On Wednesday 17 May 2006 20:26, you wrote:
> sd 3:0:0:0: SCSI error: return code = 0xb0000
> end_request: I/O error, dev sdc, sector 439427200
> device-mapper: dm-multipath: Failing path 8:32.
> end_request: I/O error, dev sdc, sector 439427208
> mptbase: ioc2: LogInfo(0x11070000): F/W: DMA Error
> mptbase: ioc2: IOCStatus(0x004b): SCSI IOC Terminated
> mptbase: ioc2: LogInfo(0x11070000): F/W: DMA Error
> mptbase: ioc2: IOCStatus(0x004b): SCSI IOC Terminated
I forgot to include the following error, which I get each time:
mptbase: ioc2: IOCStatus(0x0003): Invalid SGL
mptbase: ioc2: LogInfo(0x11070000): F/W: DMA Error
> This error appears randomly on both interfaces, under heavy I/O. And it
> does not seems to occur when using only one interface (/dev/sdb directly,
> rather than /dev/md0 or /dev/mapper/<id>.
Actually, that's not true. ;) It just took more time to appear : I get the
same errors when only using one of the two interfaces (only one SCSI link). So
it's probably not a MD-related issue. It seems to be more related to the mpt
driver (I currently use the 2.6.16-1-em64t-p4-smp debian kernel).
Does that ring a bell anywhere?
ioc0: 53C1030: Capabilities={Initiator}
scsi0 : ioc0: LSI53C1030, FwRev=01030a00h, Ports=1, MaxQ=222, IRQ=201
Vendor: IFT Model: A16U-G1410 Rev: 342J
Type: Direct-Access ANSI SCSI revision: 05
GSI 21 sharing vector 0xD1 and IRQ 21
ACPI: PCI Interrupt 0000:03:0b.1[B] -> GSI 38 (level, low) -> IRQ 209
mptbase: Initiating ioc1 bringup
ioc1: 53C1030: Capabilities={Initiator}
scsi1 : ioc1: LSI53C1030, FwRev=01030a00h, Ports=1, MaxQ=222, IRQ=209
GSI 22 sharing vector 0xD9 and IRQ 22
ACPI: PCI Interrupt 0000:05:04.0[A] -> GSI 106 (level, low) -> IRQ 217
mptbase: Initiating ioc2 bringup
ioc2: 53C1030: Capabilities={Initiator}
scsi2 : ioc2: LSI53C1030, FwRev=01030a00h, Ports=1, MaxQ=222, IRQ=217
GSI 23 sharing vector 0xE1 and IRQ 23
ACPI: PCI Interrupt 0000:05:04.1[B] -> GSI 107 (level, low) -> IRQ 225
mptbase: Initiating ioc3 bringup
ioc3: 53C1030: Capabilities={Initiator}
scsi3 : ioc3: LSI53C1030, FwRev=01030a00h, Ports=1, MaxQ=222, IRQ=225
GSI 24 sharing vector 0xE9 and IRQ 24
ACPI: PCI Interrupt 0000:00:1d.7[D] -> GSI 23 (level, low) -> IRQ 233
PCI: Setting latency timer of device 0000:00:1d.7 to 64
PCI: cache line size of 128 is not supported by device 0000:00:1d.7
sda : very big device. try to use READ CAPACITY(16).
SCSI device sda: 5466216448 512-byte hdwr sectors (2798703 MB)
sda: Write Protect is off
sda: Mode Sense: 9b 00 00 08
SCSI device sda: drive cache: write through
sda : very big device. try to use READ CAPACITY(16).
SCSI device sda: 5466216448 512-byte hdwr sectors (2798703 MB)
sda: Write Protect is off
sda: Mode Sense: 9b 00 00 08
SCSI device sda: drive cache: write through
sda: unknown partition table
sd 0:0:1:0: Attached scsi disk sda
sd 0:0:1:0: SCSI error: return code = 0xb0000
end_request: I/O error, dev sda, sector 5466216320
Buffer I/O error on device sda, logical block 683277040
sd 0:0:1:0: SCSI error: return code = 0xb0000
end_request: I/O error, dev sda, sector 5466216320
Buffer I/O error on device sda, logical block 683277040
sd 0:0:1:0: SCSI error: return code = 0xb0000
end_request: I/O error, dev sda, sector 5466216440
Buffer I/O error on device sda, logical block 683277055
sd 0:0:1:0: SCSI error: return code = 0xb0000
end_request: I/O error, dev sda, sector 5466216440
Buffer I/O error on device sda, logical block 683277055
sd 0:0:1:0: SCSI error: return code = 0xb0000
end_request: I/O error, dev sda, sector 5466216440
Buffer I/O error on device sda, logical block 683277055
sd 0:0:1:0: SCSI error: return code = 0xb0000
end_request: I/O error, dev sda, sector 5466216440
Buffer I/O error on device sda, logical block 683277055
sd 0:0:1:0: SCSI error: return code = 0xb0000
end_request: I/O error, dev sda, sector 5466216440
Buffer I/O error on device sda, logical block 683277055
sd 0:0:1:0: SCSI error: return code = 0xb0000
end_request: I/O error, dev sda, sector 5466216440
Buffer I/O error on device sda, logical block 683277055
sd 0:0:1:0: SCSI error: return code = 0xb0000
end_request: I/O error, dev sda, sector 5466216384
Buffer I/O error on device sda, logical block 683277048
What do I have to check? The RAID array seems in good shape, the cables are
brand new, as are the HBAs. I'm running out of ideas, so I'd really
appreciate any clue.
Thanks in advance,
--
Kilian CAVALOTTI Administrateur réseaux et systèmes
UPMC / CNRS - LIP6 (C870)
8, rue du Capitaine Scott Tel. : 01 44 27 88 54
75015 Paris - France Fax. : 01 44 27 70 00
-
: send the line "unsubscribe linux-scsi" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
