Steve Holdoway wrote:
Volker Kuhlmann wrote:
...
kernel: ide: failed opcode was: unknown
kernel: hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error }
kernel: hdb: dma_intr: error=0x84 { DriveStatusError BadCRC }
kernel: ide: failed opcode was: unknown
kernel: hdb: dma_intr: status=0x51 { DriveReady SeekComplete Error }
kernel: hdb: dma_intr: error=0x84 { DriveStatusError BadCRC }
kernel: ide: failed opcode was: unknown
kernel: hda: DMA disabled
kernel: ide0: reset: success

Reboot. Two concurrent dd thrashing the disk

dd bs=1k if=/dev/hdb1 of=/dev/null skip=60000000 &
dd bs=1k if=/dev/zero of=/dev/hdb1 count=1000000 &

fail to cause any trouble. What can I try next?

Any ideas much appreciated,

Volker,

Your disk's b*ggered. Switch it off for 24 hours, then get as much off as you can, print off the error messages, and then take it back under warranty!

I had the same thing happen to me, and did a load of research. What it boils down to is that is you *ever* see a disk error that hasn't been caught by the dozens of layers of error correction that it goes through, then it's dead.

I'd agree. It looks like possibly on its way out. It could also be an issue with your motherboards IDE controller and the kernel.


Try running smart tests on the disk. For the smartmontools package (don't know if there is a suse package but there are rpms, or get the source from http://smartmontools.sourceforge.net if necessary), then use:

smartctl --smart=on --offlineauto=on --saveauto=on /dev/hdb
smartctl -t short /dev/hdb
then wait the amount of time it said the test will take
smartctl -a /dev/hdb

Repeat for hda.  Use -t long if -t short doesn't report any issues.

the older smartctl has a slightly different syntax, but is much the same.

It will make it easier to give back to the store and get them to replace the disk if it under warranty rather than having them just do a low level format or something.

Regards

Daniel

Reply via email to