mount ubi volume fails: No such device

[Lars Michael <lars.michael@yahoo.com>]

> > On Mon, 2011-02-07 at 07:05 -0800, Lars Michael
> wrote:
> > > UBI: scrubbed PEB 863, data moved to PEB 1097 UBI
> warning:
> > ubi_eba_copy_leb: read data back from PEB 1092 and it
> is different UBI
> > error: wear_leveling_worker: error -22 while moving
> PEB 868 to PEB
> 1092
> > UBI warning: ubi_ro_mode: switch to read-only mode UBI
> error: do_work:
> > work failed with error code -22 UBI error: ubi_thread:
> ubi_bgt0d: work
> > failed with error code -22
> > 
> > It says you that it wrote some date, then read it
> back, compared, and
> > the data did not match. You need to start with
> validating your falash
> > drivers - use mtd tests:
> > 
> > http://www.linux-mtd.infradead.org/doc/general.html#L_mtd_tests
> > 
> 
> I formatted and wrote the image again. ubiattach
> _sometimes_ completes
> without the error code -22. In this case the mount went ok.
> But mount
> sometimes fails completely (structure needs cleaning),
> other times
> errors are reported but successfully recovered. So it seems
> to work, but
> no in
> a reliable way. Any suggestions on how to improve it? My
> kernel is
> 2.6.29 so perhaps some patches are needed?
> 

So I got 167 patches from the ubifs 2.6.29 back port tree, some of them
looking very relevant. I did manage to format, attach and mount once, but it appears very unstable reporting more and more errors, like:

UBI warning: process_eb: valid VID header but corrupted EC header at PEB 959

UBI error: check_corruption: PEB 1065 contains corrupted VID header, and the data does not contain all 0xFF, this may be a non-UBI PEB or a severe VID header corruption which requires manual inspection

I will make a new post with more details, using another mail client that dont screw the header info.