Server Fault is a question and answer site for system and network administrators. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I have problem with slow disk transfer rate. It's static files server for our website. I was making backup of data and noticed that tar is very slow. So I did hdparm -t and...

hdparm -t /dev/sda3
Timing buffered disk reads:   6 MB in  4.70 seconds =   1.28 MB/sec

It's low traffic hour now on our site, so huge I/O traffic is not a reason (iotop show less than 1 MB/s).

It's RAID10 setup (2x2 SATA drives).

Unit  UnitType  Status         %RCmpl  %V/I/M  Stripe  Size(GB)  Cache  AVrfy
u0    RAID-10   OK             -       -       64K     1396.96   W      ON     

VPort Status         Unit Size      Type  Phy Encl-Slot    Model
p0    OK             u0   698.63 GB SATA  0   -            WDC WD7500AADS-00M2 
p1    OK             u0   698.63 GB SATA  1   -            WDC WD7500AADS-00M2 
p2    OK             u0   698.63 GB SATA  2   -            WDC WD7500AADS-00M2 
p3    OK             u0   698.63 GB SATA  3   -            WDC WD7500AADS-00M2

We have recently changed almost all components of server (excluding 3ware controller + disks). And I think problems started since then. Can it be configuration problem or hardware?


I found something like that in dmesg

[166843.625843] irq 16: nobody cared (try booting with the "irqpoll" option)
[166843.625846] Pid: 0, comm: swapper Not tainted 3.1.5-gentoo #3
[166843.625847] Call Trace:
[166843.625848]  <IRQ>  [<ffffffff810859d5>] __report_bad_irq+0x35/0xc1
[166843.625856]  [<ffffffff81085cec>] note_interrupt+0x165/0x1e1
[166843.625859]  [<ffffffff8108445f>] handle_irq_event_percpu+0x16f/0x187
[166843.625861]  [<ffffffff810844a9>] handle_irq_event+0x32/0x51
[166843.625863]  [<ffffffff8108640b>] handle_fasteoi_irq+0x75/0x99
[166843.625866]  [<ffffffff810039d7>] handle_irq+0x83/0x8b
[166843.625868]  [<ffffffff810036ad>] do_IRQ+0x48/0xa0
[166843.625871]  [<ffffffff8155082b>] common_interrupt+0x6b/0x6b
[166843.625872]  <EOI>  [<ffffffff812981e8>] ? acpi_safe_halt+0x22/0x35
[166843.625877]  [<ffffffff812981e2>] ? acpi_safe_halt+0x1c/0x35
[166843.625879]  [<ffffffff81298216>] acpi_idle_do_entry+0x1b/0x2b
[166843.625881]  [<ffffffff81298276>] acpi_idle_enter_c1+0x50/0x99
[166843.625884]  [<ffffffff813b792a>] cpuidle_idle_call+0xed/0x171
[166843.625886]  [<ffffffff81001257>] cpu_idle+0x55/0x81
[166843.625888]  [<ffffffff81532a69>] rest_init+0x6d/0x6f
[166843.625891]  [<ffffffff81aa1aca>] start_kernel+0x329/0x334
[166843.625893]  [<ffffffff81aa12a6>] x86_64_start_reservations+0xb6/0xba
[166843.625894]  [<ffffffff81aa139c>] x86_64_start_kernel+0xf2/0xf9
[166843.625896] handlers:
[166843.625898] [<ffffffff812dc8de>] twl_interrupt
[166843.625900] Disabling IRQ #16

It's related to problem?

EDIT #2:

Based on feedback in comments, here is more informations.

cat /proc/interrupts

16:     390813          0          0          0   IO-APIC-fasteoi   3w-sas

Controller model:

[    1.095350] 3ware Storage Controller device driver for Linux v1.26.02.003.
[    1.095467] 3ware 9000 Storage Controller device driver for Linux v2.26.02.014.
[    1.095641] LSI 3ware SAS/SATA-RAID Controller device driver for Linux v3.26.02.000.
[    1.095787] 3w-sas 0000:01:00.0: PCI INT A -> GSI 16 (level, low) -> IRQ 16
[    1.095881] 3w-sas 0000:01:00.0: setting latency timer to 64
[    1.910801] 3w-sas: scsi0: Found an LSI 3ware 9750-4i Controller at 0xfe560000, IRQ: 16.
[    2.216537] 3w-sas: scsi0: Firmware FH9X, BIOS BE9X, Phys: 8.
[    2.216836] scsi 0:0:0:0: Direct-Access     LSI      9750-4i    DISK  5.08 PQ: 0 ANSI: 5

And motherboard:

description: Motherboard
product: P8H67-M
vendor: ASUSTeK Computer INC.
share|improve this question
What's the output of cat /proc/interrupts | grep 16:? – David Schwartz Apr 7 '12 at 7:41
It's 3ware controller (and nothing else, so it's not shared with any other component) – Nooklez Apr 7 '12 at 9:45
Then that's your problem. It would help a lot to know what 3ware controller you have, what motherboard you have, and what type of interrupt it was. (APIC edge? fasteoi? MSI? etcetera.) – David Schwartz Apr 7 '12 at 10:11
Question edited with more info. – Nooklez Apr 7 '12 at 10:28
I recompiled kernel and after 4 days there is no sign of this problem anymore. – Nooklez Apr 12 '12 at 9:04

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.