LinuxQuestions.org

LinuxQuestions.org (/questions/)
-   Linux - Hardware (http://www.linuxquestions.org/questions/linux-hardware-18/)
-   -   failing hard disk...i think... (http://www.linuxquestions.org/questions/linux-hardware-18/failing-hard-disk-i-think-153090/)

zepplin611 03-03-2004 02:31 PM

failing hard disk...i think...
 
my machine is running RH 9.0 - and the output of uname -r = 2.4.20-6

recently, the contents of /var/log/messages has been filling up with hard disk errors of the type:

*********************************************************************************
Mar 3 11:09:48 bono kernel: hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error }
Mar 3 11:09:48 bono kernel: hdb: dma_intr: error=0x40 { UncorrectableError }, LBAsect=332084242, high=19, low=13317138, sec
tor=328003672
Mar 3 11:09:48 bono kernel: end_request: I/O error, dev 03:41 (hdb), sector 328003672
Mar 3 11:12:30 bono kernel: hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error }
Mar 3 11:12:30 bono kernel: hdb: dma_intr: error=0x40 { UncorrectableError }, LBAsect=332087562, high=19, low=13320458, sec
tor=328007016
Mar 3 11:12:30 bono kernel: end_request: I/O error, dev 03:41 (hdb), sector 328007016
Mar 3 11:12:35 bono kernel: hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error }
Mar 3 11:12:35 bono kernel: hdb: dma_intr: error=0x40 { UncorrectableError }, LBAsect=332087567, high=19, low=13320463, sec
tor=328007024
***********************************************************************************

My first plan of action was to unmount the drive (which contains an ext3 filesystem) and run:

fsck -y /dev/hdb1

it ran and corrected some errors and ended.

I remounted the drive and have tried to pull all important data from it...in the midst of this, I was watching the logs for more errors and sure enough, the same ones as lited above re-surfaced.

I un-mounted the drive again and ran: fsck -y /dev/hdb1 and it returned that the disk was clean!?!?

I am also getting I/O errors from this disk (which, by the way is a brand new WD 200GB / 7200RPM / 8mb cache disk) which look like:

[root@bono data]# cp: reading `<FILE>': Input/output error
I think the inevitable is replacing this disk, but its brand new...how often do brand new (within 2 weeks of purchase) drives fail? Anomaly here? Can anyone offer an alternate plan of action to check whether this disk is indeed failing???

Thanks

Zepplin611

kilgoretrout 03-03-2004 03:35 PM

I guess the first thing to do is download the WD diagnostic utility(DLG Diagnostic) and run it on the drive:

http://support.wdc.com/download/

Next thing to do is see about getting a warranty replacement if the drive comes up bad. You'll need to contact their tech support to get authorization.

slakmagik 03-03-2004 09:17 PM

Yep. Try like kilgoretrout said, but I just had one bite the dust instantly, spewing Seek/Error messages like that.

Basically, disks fail in one of two ways - almost immediately or after years. It's not *common* for them to fail in a matter of days/weeks but more common than to fail after a year or so. It's not extraordinary. If you're under warranty and have your data backed up, you're in good shape. Or you can be like I was. ;)

zepplin611 03-03-2004 09:57 PM

thanks for the help guys...i will take a look at the WD diagnostic utility...but replacing the disk does look inevitable...

uggh...

zepplin611


All times are GMT -5. The time now is 07:29 AM.