Skip to contentSkip to conversation

dell poweredge 2400 harddisks going into offline mode when heavy I/O occurs

DISCUSSION Folkert van Heusden 3 participants 4 messages Sep 28, 2006 – Oct 6, 2006

lore
Share & tools
lore
#0 lore raw

Hi,

I have a Dell Poweredge 2400 with 6 scsi harddisks in (hw-) raid 5. 512MB ram, 2x P3.
When heavy disk i/o occurs, the system puts the harddisks into offline mode causing the filesystems to be put in readonly. The current kernel is 2.6.8, with 2.4.27 this did not occure. Googling did not help. The disks all have green lights (there's a special led for each to indicate errors - that one is off).

dmesg:
-----
SCSI subsystem initialized
ACPI: PCI interrupt 0000:01:06.0[A] -> GSI 30 (level, low) -> IRQ 193 scsi0 : Adaptec AIC7XXX EISA/VLB/PCI SCSI HBA DRIVER, Rev 6.2.36
<Adaptec aic7880 Ultra SCSI adapter>
aic7880: Ultra Single Channel A, SCSI Id=7, 16/253 SCBs

Using anticipatory io scheduler
(scsi0:A:5): 20.000MB/s transfers (20.000MHz, offset 15)
Vendor: NEC Model: CD-ROM DRIVE:466 Rev: 1.06
Type: CD-ROM ANSI SCSI revision: 02
Red Hat/Adaptec aacraid driver (1.1.2-lk2 Sep 7 2006)
ACPI: PCI interrupt 0000:00:02.1[A] -> GSI 31 (level, low) -> IRQ 177
AAC0: kernel 2.5.4 build 2991
AAC0: monitor 2.5.4 build 2991
AAC0: bios 2.5.0 build 2991
AAC0: serial 7b0c10d0fafaf001
scsi1 : percraid
Vendor: DELL Model: PERCRAID RAID5 Rev: V1.0
Type: Direct-Access ANSI SCSI revision: 02
Attached scsi generic sg0 at scsi0, channel 0, id 5, lun 0, type 5 Attached scsi generic sg1 at scsi1, channel 0, id 0, lun 0, type 0 SCSI device sda: 88855680 512-byte hdwr sectors (45494 MB)
sda: Write Protect is off
sda: Mode Sense: 03 00 00 00
SCSI device sda: drive cache: write through
/dev/scsi/host1/bus0/target0/lun0: p1 p2
Attached scsi removable disk sda at scsi1, channel 0, id 0, lun 0

