hda problems

Bryan C. Andregg (bandregg@idir.net)
Wed, 04 Sep 1996 00:24:30 -0500

'lo All,
While doing some heavy database access (Red Hat's rpm with the -qa
option) my machine loops up accessing the harddrive and I have to suspend
and interrupt it to kill. The errors given are:

Sep 4 00:14:09 hueco kernel: hda: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
Sep 4 00:14:09 hueco kernel: hda: read_intr: error=0x40 { UncorrectableError }, LBAsect=920040, sector=7732
Sep 4 00:14:09 hueco kernel: end_request: I/O error, dev 03:07, sector 7732
Sep 4 00:14:18 hueco kernel: hda: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
Sep 4 00:14:18 hueco kernel: hda: read_intr: error=0x40 { UncorrectableError }, LBAsect=920040, sector=7734
Sep 4 00:14:18 hueco kernel: end_request: I/O error, dev 03:07, sector 7734
Sep 4 00:14:28 hueco kernel: hda: read_intr: status=0x59 { DriveReady SeekComplete DataRequest Error }
Sep 4 00:14:28 hueco kernel: hda: read_intr: error=0x40 { UncorrectableError }, LBAsect=920040, sector=7736
Sep 4 00:14:28 hueco kernel: end_request: I/O error, dev 03:07, sector 7736

I have run e2fsck on all partitions and found no errors.
The changes I have made are to /usr/src/linux/drivers/block/ide.h to define
REALLY_FAST_IO and to change INITAIL_MULT_COUNT to 16. These have worked
fine in >= 2.0.11 kernels (the last I ran) and I wonder why they would
crash now?

The output of hdparm -i /dev/hda is

/dev/hda:

Model=IBM-DBOA-2528, FwRev=C1B80H38, SerialNo=1E12E024
Config={ HardSect NotMFM HdSw>15uSec Fixed DTR>10Mbs }
RawCHS=1024/16/63, TrkSize=0, SectSize=0, ECCbytes=4
BuffType=3(DualPortCache), BuffSize=64kB, MaxMultSect=32, MultSect=16
DblWordIO=no, maxPIO=2(fast), DMA=yes, maxDMA=0(slow)
CurCHS=1024/16/63, CurSects=1032192, LBA=yes, LBAsects=1032192
tDMA={min:180,rec:180}, DMA modes: sword0 sword1 sword2 mword0 mword1
IORDY=on/off, tPIO={min:240,w/IORDY:180}, PIO modes: mode3

Which I believe will support this INITIAL_MULTI but there isn't much info.
Can anyone explain what would cause these I/O errors outside of physical
damage.

-------------------------------------
Bryan C. Andregg [] Network Administrator
Internet Direct Communications [] System Administrator
Lawrence, Kansas [] Internet Janitor
-------------------------------------