Comment 105 for bug 599255

Hi Phillip and Henry,

I have taken a quick look at this and compared the latest metadata with Nishes from before and it looks like the offset for the high bits might actually be at 0x2E8 (in filler 2).

Basically we have 3 metadata sets in this bug report.

Nishes exist in metadata.tar.gz, the first set (2TB set) is sda1.dat, second (2.5TB set) is sda2.dat, and the latest from Henry.

Comparing all three:
The high bits for the 2TB array will be 0x0000
The high bits for the 2.5TB array should be 0x0001
The high bits for the 8 TB array should be 0x0003

Compare each metadata and 0x2E8 is correct in each instance.

Obviously this is within filler2 but occurs in the section that appears to be all 0's.

Are there any other bugs with metadata that we can compare this value from?
I remember the original patch broke a lot of arrays, were there dumps in those bug reports?

Best regards,
Danny