errors:
------

Sep 28 16:05:12 kasparov kernel: aacraid: Host adapter reset request. SCSI hang ?
Sep 28 16:06:12 kasparov kernel: aacraid: SCSI bus appears hung
Sep 28 16:06:12 kasparov kernel: scsi: Device offlined - not ready after error recovery: host 1 channel 0 id 0 lun 0
Sep 28 16:06:12 kasparov last message repeated 99 times
Sep 28 16:06:12 kasparov kernel: SCSI error : <1 0 0 0> return code = 0x6000000
Sep 28 16:06:12 kasparov kernel: end_request: I/O error, dev sda, sector 8827879
Sep 28 16:06:12 kasparov kernel: Buffer I/O error on device dm-2, logical block 677445
Sep 28 16:06:12 kasparov kernel: lost page write due to I/O error on dm-2
Sep 28 16:06:12 kasparov kernel: scsi1 (0:0): rejecting I/O to offline device
Sep 28 16:06:12 kasparov kernel: Buffer I/O error on device dm-2, logical block 677446
Sep 28 16:06:12 kasparov kernel: lost page write due to I/O error on dm-2
Sep 28 16:06:12 kasparov kernel: Buffer I/O error on device dm-2, logical block 677447
Sep 28 16:06:12 kasparov kernel: lost page write due to I/O error on dm-2
Sep 28 16:06:12 kasparov kernel: Buffer I/O error on device dm-2, logical block 677448
Sep 28 16:06:12 kasparov kernel: lost page write due to I/O error on dm-2
Sep 28 16:06:12 kasparov kernel: Buffer I/O error on device dm-2, logical block 677449
Sep 28 16:06:12 kasparov kernel: lost page write due to I/O error on dm-2
Sep 28 16:06:12 kasparov kernel: Buffer I/O error on device dm-2, logical block 677450
Sep 28 16:06:12 kasparov kernel: lost page write due to I/O error on dm-2
Sep 28 16:06:12 kasparov kernel: Buffer I/O error on device dm-2, logical block 677451
Sep 28 16:06:12 kasparov kernel: lost page write due to I/O error on dm-2
Sep 28 16:06:12 kasparov kernel: Buffer I/O error on device dm-2, logical block 677452
Sep 28 16:06:12 kasparov kernel: lost page write due to I/O error on dm-2
Sep 28 16:06:12 kasparov kernel: Buffer I/O error on device dm-2, logical block 677453
Sep 28 16:06:13 kasparov kernel: lost page write due to I/O error on dm-2
Sep 28 16:06:13 kasparov kernel: Buffer I/O error on device dm-2, logical block 677454
Sep 28 16:06:13 kasparov kernel: lost page write due to I/O error on dm-2
Sep 28 16:06:13 kasparov kernel: scsi1 (0:0): rejecting I/O to offline device
Sep 28 16:06:13 kasparov last message repeated 88 times
Sep 28 16:06:13 kasparov syslogd: /var/log/syslog: Read-only file system
Sep 28 16:06:13 kasparov syslogd: /var/log/kern.log: Read-only file system
Sep 28 16:06:13 kasparov syslogd: /var/log/messages: Read-only file system
Sep 28 16:06:13 kasparov kernel: SCSI error : <1 0 0 0> return code = 0x6000000
Sep 28 16:06:13 kasparov kernel: end_request: I/O error, dev sda, sector 8828007
Sep 28 16:06:13 kasparov kernel: scsi1 (0:0): rejecting I/O to offline device
...
Sep 28 16:06:13 kasparov kernel: SCSI error : <1 0 0 0> return code = 0x6000000
Sep 28 16:06:13 kasparov kernel: end_request: I/O error, dev sda, sector 8830951
Sep 28 16:06:13 kasparov kernel: scsi1 (0:0): rejecting I/O to offline device
Sep 28 16:06:13 kasparov last message repeated 2 times
Sep 28 16:06:13 kasparov kernel: Aborting journal on device dm-4.
Sep 28 16:06:13 kasparov kernel: scsi1 (0:0): rejecting I/O to offline device
Sep 28 16:06:13 kasparov last message repeated 3 times
Sep 28 16:06:13 kasparov kernel: Aborting journal on device dm-1.
Sep 28 16:06:13 kasparov kernel: scsi1 (0:0): rejecting I/O to offline device
Sep 28 16:06:13 kasparov last message repeated 31 times
Sep 28 16:06:13 kasparov kernel: ext3_abort called.
Sep 28 16:06:13 kasparov kernel: EXT3-fs abort (device dm-4): ext3_journal_start: Detected aborted journal
Sep 28 16:06:13 kasparov kernel: Remounting filesystem read-only
Sep 28 16:06:13 kasparov kernel: EXT3-fs error (device dm-4) in start_transaction: Journal has aborted
Sep 28 16:06:13 kasparov kernel: scsi1 (0:0): rejecting I/O to offline device
Sep 28 16:06:13 kasparov last message repeated 2 times
Sep 28 16:06:13 kasparov kernel: EXT3-fs error (device dm-4) in start_transaction: Journal has aborted
Sep 28 16:06:13 kasparov kernel: scsi1 (0:0): rejecting I/O to offline device
Sep 28 16:06:13 kasparov kernel: EXT3-fs error (device dm-4) in start_transaction: Journal has aborted
Sep 28 16:06:13 kasparov kernel: scsi1 (0:0): rejecting I/O to offline device
Sep 28 16:06:13 kasparov last message repeated 8 times
Sep 28 16:06:13 kasparov kernel: SCSI error : <1 0 0 0> return code = 0x6000000
Sep 28 16:06:13 kasparov kernel: end_request: I/O error, dev sda, sector 8831079
Sep 28 16:06:13 kasparov kernel: scsi1 (0:0): rejecting I/O to offline device
...
Sep 28 16:06:14 kasparov kernel: SCSI error : <1 0 0 0> return code = 0x6000000
Sep 28 16:06:14 kasparov kernel: end_request: I/O error, dev sda, sector 11177631
Sep 28 16:06:14 kasparov kernel: ext3_abort called.
Sep 28 16:06:14 kasparov kernel: EXT3-fs abort (device dm-1): ext3_journal_start: Detected aborted journal
Sep 28 16:06:14 kasparov kernel: Remounting filesystem read-only

