JFFS2 messages

[Eric]

Hi, all:

We are developing AT91RM9200-DK based device, 32M SDRAM, 8M Nor flash for linux kernel image and INITRD, Nand flash for applications and storage. Our application will frequently write something to the Nand. 

Now we are performing critical test. The device may lose power at any time and it will reboot after 10sec. Two days has gone.
While rebooting and mounting Nand flash, Jffs2 gives many messages, like:
---------------------------------------------------------------------
Empty flash at 0x000cc738 ends at 0x000cc800
....

jffs2_scan_eraseblock(): Node at 0x00ca0ffc {0x1985, 0xe002, 0xe0021985) has invalid CRC 0x00000075 (calculated 0xd7cd6a7b)
.....

JFFS2 notice: (179) check_node_data: wrong data CRC in data node at 0x01b25a00: read 0x58d6423f, calculated 0xb3eebebd.
JFFS2 notice: (179) read_dnode: node CRC failed on dnode at 0x1057bdc: read 0xffffffff, calculated 0x32a8b981
.......
---------------------------------------------------------------------

Now these message becomes more and more.

Another frustrating thing is that we find one log file (produced by our application) is corrupted. Its size is over 800M and cannot cat it. However, can rm it.

I am wondering whether the device can survive the test, and JFFS2 is as robust as the webpage (http://www.denx.de/wiki/DULG/RootFileSystemOnAJFFS2FileSystem) says: JFFS2 (Journalling Flash File System version 2) was specifically designed for use on flash memory devices in embedded systems. It is a log-structured file system which means that it is robust against loss of power, crashes or other unorderly shutdowns of the system ("robust" means that data that is just being written when the system goes down may be lost, but the file system itself does not get corrupted and the system can be rebootet without need for any kind of file system check).

Thanks and Regards
Eric