Re: Journaling pointless with today's hard disks?

[Wayne Whitney]

In mailing-lists.linux-kernel, Andre Hedrick wrote:

> By the time an ATA device gets to generating this message, either the bad
> block list is full or all reallocation sectors are used.  Unlike SCSI
> which has to be hand held, 90% of all errors are handle by the device.

Perhaps you or one of the other gurus could explain the following
observations, which I am sure that I and many other readers would find
very enlightening:

I have an IBM-DTLA-307045 drive connected to a PDC20265 controller on
an ia32 machine running 2.4.16.  After reading this discussion and
hearing about the problems that others have had with the IBM 75GXP
series, I thought that I should test out my drive to see if it is OK.
So I ran 'dd if=/dev/hde of=/dev/null bs=128k'.  Every thing went
fine, except for about five seconds in the middle, when the disk made
a lot of grinding sounds and the system was unresponsive.  That
generated the log messages messages appended below.

However, running the dd command again (after a reboot) produced no
errors.  So the drive remapped some bad sectors the first time
through?  The common wisdom here is that once you get to the point
where the drive reports a bad sector, you are in trouble.  If so, why
did the second dd command work OK?  I have had no other problems with
this drive.

Thanks, Wayne

hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x40 { UncorrectableError }, LBAsect=12939888, sector=12939804
end_request: I/O error, dev 21:00 (hde), sector 12939804
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x40 { UncorrectableError }, LBAsect=12939888, sector=12939806
end_request: I/O error, dev 21:00 (hde), sector 12939806
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x40 { UncorrectableError }, LBAsect=12939888, sector=12939808
end_request: I/O error, dev 21:00 (hde), sector 12939808
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x40 { UncorrectableError }, LBAsect=12939888, sector=12939810
end_request: I/O error, dev 21:00 (hde), sector 12939810
hde: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hde: dma_intr: error=0x40 { UncorrectableError }, LBAsect=12939888, sector=12939812
end_request: I/O error, dev 21:00 (hde), sector 12939812