i remembered something I saw on SGVLUG a few days ago...

http://blogs.zdnet.com/storage/?p=162

from that article....

"SATA drives are commonly specified with an unrecoverable read error
rate (URE) of 10^14. Which means that once every 100,000,000,000,000
bits, the disk will very politely tell you that, so sorry, but I
really, truly can't read that sector back to you."

"With a 7 drive RAID 5 disk failure, you'll have 6 remaining ### TB
drives. As the RAID controller is busily reading through those
remaining disks to reconstruct the data from the failed drive, it is
almost certain it will see an URE."

(Which is now what I think happened...)

"So the read fails. And when that happens, you are one unhappy camper.
The message "we can't read this RAID volume" travels up the chain of
command until an error message is presented on the screen."

Scheiße!

-Chris




On Wed, Jan 14, 2009 at 5:17 PM, Chris Louden <[email protected]> wrote:
> On Wed, Jan 14, 2009 at 4:59 PM, Peter Manis <[email protected]> wrote:
>> When I built my file server the raid card kept swapping with the boot
>> drive.  Despite mounting with UUID and spending a lot of time on it I never
>> got it fixed until I moved it to CentOS.  If something like that was
>> happening it would explain the movement in slots.  I would check the serial
>> numbers a couple times after rebooting to see if this is happening.  You may
>
> I think you are referring to SCSI order. This is eSATA in a DAS. Not
> sure it works the same way.
>
>> need to erase the drives to clean all possible information about the array.
>
> I need to make every effort to save the data. This was the backup
> location for production data.
>
>> I had to when I created a test array once, erasing the MBR wasn't enough for
>> some reason.
>>
>> On Wed, Jan 14, 2009 at 7:51 PM, Chris Louden <[email protected]> wrote:
>>>
>>> On Wed, Jan 14, 2009 at 4:46 PM, Peter Manis <[email protected]> wrote:
>>> > Have you moved any drives around?  What distro is this?
>>> >
>>>
>>> No movement, possible drive failure. SLES10
>>> _______________________________________________
>>> LinuxUsers mailing list
>>> [email protected]
>>> http://socallinux.org/cgi-bin/mailman/listinfo/linuxusers
>>
>>
>>
>> --
>> Peter Manis
>> (678) 269-7979
>>
>> _______________________________________________
>> LinuxUsers mailing list
>> [email protected]
>> http://socallinux.org/cgi-bin/mailman/listinfo/linuxusers
>>
>>
>

Reply via email to