[CentOS] High softirq usage in Centos 5

Oguz Yilmaz oguzyilmazlist at gmail.com
Wed Jan 7 07:14:45 UTC 2009


Hi,

I have a machine under heavy network traffic.
Kernel is Centos kernel 2.6.18 SMP 32 bit.

Ethernets are
05:00.0 Ethernet controller: Intel Corporation 80003ES2LAN Gigabit Ethernet
Controller (Copper) (rev 01)
05:00.1 Ethernet controller: Intel Corporation 80003ES2LAN Gigabit Ethernet
Controller (Copper) (rev 01)
06:01.0 Ethernet controller: Intel Corporation 82541PI Gigabit Ethernet
Controller (rev 05)
06:02.0 Ethernet controller: Intel Corporation 82541PI Gigabit Ethernet
Controller (rev 05)

The mainboard is with Intel 5000 series chipset Asus board.

When I use 2.6.9 kernel with the same location and traffic I have had no
softirq usage. Also in 2.4 Centos 3 machine I have had no softirq usage.

With the same location and traffic I have softirq usage %25:
Cpu0  :  2.0%us,  2.0%sy,  0.0%ni, 71.3%id,  0.0%wa,  2.7%hi, 22.0%si,
0.0%st
Cpu1  :  1.0%us,  1.3%sy,  0.0%ni, 79.7%id,  0.0%wa,  1.3%hi, 16.6%si,
0.0%st

I have no extra parameters in grub.conf.

Is it a known problem?

What do you suggest?

Regards,

Oguz Yilmaz



# cat /proc/interrupts
           CPU0       CPU1
  0:   12475556   27878448    IO-APIC-edge  timer
  1:          2          1    IO-APIC-edge  i8042
  4:         20         15    IO-APIC-edge  serial
  8:          3          0    IO-APIC-edge  rtc
  9:          0          0   IO-APIC-level  acpi
 12:          1          3    IO-APIC-edge  i8042
 58:       3194     551096   IO-APIC-level  ahci
 66:          0          0   IO-APIC-level  ehci_hcd:usb1, uhci_hcd:usb2,
uhci_hcd:usb4
 74:          0          0   IO-APIC-level  uhci_hcd:usb3, uhci_hcd:usb5
 82:         13      20148   IO-APIC-level  eth0
 98:   49173115       8304   IO-APIC-level  eth1
106:  165302876          0         PCI-MSI  eth3
114:      20091  188108320         PCI-MSI  eth4
169:         61      22012   IO-APIC-level  eth2
NMI:          0          0
LOC:   40353900   40353898
ERR:          0
MIS:          0

# cat /proc/stat
cpu  109394 94 48981 7357582 21023 59947 475284 0
cpu0 54410 40 25805 3688679 3670 33482 230074 0
cpu1 54984 53 23176 3668902 17353 26465 245210 0
intr 443941445 40367708 3 0 2 35 0 0 0 3 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 554483 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 20168 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
49274302 0 0 0 0 0 0 0 165429410 0 0 0 0 0 0 0 188273249 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 22080 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0
ctxt 35284048
btime 1231271275
processes 292966
procs_running 1
procs_blocked 0

# vmstat
procs -----------memory---------- ---swap-- -----io---- --system--
-----cpu------
 r  b   swpd   free   buff  cache   si   so    bi    bo   in   cs us sy id
wa st
 0  0      0 3220064 163084 412172    0    0     5    29  183  438  1  7 91
0  0
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.centos.org/pipermail/centos/attachments/20090107/2255e3f9/attachment.html 


More information about the CentOS mailing list