checksum error

[Phillip Susi]

I took a look at this, and I can't see how this error would ever NOT 
happen ( which is really strange since I used to have a via chipset that 
worked fine ).  via.c computes a sum of the first 50 bytes ( comment 
says this, but it actually starts at offset 50 and goes back to zero, so 
it's really 51 bytes ), then compares it with the sum stored in the 
metadata, which is at byte 50.  There is no value you could possibly 
have for the checksum byte that when added to the sum of the preceding 
bytes, would give you that same checksum, unless the sum of the first 50 
bytes happened to be zero, any value at all would be valid for the 
checksum byte.

On 12/6/2011 12:29 PM, Heinz Mauelshagen wrote:
> On 12/05/2011 02:38 PM, Bad Bod wrote:
>> Hi,
>> I don't know if this is the correct place to send this, please send me
>> information on where I should post of this is not the right place.
>>
>> dmraid version
>> 1.0.0.rc16 (2009.09.16) shared
>>
>> RAID 0, Asus M2V motherboard, via 8237, 2 * hitachi 250Gb HD
>>
>> ubuntu 11.04 and fedora 15
>>
>> running 'dmraid - ay' gives the following output.
>>
>> ERROR: via: invalid checksum on /dev/sdb
>> ERROR: via: invalid checksum on /dev/sda
>> no raid disks
>>
>> windows via v-raid tool shows no errors, array boots works fine in
>> windows 7. Have rebuilt array several times to ensure metadata should
>> not be corrupt.
>>
>> attached is dd of /dev/sda with metadata. Sorry, dumped 3000 sectors
>> so a bit more than you need ;)
>>
>> Any assistance greatly appreciated, even if this is just my setup that
>> is weird, I am happy to patch around it to get it working. I looked at
>> the via.h and could not work out where to start :(
>>
>> As I said at start, if this is not the correct place then please
>> inform me where is, and please accept my apologies.
>
> That's fine.
> I'll try to have a look...
>
> Heinz