proc:
----

kasparov:/home/folkert# cat /proc/interrupts
           CPU0       CPU1
  0:     159442     182242    IO-APIC-edge  timer
  1:          0          9    IO-APIC-edge  i8042
  7:          2          0    IO-APIC-edge  parport0
  8:          3          1    IO-APIC-edge  rtc
  9:          0          0   IO-APIC-level  acpi
 10:          0          0   IO-APIC-level  ohci_hcd
177:       3314       3540   IO-APIC-level  aacraid
185:      33638        839   IO-APIC-level  eth0
193:         53         53   IO-APIC-level  aic7xxx
NMI:          0          0
LOC:     341531     341530
ERR:          0
MIS:          0
kasparov:/home/folkert# cat /proc/scsi/aic7xxx/0
Adaptec AIC7xxx driver version: 6.2.36
Adaptec aic7880 Ultra SCSI adapter
aic7880: Ultra Single Channel A, SCSI Id=7, 16/253 SCBs
Allocated SCBs: 4, SG List Length: 128

Serial EEPROM:
0xc358 0xc358 0xc358 0xc358 0xc358 0xc358 0xc358 0xc358 0xc378 0xc378 0xc378 0xc378 0xc378 0xc378 0xc378 0xc378 0x58e6 0x5c5e 0x2807 0x0008 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0xffff 0x0250 0x1619

Target 0 Negotiation Settings

        User: 20.000MB/s transfers (20.000MHz, offset 127)
Target 1 Negotiation Settings
        User: 20.000MB/s transfers (20.000MHz, offset 127)
Target 2 Negotiation Settings
        User: 20.000MB/s transfers (20.000MHz, offset 127)
Target 3 Negotiation Settings
        User: 20.000MB/s transfers (20.000MHz, offset 127)
Target 4 Negotiation Settings
        User: 20.000MB/s transfers (20.000MHz, offset 127)

Target 5 Negotiation Settings
User: 20.000MB/s transfers (20.000MHz, offset 127)
Goal: 20.000MB/s transfers (20.000MHz, offset 15)
Curr: 20.000MB/s transfers (20.000MHz, offset 15) Channel A Target 5 Lun 0 Settings
Commands Queued 8
Commands Active 0
Command Openings 1
Max Tagged Openings 0
Device Queue Frozen Count 0
Target 6 Negotiation Settings

        User: 20.000MB/s transfers (20.000MHz, offset 127)
Target 7 Negotiation Settings
        User: 20.000MB/s transfers (20.000MHz, offset 127)
kasparov:/home/folkert# cat /proc/scsi/aacraid/1
percraid

Folkert van Heusden

Signature · 2 lines
--------------------------------------------------------------------
Phone: +31-6-41278122, PGP-key: 1F28D8AE, www.vanheusden.com
#1 lore raw
Quoted 9 lines · Folkert van Heusden

Hi,

I have a Dell Poweredge 2400 with 6 scsi harddisks in (hw-) raid 5. 512MB ram, 2x P3.
When heavy disk i/o occurs, the system puts the harddisks into offline mode causing the filesystems to be put in readonly. The current kernel is 2.6.8, with 2.4.27 this did not occure. Googling did not help. The disks all have green lights (there's a special led for each to indicate errors - that one is off).

