Libguestfs - Feb 2015 - Re: Mounting disk images with ext2 filesystems on RHEL7

On Wed, Feb 18, 2015 at 9:47 AM, Richard W.M. Jones <rjones@redhat.com>
wrote:
> On Wed, Feb 18, 2015 at 09:41:52AM -0500, Leonard Basuino wrote:
> > On Wed, Feb 18, 2015 at 6:52 AM, Richard W.M. Jones
<rjones@redhat.com>
> > wrote:
> >
> > > On Tue, Feb 17, 2015 at 01:37:55PM -0500, Leonard Basuino wrote:
> > > > I get the following debug info:
> > > >
> > > > mount -o /dev/sda1 /
> > > > [ <time> ] EXT4-fs (sda1): mounting ext2 file system
using the ext4
> > > > subsystem
> > > > [ <time> ] EXT4-fs (sda1): bad geometry: block count
104388 exceeds
> size
> > > of
> > > > device (103408 blocks)
> > > > mount: wrong fs type, bad option, bad superblock on
/dev/sda1
> > >
> > > It looks as if the image is properly corrupt.  My suggestion
> > > is to try 'virt-rescue' on it.
> > >
> > > virt-rescue --suggest -a <image>
> >
> > Inspecting the virtual machine or disk image ...
> >
> > This disk contains one or more filesystems, but we don't recognize
any
> > operating system.  You can use these mount commands in virt-rescue (at
> the
> > ><rescue> prompt) to mount these filesystems.
> >
> > # /dev/sda1 has type 'ext2'
> > mount /dev/sda1 /sysroot
> >
> > So I tried ...
> >
> > ><rescue> mount /dev/sda1 /sysroot
> > [<time>] EXT4-fs (sda1): mounting ext2 filesystem using the ext4
> subsystem
> > [<time>] EXT4-fs (sda1): bad geometry: block count 104388
exceeds size of
> > device (103408 blocks)
> > mount: wrong fs type, bad option, bad superblock on /dev/sda1,
> >            missing codepage or helper program, or other error
> >
> >           In some cases useful info is found in syslog - try
> >           dmesg | tail
> >
> > I don't think the image is corrupt because a colleague is able to
mount
> it
> > with guestfish on a RHEL 6 box.
>
> This is pretty strange, and may, I suppose, indicate a kernel bug of
> some kind.  I imagine it's possible that (a) the ext4 driver
doesn't
> understand some very obscure ext2 feature, and (b) since hardly anyone
> uses ext2, this bug has never been discovered before.  Or it may be as
> Pino says that the filesystem is really truncated and ext4.ko is
> correctly moaning about that, but ext2.ko in RHEL 6 ignored it.
>
> Does running `e2fsck -n /dev/sda1' under virt-rescue show up any
> errors?
>
> ><rescue> e2fsck -n /dev/sda1
e2fsck 1.42.9 (28-Dec-2013)
The filesystem size (according to the superblock) is 104388 blocks
The physical size of the device is 103408 blocks
Either the superblock or the partition table is likely to be corrupt!
Abort? no

/boot contains a file system with errors, check forced.
Pass 1: Checking inodes, blocks, and sizes
Pass 2: Checking directory structure
Pass 3: Checking directory connectivity
Pass 4: Checking reference counts
Pass 5: Checking group summary information
/boot: 36/26104 files (2.8% non contiguous), 12201/104388 blocks

> Anything interesting in the output of `tune2fs -l /dev/sda1'?
>
I'm not retyping all of the output ... interesting?
Filesystem state: not clean
Block count: 104388
>
> Rich.
>
> --
> Richard Jones, Virtualization Group, Red Hat
> http://people.redhat.com/~rjones
> Read my programming and virtualization blog: http://rwmj.wordpress.com
> virt-builder quickly builds VMs from scratch
> http://libguestfs.org/virt-builder.1.html
>

On Wed, Feb 18, 2015 at 10:07:20AM -0500, Leonard Basuino
wrote:
> > ><rescue> e2fsck -n /dev/sda1
> e2fsck 1.42.9 (28-Dec-2013)
> The filesystem size (according to the superblock) is 104388 blocks
> The physical size of the device is 103408 blocks
> Either the superblock or the partition table is likely to be corrupt!
> Abort? no
> 
> /boot contains a file system with errors, check forced.
> Pass 1: Checking inodes, blocks, and sizes
> Pass 2: Checking directory structure
> Pass 3: Checking directory connectivity
> Pass 4: Checking reference counts
> Pass 5: Checking group summary information
> /boot: 36/26104 files (2.8% non contiguous), 12201/104388 blocks
I would say this is (fairly) conclusive proof that the filesystem is
really truncated, and RHEL 6 is just ignoring that hard fact.  Note
that the code paths used by ext4 kernel and e2fsck (ie. e2fsprogs) are
quite different, and both think the filesystem is longer than the
containing device.

Sorry :-(

Rich.

-- 
Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones
Read my programming and virtualization blog: http://rwmj.wordpress.com
virt-p2v converts physical machines to virtual machines.  Boot with a
live CD or over the network (PXE) and turn machines into KVM guests.
http://libguestfs.org/virt-v2v

On Wed, Feb 18, 2015 at 10:18 AM, Richard W.M. Jones <rjones@redhat.com>
wrote:
> On Wed, Feb 18, 2015 at 10:07:20AM -0500, Leonard Basuino wrote:
> > > ><rescue> e2fsck -n /dev/sda1
> > e2fsck 1.42.9 (28-Dec-2013)
> > The filesystem size (according to the superblock) is 104388 blocks
> > The physical size of the device is 103408 blocks
> > Either the superblock or the partition table is likely to be corrupt!
> > Abort? no
> >
> > /boot contains a file system with errors, check forced.
> > Pass 1: Checking inodes, blocks, and sizes
> > Pass 2: Checking directory structure
> > Pass 3: Checking directory connectivity
> > Pass 4: Checking reference counts
> > Pass 5: Checking group summary information
> > /boot: 36/26104 files (2.8% non contiguous), 12201/104388 blocks
>
> I would say this is (fairly) conclusive proof that the filesystem is
> really truncated, and RHEL 6 is just ignoring that hard fact.  Note
> that the code paths used by ext4 kernel and e2fsck (ie. e2fsprogs) are
> quite different, and both think the filesystem is longer than the
> containing device.
>
> Sorry :-(
>
> No need to be sorry ... you have put me on a promising path ...
I was able to mount it after resizing the filesystem ...
within virt-rescue ...
><rescue> resize2fs -f /dev/sda1
then I was able to mount it with ...
guestmount -a <image> -m /dev/sda1 <mount point>
Not ideal for my needs, but work in progress.

> Rich.
>
> --
> Richard Jones, Virtualization Group, Red Hat
> http://people.redhat.com/~rjones
> Read my programming and virtualization blog: http://rwmj.wordpress.com
> virt-p2v converts physical machines to virtual machines.  Boot with a
> live CD or over the network (PXE) and turn machines into KVM guests.
> http://libguestfs.org/virt-v2v
>