ts-7000
[Top] [All Lists]

Re: [ts-7000] SD card corruption, please help!

To:
Subject: Re: [ts-7000] SD card corruption, please help!
From: Heinrich du Toit <>
Date: Mon, 10 Mar 2008 08:20:34 +0200
I've your running ext3 (which is a journaling filesystem) you should not 
get file corruption that is not easily fixable. That is the whole point 
of journaling :)

If you don't really need that data past a crash/restart then you might 
consider running an tmpfs in RAM.
Or you could alternatively run you live-backups on an nfs filesystem 
mounted on a PC/server somewhere - if you have network.

The TS-7800 board has SATA points on so in theory you can plug in a 
harddrive..
Else a few of the board have USB points so you could consider a usb 
harddrive.

I guess it all depends on what you have available...
good luck.

I agree with Richard - using a flash filesystem can possibly extend the 
lifetime.

In case this is not really a flash write cycle problem:
(If you write the same block 1's a second then you should be able to 
last at least 24 hours before the flash cracks and I figure then it 
still should only make one bad block.)
All linux filesystems can mark bad blocks (although I'm not sure if this 
is auto) and I know that yaffs running on the onboard flash can track 
bad blocks.
So there could be something else.
On the other hand if the bad block is somewhere in the filesystem 
structure code then it could in theory destroy large parts of the 
filesystem.

Firstly maybe there is something wrong with your code? (Although I guess 
that is unlikely - file stuff is pretty easy)

There is a remote possibility that the linux drivers for the SD card has 
some bug.
(There was one on the NAND drivers in the TS-7800 :/ )


tedapt wrote:
>
> I'm running Debian on an TS-7260, booting to an SD card via an initrd
> image stored in the on-board flash. The SD cards are becoming
> corrupted (binaries won't execute, files becoming garbled, attempts to
> login result in "Login: exec format error", etc.) Does anyone else
> have these issues? I've converted to an ext3 (ordered) kernel and
> filesystem from ext2 in an attempt to correct the problem, and while
> it has helped to reduce the frequency of corruption it still occurs.
>
> My application writes a small amount of data continually to a file
> on the SD card (~4k every few seconds). I can understand that a sudden
> power loss could contribute to corruption of that file, but how could
> it corrupt the entire file system in such a dramatic way?
>
> Can anyone recommend a solution or other steps I can take to reduce
> the potential for damage? In absence of a solution, are there
> any strategies for enabling auto-recovery of a corrupt SD card (i.e.,
> anything I can do in the initrd image to detect and fix/replace the
> image?) or perhaps create another read-only partition for most of the
> Debian distribution? I'm not confident the latter solution would help
> given the catastrophic nature of the failures I'm seeing, I could
> imagine that even a read-only partition could become corrupt...
>
>  



 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/ts-7000/

<*> Your email settings:
    Individual Email | Traditional

<*> To change settings online go to:
    http://groups.yahoo.com/group/ts-7000/join
    (Yahoo! ID required)

<*> To change settings via email:
     
    

<*> To unsubscribe from this group, send an email to:
    

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/
 

<Prev in Thread] Current Thread [Next in Thread>
Admin

Disclaimer: Neither Andrew Taylor nor the University of NSW School of Computer and Engineering take any responsibility for the contents of this archive. It is purely a compilation of material sent by many people to the birding-aus mailing list. It has not been checked for accuracy nor its content verified in any way. If you wish to get material removed from the archive or have other queries about the archive e-mail Andrew Taylor at this address: andrewt@cse.unsw.EDU.AU