Gfilui

I tried to expand a VMWARE ESXI GUEST LINUX DISK without success.

After changing the size of the disk on the vSphere client, I rebooted the server and then lost the disk.

VMWARE GUEST says at the BOOT:

NO OPERATING SYSTEM FOUND

The weirdest thing is that I can't see any disk at the BIOS, but after rebooting with a Ubuntu live CD, I'm able to see SDB1 and SDA.

After the live boot:

Note that the SDA1 has a sub partition called SDA1P1 and SDA1P2. Another strange thing is that the first sector starts at 0 (ZERO).

So whenever I try to mount this unit it throws some errors out:

mount: mounting /dev/sda1 on /mnt/disk iled: invalid argument

And then I'm unable to mount the disk.

I tried some other stuff, like exporting the .VMDK file to another linux box.

So I had to mount a loopback block-device on my linux machine in order to see the content of the .VMDK.

To do that, I tried:

# kpartx -av "MYBOX.vmdk";
add map loop0p1 (252:2): 0 62814150 linear /dev/loop0 0
add map loop0p2 (252:3): 0 96390 linear /dev/loop0 62814150

Then I was capable to see the VMDK inside my linux box.

# fdisk -l /dev/loop0

Disk /dev/loop0 (Sun disk label): 255 heads, 63 sectors, 3916 cylinders
Units = sectors of 1 * 512 bytes

 Device Flag Start End Blocks Id System
/dev/loop0p1 0 62814150 31407075 83 Linux native
/dev/loop0p2 u 62814150 62910540 48195 82 Linux swap

Now I got really excited! I can almost touch the disk, its juts a matter of mounting it, right?

Nope! :/

# mount /dev/mapper/loop0p1 /mnt/
mount: you must specify the filesystem type

Ok, it doesn't sound pretty good.

So I tried to check the superblocks on /dev/mapper/loop0p1:

# mke2fs -n /dev/mapper/loop0p1
mke2fs 1.42.9 (4-Feb-2014)
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=0 blocks, Stripe width=0 blocks
1966080 inodes, 7851768 blocks
392588 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=4294967296
240 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Superblock backups stored on blocks: 
 32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208, 
 4096000

Then I tried to run a FSCK on those blocks:

# e2fsck -b 32768 /dev/mapper/loop0p1 
e2fsck 1.42.9 (4-Feb-2014)
e2fsck: Bad magic number in super-block while trying to open /dev/mapper/loop0p1

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 actually got pretty far, but still no success.

Any help would be much appreciated.

UPDATES ON 03-27-18

I've been doing a lot of research lately.

Between a lot of things that didn't take me further, here's what I've done and got positive results.

mkfs.ext4 -S /dev/sdb1 -y

By doing this, I re-made the disk's filesystem. Nice, but very risky (do not do this if you don't have a copy of this disk -- when I did it, it was on a copied image). Please, read MKFS man page regarding the usage of "-S" option.

So OK, I got pretty far guys. I can almost see the light at the end of the tunnel. But before that, we need to do a full scan on the disk.

Let's do it!

fsck.ext4 /dev/sdb1 -y

Then I got a lot of output, basically the disk was entirely corrupted.

At this point, I've tried a lot of technics to recover just 10 files from the disk. Over a week working very hard on this.

Well, let's keep walking!

After all the output, I tried to mount the disk and then.. tchaaaannaaan!

mount /dev/sdb1 /mnt/

The disk was mounted.

Looking at its content, I can see lots of weird files and folders inside "lost+found" folder. Some of them starts with "#". I need more time to study this but basically, all human visible content was gone.

When I say BASICALLY, I really mean it.

I could find about 8GB of data on this VMDK image.

Now I've gotta figure out a way to find my miserable 10 files and re-install everything on another server.

Cheers!

-- If its not done yet, its 'cause it's not finished.