On 09/05/2006 02:43 AM, thomass@remove.bigpond.com wrote:
> Hello All,
>
> Whilst trying to back up my drive i corrupted the dba2 partition and
> Fedora Core 3 will now not boot. So i have a new install on another HD
> disk and am trying to mount the original disk so that i can retrieve
> some data from it. I can see these files under /mnt/sysimage when i
> boot with the old disk as primary and "linux rescue". It does warn me
> that not all files are avalible as some are lost but I can see and
> read the ones which I need. When i make that disk a slave and try to
> mount drive it fails:
>
> mount -t ext3 /dev/dbb2 /mnt/oldrive
>

Since you created an ext2 partition, try to mount it as ext2:

mount -t ext2 /dev/dbb2 /mnt/olddrive


> mount: wrong fs type, bad option, bad superblock on /dev/dbb2 or too
> many mounted file systems
>
> I have tried
>
> e2fsck -f /dev/dbb2
> e2fsck -b XXXXX /dev/dbb2
>
> How do i mount it (exactly like linux resue boot) so as to retrieve
> the still avalible data from the partition. I can then reformat the
> drive and get on with my life.
>
> thanks for any help.

You can also let mount auto-detect the filesystem type:

mount /dev/ddb2 /mnt/olddrive

On Tue, 05 Sep 2006 11:22:56 GMT, "Mumia W."
wrote:

>On 09/05/2006 02:43 AM, thomass@remove.bigpond.com wrote:
>> Hello All,
>>
>> Whilst trying to back up my drive i corrupted the dba2 partition and
>> Fedora Core 3 will now not boot. So i have a new install on another HD
>> disk and am trying to mount the original disk so that i can retrieve
>> some data from it. I can see these files under /mnt/sysimage when i
>> boot with the old disk as primary and "linux rescue". It does warn me
>> that not all files are avalible as some are lost but I can see and
>> read the ones which I need. When i make that disk a slave and try to
>> mount drive it fails:
>>
>> mount -t ext3 /dev/dbb2 /mnt/oldrive
>>
>
>Since you created an ext2 partition, try to mount it as ext2:
>
>mount -t ext2 /dev/dbb2 /mnt/olddrive
>
>
>> mount: wrong fs type, bad option, bad superblock on /dev/dbb2 or too
>> many mounted file systems
>>
>> I have tried
>>
>> e2fsck -f /dev/dbb2
>> e2fsck -b XXXXX /dev/dbb2
>>
>> How do i mount it (exactly like linux resue boot) so as to retrieve
>> the still avalible data from the partition. I can then reformat the
>> drive and get on with my life.
>>
>> thanks for any help.
>
>You can also let mount auto-detect the filesystem type:
>
>mount /dev/ddb2 /mnt/olddrive


I have tried

mount -t ext3 /dev/dbb2 /mnt/oldrive
mount -t ext2 /dev/dbb2 /mnt/oldrive
mount /dev/dbb2 /mnt/oldrive
mount -t ext3 /dev/dbb2 /mnt/oldrive -o ro
mount -t ext2 /dev/dbb2 /mnt/oldrive -o ro
mount -t ext3 -o ro.errors=recover,errors=continue /dev/dbb2
/mnt/oldrive

always the above error msg

if i run "dmesg | more" after

"mount -t ext3 /dev/dbb2 /mnt/oldrive"

I get:

EXT3-fs: (DEVICE hdb2) ext3_check_descripters: Block bitmap for group
0 not in group (block 1853189987)
EXT3-fs: Group descriptors corupted !

nothing works. If I can replicate the mount on the linux rescue boot I
can get to my data. I do not care about recovering the disk itself.

thanks for any help

On 09/05/2006 08:03 AM, thomass@remove.bigpond.com wrote:
> On Tue, 05 Sep 2006 11:22:56 GMT, "Mumia W."
> wrote:
>
>> On 09/05/2006 02:43 AM, thomass@remove.bigpond.com wrote:
>>> Hello All,
>>>
>>> Whilst trying to back up my drive i corrupted the dba2 partition and
>>> Fedora Core 3 will now not boot. So i have a new install on another HD
>>> disk and am trying to mount the original disk so that i can retrieve
>>> some data from it. I can see these files under /mnt/sysimage when i
>>> boot with the old disk as primary and "linux rescue". It does warn me
>>> that not all files are avalible as some are lost but I can see and
>>> read the ones which I need. When i make that disk a slave and try to
>>> mount drive it fails:
>>>
>>> mount -t ext3 /dev/dbb2 /mnt/oldrive
>>>
>> Since you created an ext2 partition, try to mount it as ext2:
>>
>> mount -t ext2 /dev/dbb2 /mnt/olddrive
>>
>>
>>> mount: wrong fs type, bad option, bad superblock on /dev/dbb2 or too
>>> many mounted file systems
>>>
>>> I have tried
>>>
>>> e2fsck -f /dev/dbb2
>>> e2fsck -b XXXXX /dev/dbb2
>>>
>>> How do i mount it (exactly like linux resue boot) so as to retrieve
>>> the still avalible data from the partition. I can then reformat the
>>> drive and get on with my life.
>>>
>>> thanks for any help.
>> You can also let mount auto-detect the filesystem type:
>>
>> mount /dev/ddb2 /mnt/olddrive
>
>
> I have tried
>
> mount -t ext3 /dev/dbb2 /mnt/oldrive
> mount -t ext2 /dev/dbb2 /mnt/oldrive
> mount /dev/dbb2 /mnt/oldrive
> mount -t ext3 /dev/dbb2 /mnt/oldrive -o ro
> mount -t ext2 /dev/dbb2 /mnt/oldrive -o ro
> mount -t ext3 -o ro.errors=recover,errors=continue /dev/dbb2
> /mnt/oldrive
>
> always the above error msg
>
> if i run "dmesg | more" after
>
> "mount -t ext3 /dev/dbb2 /mnt/oldrive"
>
> I get:
>
> EXT3-fs: (DEVICE hdb2) ext3_check_descripters: Block bitmap for group
> 0 not in group (block 1853189987)
> EXT3-fs: Group descriptors corupted !
>
> nothing works. If I can replicate the mount on the linux rescue boot I
> can get to my data. I do not care about recovering the disk itself.
>
> thanks for any help

Since e2fsck does not work on that disk, I advise copying the
device data using "dd" to another disk. For the time being,
your data may be lost; however, future improvements to e2fsck
may allow some data retrieval.

Also, an Internet search may reveal better tools for ext2 data
retrieval than e2fsck.

But while you are stuck with e2fsck, use dumpe2fs and get as
much information as you can (including the blocksize). When
you go back to e2fsck, you might need to specify the blocksize
since your filesystem is very badly corrupted; don't blindly
trust what dumpe2fs says; what you remember about the
blocksize is more reliable than dumpe2fs at this point.
Dumpe2fs might also tell you where the backup superblocks are.

After you've copied the partition image to another drive, you
can start grasping at straws--trying even more options of
e2fsck, such as the "-p" option (probably ineffective alone)
and combinations of options, such as "e2fsck -p -b XXXXXX -f
/dev/dbb2"

Good luck.