[snip]

Quoted 3 lines
Sep 28 16:05:12 kasparov kernel: aacraid: Host adapter reset request. SCSI hang ?
Sep 28 16:06:12 kasparov kernel: aacraid: SCSI bus appears hung
Sep 28 16:06:12 kasparov kernel: scsi: Device offlined - not ready after error recovery: host 1 channel 0 id 0 lun 0

Yes, this is familiar. See:

http://lists.us.dell.com/pipermail/linux-poweredge/2004-May/014694.html

In addition, please consider mounting your file systems with 'noatime', as this reduces the number of small writes being sent to the disks.

2.6.x kernels have the ability to swamp the RAID controller firmware with requests where 2.4.x kernels couldn't so easily.

Thanks,
Matt

Signature · 4 lines
Matt Domsch
Software Architect
Dell Linux Solutions linux.dell.com & www.dell.com/linux
Linux on Dell mailing lists @ http://lists.us.dell.com
#2 lore raw
Quoted 18 lines · Matt Domsch
Quoted 9 lines · Folkert van Heusden

Hi,

I have a Dell Poweredge 2400 with 6 scsi harddisks in (hw-) raid 5. 512MB ram, 2x P3.
When heavy disk i/o occurs, the system puts the harddisks into offline mode causing the filesystems to be put in readonly. The current kernel is 2.6.8, with 2.4.27 this did not occure. Googling did not help. The disks all have green lights (there's a special led for each to indicate errors - that one is off).

[snip]

Quoted 3 lines
Sep 28 16:05:12 kasparov kernel: aacraid: Host adapter reset request. SCSI hang ?
Sep 28 16:06:12 kasparov kernel: aacraid: SCSI bus appears hung
Sep 28 16:06:12 kasparov kernel: scsi: Device offlined - not ready after error recovery: host 1 channel 0 id 0 lun 0

Yes, this is familiar. See:

Quoted 8 lines

http://lists.us.dell.com/pipermail/linux-poweredge/2004-May/014694.html

In addition, please consider mounting your file systems with 'noatime', as this reduces the number of small writes being sent to the disks.

2.6.x kernels have the ability to swamp the RAID controller firmware with requests where 2.4.x kernels couldn't so easily.

Can you configure the controller as JBOD and use software raid. Would the controller keep up with that?

Quoted 4 lines

Thanks,
Matt

#3 lore raw
Quoted 23 lines · Bill Davidsen
Quoted 22 lines · Matt Domsch
Quoted 9 lines · Folkert van Heusden

Hi,

I have a Dell Poweredge 2400 with 6 scsi harddisks in (hw-) raid 5. 512MB ram, 2x P3.
When heavy disk i/o occurs, the system puts the harddisks into offline mode causing the filesystems to be put in readonly. The current kernel is 2.6.8, with 2.4.27 this did not occure. Googling did not help. The disks all have green lights (there's a special led for each to indicate errors - that one is off).

[snip]

Quoted 5 lines
Sep 28 16:05:12 kasparov kernel: aacraid: Host adapter reset request. 
SCSI hang ?
Sep 28 16:06:12 kasparov kernel: aacraid: SCSI bus appears hung
Sep 28 16:06:12 kasparov kernel: scsi: Device offlined - not ready after 

error recovery: host 1 channel 0 id 0 lun 0

Yes, this is familiar. See:

http://lists.us.dell.com/pipermail/linux-poweredge/2004-May/014694.html

Quoted 8 lines
Quoted 6 lines

In addition, please consider mounting your file systems with 'noatime', as this reduces the number of small writes being sent to the disks.

2.6.x kernels have the ability to swamp the RAID controller firmware with requests where 2.4.x kernels couldn't so easily.

Can you configure the controller as JBOD and use software raid.

Yes, you can. It's an option in the BIOS SETUP screen during POST.

Quoted 1 line

Would the controller keep up with that?

Yes. (This is what I had done on my PE2400 for years too, before I replaced it entirely). The SCSI controller is then driven by aic7xxx, 80MB/sec capable, though in that unit it's only a single SCSI channel and the backplane is a single 1x6 backplane, so you can't split the bus in half; but that's no different than what you've already got.

Thanks,
Matt

Signature · 4 lines
Matt Domsch
Software Architect
Dell Linux Solutions linux.dell.com & www.dell.com/linux
Linux on Dell mailing lists @ http://lists.us.dell.com