RE: Problems with jffs corruption

Hi Tim

This is a known problem. It happens when the JFFS2 image is not complete.
With the correct options mkfs.jffs2 creates a image that contains all the
data in all blocks, even the "empty" ones. Then the kernel will not complain.
Else you need to erase the FLASH before burning a new one.

Walter

> -----Original Message-----
> From: [email protected]
> [mailto:[email protected]]On Behalf Of Tim
> Wentford
> Sent: Monday, August 23, 2004 4:24 PM
> To: [email protected]
> Subject: [Simpad-general] Problems with jffs corruption
> 
> 
> I've been posting to the forums but haven't got much response yet so I
> thought that maybe the mailing list is more active.
> 
> Anyway, the short version of what is happening to me is that 
> I get jffs
> corruption similar to that which you get when you don't first 
> erase the
> flash or have a corrupt image or even damaged serial port. 
> However, I have
> eliminated all these as potential sources of my problems by 
> various cross
> checks not least of which is observing similar behaviour on a 
> colleagues
> simpad from the same batch when he also tried installing opie.
> 
> My best success so far has been by installing the tinyx 
> cramfs image and
> then running mount_jffs2.sh mount and then reflashing opie 
> without erasing
> the flash. This almost works with two relatively minor problems:
> 
> 1) If I suspend I can't open any apps after resuming (I have 
> to shutdown and
> then reboot)
> 
> 2) If I use a kernel with a ram-disk the ram-disk disappears 
> after the first
> reboot.
> 
> I still get a couple of jffs2 error messages but nothing like 
> the flood I
> used to get when relying on erasing the flash. One pops up 
> when starting up:
> 
> jffs2_read_inode(): No data nodes found for ino #3722
> 
> One when terminating (couldn't get the whole message for this 
> because it
> disappears too quickly)
> 
> totlen at node# NNNN != totlen at NNNN
> 
> I think what may be happening is that the erase procedure 
> isn't working
> (perhaps there are some lock-bits set in the flash which 
> aren't unset by
> erasing but are by serloading a new image?).
> 
> So, a few questions:
> 
> 1) Does anyone know how to fix this?
> 
> 2) Is it worth doing the mount_jffs2 thing but then also 
> erasing the flash
> before installing opie?
> 
> 3) Is there likely to be a padded SL4 image soon (which, if I 
> am right, may
> fix my problems)?
> 
> 4) Is there anything else I can do to diagnose the problem?
> 
> Cheers
> Tim
> 
> 
> -------------------------------------------------------
> SF.Net email is sponsored by Shop4tech.com-Lowest price on Blank Media
> 100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for only $33
> Save 50% off Retail on Ink & Toner - Free Shipping and Free Gift.
> http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285
> _______________________________________________
> Simpad-general mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/simpad-general
> 


-------------------------------------------------------
SF.Net email is sponsored by Shop4tech.com-Lowest price on Blank Media
100pk Sonic DVD-R 4x for only $29 -100pk Sonic DVD+R for only $33
Save 50% off Retail on Ink & Toner - Free Shipping and Free Gift.
http://www.shop4tech.com/z/Inkjet_Cartridges/9_108_r285