Help with UBI failure analysis

[Richard Weinberger]

On Thu, Sep 24, 2020 at 10:41 AM Sergei Poselenov
<sposelenov at emcraft.com> wrote:
> Freeing unused kernel memory: 280K (80898000 - 808de000)
> UBIFS error (ubi0:0 pid 1): ubifs_read_node: bad node type (0 but expected 9)
> UBIFS error (ubi0:0 pid 1): ubifs_read_node: bad node at LEB 12:100352, LEB mapping status 1
> Not a node, first 24 bytes:
> 00000000: 55 42 49 23 01 00 00 00 00 00 00 00 00 00 00 02 00 00 08 00 00 00 10 00                          UBI#....................

So, UBIFS reads from a node and finds a UBI EC header, this must not happen.
UBI headers are invisible to UBIFS. If UBIFS reads from a LEB and a
given offset, the offset is translated
to read data after the page(s) where EC/VID headers reside.

*unless* LEB 12 offset 100352 contains an UBIFS data node for a
nanddump you did before
and the index tree is corrupted. But I don't think so.

Looks more like a problem on the mtd side.

-- 
Thanks,
//richard