Data corrupted

[Artem B. Bityuckiy]

Hello

1. Messages like "Node header CRC failed at 001f4ec8." are normal after 
unclean reboots. 

But the fact that you lost data which weren't engaged is *not normal*. 
Theoretically, if JFFS2 has no bugs, static data can't be lost.

I suspect JFFS2 may have bug: after block is GCed, it is erased, and wbuf 
isn't sync-ed before erase. I'll see the relative places in JFFS2 and 
check if it possible. If it is, it definitely must be fixed.

Bobzin,Heiko wrote:
> Hello mtd-fellows,
> 
> I ran into trouble with a test I've implemented. It is basically the 
same
> as the mtd-util checkfs: writing into the flash while powering down
> the device. The problem is, that files which are never touched are
> corrupted, not the data I write.
> I diff'd the data and exactly one 4K block is set to zero in a 500k file
> in a ~7MB JFFS2 flash partition.
> It is reproducible in some hours, but the location differs.

--
Best Regards,
Artem B. Bityuckiy,
St.-Petersburg, Russia.