" 'initrd.img' not found" after a kernel upgrade hang (& a re-boot)

[watchpocket]

So a few days ago I hit the software updater from my drive #2 (running Precise), and some kernel files were set to be upgraded, I think to 3.13.0-58.

Then an Adobe Flash installer started to upgrade, but hung.

And hung, and hung. There seemed to be no way to exit gracefully from the software updater, or to get out of it at all.

So I re-booted. But it seems the kernel upgrade process hadn't completed, so when I next tried to log in, I selected my #2 drive from the GRUB2 menu, and saw this:


error: file '/initrd.img' not found. Press any key to continue

Upon pressing any key, big BIOS-type lettering threw out a bunch of stuff, including:


kernel panic - not syncing: VFS: unable to mount root fs on unknown-block (0,0)

CPU: 1 PID: 1: Comm: swapper/0 Not tainted 3.13.0.57-generic #95~precise1-Ubuntu

Call Trace: dump_stack+0x46/0x58

panic+ [numbers]
mount_block_root [numbers]
mount_root [nos]
prepare_namespace [nos]
kernel_init [nos]
ret_from_fork [nos]
rest_init [nos]

I have 3 drives. Drive #1 runs Ubuntu-MATE 14.04; drive #2 runs Precise; and drive #3 is for storage. I have BIOS booting, not EFI.

From drive #1, I can mount drive #2 and access all of drive #2's files with no problems.

But I cannot now boot into drive #2.

When I look in the root dir of my good drive (drive #1), I see this file:

initrd.img 19.2 MB Link to Raw disk image application/x-raw-disk-imageBut when I look in the root dir of my "problem" drive (drive #2), I see this:
initrd.img 33 bytes link (broken) inode/symlink My question: How can I repair this to return to normal booting from my Precise drive (drive #2)?

Some command output:
sudo blkid
/dev/sda1: LABEL="UbuntuMATE-14.04" UUID="1742aa67-b8ce-45d2-94ca-05e3579f7e2f" TYPE="ext4"
/dev/sda2: UUID="a88c33fd-88b5-43b2-92f6-457430d206fd" TYPE="swap"
/dev/sdb1: LABEL="Ubuntu-12.04.5" UUID="4d4f7665-f633-4409-82f9-af43d0143823" TYPE="ext4"
/dev/sdb2: UUID="106dae89-e837-4063-9f9c-bcdeae4b2d5a" TYPE="swap"
/dev/sdc1: LABEL="WeirdBeard" UUID="c801726e-7a01-4963-989a-38482e6658cf" TYPE="ext4"
------------------------------------------------------------------------------------------

sudo fdisk -l
Disk /dev/sda: 1000.2 GB, 1000204886016 bytes
255 heads, 63 sectors/track, 121601 cylinders, total 1953525168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x3aec6e77

Device Boot Start End Blocks Id System
/dev/sda1 * 2048 1932552191 966275072 83 Linux
/dev/sda2 1932552192 1953525167 10486488 82 Linux swap / Solaris

Disk /dev/sdb: 1000.2 GB, 1000204886016 bytes
255 heads, 63 sectors/track, 121601 cylinders, total 1953525168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x000b03b5

Device Boot Start End Blocks Id System
/dev/sdb1 2048 1932554239 966276096 83 Linux
/dev/sdb2 1932554240 1953525167 10485464 82 Linux swap / Solaris

Disk /dev/sdc: 1000.2 GB, 1000204886016 bytes
255 heads, 63 sectors/track, 121601 cylinders, total 1953525168 sectors
Units = sectors of 1 * 512 = 512 bytes
Sector size (logical/physical): 512 bytes / 512 bytes
I/O size (minimum/optimal): 512 bytes / 512 bytes
Disk identifier: 0x41031e20

Device Boot Start End Blocks Id System
/dev/sdc1 2048 1953523711 976760832 83 Linux
---------------------------------------------------------------------------------


sudo xxd -l 2 -p /dev/sda <-- Replacing "sda" with "sdb" here (& just below) gives the same output.
eb63

sudo xxd -s 1049 -l 2 -p /dev/sda
ffff


Boot Info Script output is here:
(http://paste.ubuntu.com/11865186/plain/)http://paste.ubuntu.com/11865226/

Thanks loads for any help.

[dino99]

if you have at least two kernels installed, boot with the working one, otherwise goto the 'recovery' process, then fix that Adobe trouble:

http://askubuntu.com/questions/603295/how-to-fix-dpkg-error-2

[watchpocket]

> if you have at least two kernels installed, boot with the working one

Well, I need to be able to boot from both drives.

Would it be a mistake to simply copy the initrd.img file from my Trusty root directory (on drive #1) to my Precise root dir (on drive #2), where it appears there is no working initrd.img file?

> goto the 'recovery' process

I did boot into recovery mode on drive 2, if that's what you mean. That returned roughly the same "BIOS"-looking big-print info I quoted above.

Also I ran these three commands from the Trusty disk:

sudo fsck /media/rj/Ubuntu-12.04.5
sudo e2fsck -b 8193 /media/rj/Ubuntu-12.04.5
sudo e2fsck -b 32768 /media/rj/Ubuntu-12.04.5

All three returned this:

e2fsck 1.42.9 (4-Feb-2014)
e2fsck: Is a directory while trying to open /media/rj/Ubuntu-12.04.5

The superblock could not be read or does not describe a valid ext2/ext3/ext4
filesystem. If the device is valid and it really contains an ext2/ext3/ext4
filesystem (and not swap or ufs or something else), then the superblock
is corrupt, and you might try running e2fsck with an alternate superblock:
e2fsck -b 8193 <device>
or
e2fsck -b 32768 <device>

I've forgotten how to run fsck on drive 2 if that drive isn't mounted. Can that only be done from a USB stick?

[watchpocket]

Can an initrd.img file be copied from one OS to a previous one? Is this a bad idea? Are they specific to the OS?

My Trusty has kernal version 3.13.0-58. My precise has -57.


> then fix that Adobe trouble

To do that I can just go into synaptic and remove flashplugin-installer and re-install it, no?

[watchpocket]

For anyone else having problems with a corrupted or broken initrd.img, the most useful info I've found so far is here:

http://www.thegeekstuff.com/2009/07/how-to-view-modify-and-recreate-initrd-img/