Hi,
The error: (/var/logs/messages)
ata3: DMA timeout, stat 0x1
ATA: abnormal status 0xD8 on port 0xF8B09287
scsi2: ERROR on channel 0, id 0, lun 0, CDB: 0x2a 00 00 23 10 3d 00 04 00
00
Current sdc: sense = 70 3
ASC= c ASCQ= 2
Raw sense data:0x70 0x00 0x03 0x00 0x00 0x00 0x00 0x06 0x00 0x00 0x00 0x00
0x0c 0x02
end_request: I/O error, dev sdc, sector 2297917
ATA: abnormal status 0xD8 on port 0xF8B09287
ATA: abnormal status 0xD8 on port 0xF8B09287
ATA: abnormal status 0xD8 on port 0xF8B09287
ata2: DMA timeout, stat 0x61
ATA: abnormal status 0xD8 on port 0xF8B090C7
scsi1: ERROR on channel 0, id 0, lun 0, CDB: 0x2a 00 00 25 b9 3d 00 04 00
00
Current sdb: sense = 70 3
ASC= c ASCQ= 2
Raw sense data:0x70 0x00 0x03 0x00 0x00 0x00 0x00 0x06 0x00 0x00 0x00 0x00
0x0c 0x02
end_request: I/O error, dev sdb, sector 2472253
ATA: abnormal status 0xD8 on port 0xF8B090C7
ATA: abnormal status 0xD8 on port 0xF8B090C7
ATA: abnormal status 0xD8 on port 0xF8B090C7
System:
Abit MAX3 , P4,1GB ram,
Distri: mandrake 10, Yoper
4x160GB SATA on Sil3114 as SCSI /dev/sda ..sdd
1x80GB ATA on /dev/hda
Created RAID5 and RAID0 on 3 HDs,
md0 *raid0* (sda2,sdb2,sdc3)
md1 *raid5* (sda3,sdb3,sdc3)
What happens:
The problem is reproducable:
Each time I copy more than 1GB in KDE to md0 the PC almost freezes.
Interesting is that it seems KDE System Guard shows me physical memory
running full, and after that it shows {stilled} on the copy window.
The HD leds are on (not blinking!) and its impossible to do anything with
this mount.
root@t-rex root # hdparm -tT /dev/sda
/dev/sda:
Timing buffer-cache reads: 3480 MB in 2.00 seconds = 1739.40 MB/sec
Timing buffered disk reads: 164 MB in 3.00 seconds = 54.64 MB/sec
root@t-rex root #
After reboot I can see this:
root@t-rex root # cat /proc/mdstat
Personalities : [raid0] [raid5] [multipath]
md1 : active raid5 sdc3[2] sdb3[1] sda3[0]
261730816 blocks level 5, 128k chunk, algorithm 2 [3/3] [UUU]
[=========>...........] resync = 49.5% (64786432/130865408)
finish=31.7min speed=34683K/sec
md0 : active raid0 sdc2[2] sdb2[1] sda2[0]
73256064 blocks 128k chunks
unused devices: <none>
My guess is that the driver sata_sil caches the transactions and doesnt
writes the data fast enough... and that leads to freezing.
Something maybe related:
I dont use the ata driver cause it never recognised 2 hds:
/dev/hde,f,g,h should be there but it took min 10 min to recognise them
and only e and g were accessible...
My kernel is: 2.6.7 from kernel.org
Ive googled a bit and fount that some ppl out there have similar problems.
Some explain they got over the problem with some modules compiled/not
compiled...
It doesnt work.
the config file is attached.
Sebastian
--
"Sie haben neue Mails!" - Die GMX Toolbar informiert Sie beim Surfen!
Jetzt aktivieren unter http://www.gmx.net/info
--
+++ Jetzt WLAN-Router f?r alle DSL-Einsteiger und Wechsler +++
GMX DSL-Powertarife zudem 3 Monate gratis* http://www.gmx.net/dsl
On Fri, 25 Jun 2004, Sebastian Slota wrote:
>ata3: DMA timeout, stat 0x1
>ATA: abnormal status 0xD8 on port 0xF8B09287
...
>4x160GB SATA on Sil3114 as SCSI /dev/sda ..sdd
I'm guessing Seagate drives? This is fixed in 2.6.7-bk7 and beyond.
(Your specific model drives may not be on the blacklist.)
--Ricky