LBA NAND corruption

[Mitch Bradley]

John Watlington wrote:
>
> Mitch,
>    One of the LBA NAND test machines killed it's MBR.
> It started with a failed comparison of the commonly
> written blocks, then stopped talking to the device at
> all.
>
> On reboot, fdisk showed no partition table.
> dd of /dev/lba showed all FFs for the first 16K,
> then 00 for the next 2K, then data.
>
> Suggestions on how to proceed w. debugging
> are welcome.
>
> wad
>

It would be interesting to see if we can recover the device by replacing 
the MBR with one from another of the units.

At one point I tried to setup the layout on my LBA test system so that 
the MBR is in a separate NAND erase block, the data partition starts on 
an erase block boundary, and the filesystem blocks are a multiple of and 
aligned to NAND pages.  But it's possible that I might not have 
succeeded on all points, and it's also possible that your test systems 
could have been formatted differently.

Another thing we could do would be to put the MBR, and perhaps some 
other critical index blocks, in the "SLC" segment of the device.  That 
would require some changes to the driver, but they wouldn't be difficult.