler way to tell the MBR to run grub on /dev/sda1 and get me going here.
Thoughts, advice?
Thanks!
- 03-08-2012 #2Just Joined!
- Join Date
- Mar 2012
- Posts
- 3
I got a little further.
I tried Section 2 in that document and got it so that grub would load, and I had to correct the root= parameter to accommodate for a slightly different naming convention.
From: set root='(hd0,msdos1)'
To: set root='(hd0,1)'
Now I get the start of a boot, but things just sit there after this message:
udhcpc[598]: Lease of 10.0.2.15 obtained, lease time 86400
...then nothing.
- 03-08-2012 #3Just Joined!
- Join Date
- Mar 2012
- Posts
- 3
had to modify /etc/fstab to not use uuids from old disk structure.
All good now.


Reply With Quote
