High CPU Usage percentage.

SUSE Linux Enterprise Server SLES Point of Service
1

Hello,
I have a Suse 10.2 system that is running with a very high CPU usage percentage. If i do a “vgs” or a "fdisk -l " my request hangs. We recently had some SAN work done and i believe something in the disk subsystem is confused.
i have no one process that i feel i can point to. I’m considering rebooting the system. Anyone have any suggestions prior to my just doing a reboot?

Thanks.

–#vmstat 10
procs -----------memory---------- —swap-- -----io---- -system-- -----cpu------
r b swpd free buff cache si so bi bo in cs us sy id wa st
1 8 691232 184340 163964 31337548 4 4 4403 435 0 0 6 1 78 15 0
0 8 691232 183736 164068 31338472 0 0 2 64 272 840 1 0 0 99 0
0 8 691232 183704 164132 31338408 0 0 1 80 287 765 0 0 0 100 0
0 8 691232 183728 164216 31338324 0 0 0 67 274 716 0 0 0 100 0

top - 11:04:11 up 507 days, 9:20, 3 users, load average: 8.01, 8.04, 8.00
Tasks: 198 total, 1 running, 196 sleeping, 0 stopped, 1 zombie
Cpu(s): 0.0%us, 0.1%sy, 0.0%ni, 0.0%id, 99.9%wa, 0.0%hi, 0.0%si, 0.0%st
Mem: 32936168k total, 32752300k used, 183868k free, 163884k buffers
Swap: 50336984k total, 691232k used, 49645752k free, 31337628k cached

–>uptime
11:12am up 507 days 9:29, 7 users, load average: 8.71, 8.24, 8.07

2

Hi fbngnzlz,

[QUOTE=fbngnzlz;17030]Hello,
I have a Suse 10.2 system that is running with a very high CPU usage percentage. If i do a “vgs” or a "fdisk -l " my request hangs. We recently had some SAN work done and i believe something in the disk subsystem is confused.
i have no one process that i feel i can point to. I’m considering rebooting the system. Anyone have any suggestions prior to my just doing a reboot?

Thanks.

–#vmstat 10
procs -----------memory---------- —swap-- -----io---- -system-- -----cpu------
r b swpd free buff cache si so bi bo in cs us sy id wa st
1 8 691232 184340 163964 31337548 4 4 4403 435 0 0 6 1 78 15 0
0 8 691232 183736 164068 31338472 0 0 2 64 272 840 1 0 0 99 0
0 8 691232 183704 164132 31338408 0 0 1 80 287 765 0 0 0 100 0
0 8 691232 183728 164216 31338324 0 0 0 67 274 716 0 0 0 100 0[/QUOTE]

you actually have no high CPU usage - close to 0% (add “us” and “sy” columns, as in “user-space” and “system CPU”). All time is eaten up waiting for disk I/O to complete (99 to 100% “wa” as in “waiting for I/O”).

You might try to pin-point the cause by checking with “iostat” and more important “dmesg”… but as you already wrote, the SAN works are the most probable cause.

If it doesn’t hurt too much, I’d go for the reboot.

Regards,
Jens