I tried to see how my raid configuration stood up to a bit of stress so I
created a tarball of the filesystem on a mounted partition. Out of four such
attempts, two caused raid failures. In both failures, one of the partitions was
removed from the array and operation continued in degraded mode. After the first
failure, I checked the partition that had been removed from the array and found
it to be without disk errors and re-created the array.
The second failure was on a different partition. Again, subsequent checks
indicate that the partition and disk are OK. In the second failure I saved the
error messages and they are attached. In all four cases the tarball was
created successfully.
My question is: Do I have a bad disk or is the raid software failing under
stress? How can I tell?
alex
[root@newlix /]# tar --exclude proc --exclude mnt/hdc -czf
/mnt/hdc/backup.tar.gz .
hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc: dma_intr: error=0x40 { UncorrectableError }, LBAsect=55630,
sector=13214
end_request: I/O error, dev 16:05 (ide), sector 13214
raid1: Disk failure o hdc5, disabling device.
Operation continuing on 1 devices
kmalloc called nonatomically from interrupt 0000002c
raid1: md3: rescheduling block 6607
(skipping faulty hdc5)
hda9 [events: 00000008](write) hda9s sb offset: 307328
md3: no spare disk to reconstruct array! -- continuing in degraded
mode.
raid1: md3: redirecting sector 6607 to another mirror
hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc: dma_intr: error=0x40 { UncorrectableError }, LBAsect=55630,
sector=13216
end_request: I/O error, dev 16:05 (ide), sector 13216
raid1: md3: rescheduling block 6608
md3: no spare disk to reconstruct array! -- continuing in degraded
mode.
raid1: md3: redirecting sector 6607 to another mirror
hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc: dma_intr: error= 0x01 { AddrMarkNotFound }, LBAsect55630,
sector=13218
hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc: dma_intr: error=0x40 { UncorrectableError }, LBAsect=55630,
sector=13218
end_request: I/O error, dev 16:05 (ide), sector 13218
raid1: md3: rescheduling block 6609
md3: no spare disk to reconstruct array! -- continuing in degraded
mode.
raid1: md3: redirecting sector 6609 to another mirror
hdc: dma_intr: status=0x51 { DriveReady SeekComplete Error }
hdc: dma_intr: error= 0x01 { AddrMarkNotFound }, LBAsect55630,
sector=13220
[root@newlix /]#
