ide fails

[Ralf Saalmüller <ralf@mail.ralf-saalmueller.de>]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Hello to the debian people,

I'm running a PowerMac G3 300 MHz:

cat /proc/cpuinfo
processor       : 0
cpu             : 740/750
temperature     : 32-35 C (uncalibrated)
clock           : 300.690000MHz
revision        : 2.2 (pvr 0008 0202)
bogomips        : 33.28
timebase        : 16708566
platform        : PowerMac
machine         : Power Macintosh
motherboard     : AAPL,Gossamer MacRISC
detected as     : 48 (PowerMac G3 (Gossamer))
pmac flags      : 00000000
L2 cache        : 1024K unified pipelined-syncro-burst
pmac-generation : OldWorld

with two hard drives:

hda: IC35L060AVVA07-0, ATA DISK drive
hda: Enabling MultiWord DMA 2
hdc: SAMSUNG HD300LD, ATA DISK drive
hdc: Enabling MultiWord DMA 2
hda: max request size: 128KiB
hda: 120103200 sectors (61492 MB) w/1863KiB Cache, CHS=65535/16/63,  
(U)DMA
hda: cache flushes supported
hda: [mac] hda1 hda2 hda3 hda4
hdc: max request size: 512KiB
hdc: 586072368 sectors (300069 MB) w/8192KiB Cache, CHS=36481/255/63,  
(U)DMA
hdc: cache flushes supported
hdc: [mac] hdc1 hdc2 hdc3 hdc4 hdc5

For the record:
Etch installation - Linux platon 2.6.18-5-powerpc #1 Wed Oct 3  
04:56:19 UTC 2007 ppc GNU/Linux

hdc is less then a year old, had an hitachi (and replacement from  
hitachi) that died in that place twice. Both drives have a  
temperature about 38-42°C over the year. Yesterday I found that a  
package was brocken an elf binary wasn't in the right place. As a new  
kernel was on the update list, I rebooted the computer, fsck kicked  
in and was done with -fc options ... which found some bad blocks on  
hda. Fixed. Packages fixed, too. Those I noticed.

And a day later I found on the console:

Oct 21 07:20:55 platon kernel: hda: dma_intr: status=0xd0 { Busy }
Oct 21 07:20:55 platon kernel: ide: failed opcode was: unknown
Oct 21 07:20:55 platon kernel: hda: DMA disabled
Oct 21 07:20:55 platon kernel: ide0: reset: success
...
Oct 21 11:20:52 platon kernel: hdc: dma_intr: status=0x58  
{ DriveReady SeekComplete DataRequest }
Oct 21 11:20:52 platon kernel: ide: failed opcode was: unknown
Oct 21 11:20:52 platon kernel: hdc: dma_intr: status=0x58  
{ DriveReady SeekComplete DataRequest }
Oct 21 11:20:52 platon kernel: ide: failed opcode was: unknown
Oct 21 11:20:52 platon kernel: hdc: dma_intr: status=0x58  
{ DriveReady SeekComplete DataRequest }
Oct 21 11:20:52 platon kernel: ide: failed opcode was: unknown
Oct 21 11:20:52 platon kernel: hdc: dma_intr: status=0xd0 { Busy }
Oct 21 11:20:52 platon kernel: ide: failed opcode was: unknown
Oct 21 11:20:52 platon kernel: hdc: DMA disabled
Oct 21 11:20:52 platon kernel: ide1: reset: success

I know that drives break. It's possible that two drives break at the  
same time. But 4 drives in all and two at the same time? Is there a  
problem with the PMac G3 ide? Do I need a custom kernel? Sarge was up  
over a year without a problem.

I'm a little afraid that I will loose the discs, the data (rdiff- 
backup done), or the server again.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.7 (Darwin)

iD8DBQFHG5sWtKcsE7vsbMERAtVBAJ9maykTAKboWuruYaFcBOQDvLQXBwCeKohZ
NiQUDodiNggMaqWSDSaM69E=
=XAK1
-----END PGP SIGNATURE-----