How to fix bad superblock or xfs_repair: error - read only 0 of 512 bytes

[Eric Sandeen]

On 1/24/12 9:46 AM, Christian Kildau wrote:
> 
> On Jan 24, 2012, at 3:12 PM, Roger Willcocks wrote:
> 
>>
>> On Tue, 2012-01-24 at 11:13 +0100, Christian Kildau wrote:
>>> Top posting... sorry.
>>>
>>> I have now found dozens of other users with a similar issue! e.g.
>>> http://www.linuxquestions.org/questions/linux-general-1/cannot-mount-hard-disk-block-count-exceeds-size-of-device-bad-partition-table-880149/
>>>
>>> To make it short all of these users were running ext4 and a fs resize to the new geometry fixed their problems! Sadly XFS doesn't support shrinking the fs(?).
>>>
>>
>> It seems fairly clear that your drive or the bios is lying about its
>> capacity. The filesystem occupies the entire disk, but the disk has
>> become 'smaller'. A quick web search suggests a 'hidden protected area'
>> - the two block counts in this link line up with the before and after
>> sizes you're seeing:
>>
>> http://lime-technology.com/forum/index.php?topic=13440.0;wap2
>>
>> It would be instructive to see what 'hdparm -N /dev/sdd' says on your
>> system. And a dmesg log would be handy too.
>>
>> Note that this is /not/ a problem with xfs. The right fix is to tell the
>> drive to report its actual capacity, not to shrink the filesystem.
> 
> I do understand that is definitely not an XFS issue, but some strange issue with ubuntu or their kernel patches...
> 
> I got my data back by dumping the entire hdd (it was partitionless nevertheless) to a bigger 2TB hdd.
> XFS mounts without any problems and I can restore my data.
> 
> Thanks all for your help!

It's likely still missing the end of the filesystem, though.

Can you run the hdparm command Roger suggested on your original hard drive, please?