Error in Linux CPU script osspi_cpuutil.sh (opclxcputil)

[Giirish]

Server Info :
[root@hpovtsmfb instrumentation]# uname -a
Linux hpovtsmfb 2.6.9-22.ELsmp #1 SMP Mon Sep 19 18:00:54 EDT 2005 x86_64 x86_64 x86_64 GNU/Linux

Script used : osspi_cpuutil.sh

Command Used in script :
"/opt/OV/bin/OpC/utils/opclxcputil 1 10"

Output of above command :
==============================================
root@hpovtsmfb instrumentation]# /opt/OV/bin/OpC/utils/opclxcputil 1 10
CPU states: %user %system %nice %idle
34 65 0 0
34 65 0 0
33 66 0 0
35 64 0 0
34 65 0 0
37 62 0 0
37 62 0 0
33 66 0 0
33 66 0 0
Floating point exception
==============================================


Two errors observed are :
1.Floating point exception error during script execution
2.Output shown by command used in script is showing wong value. Actual CPU utilization of system is 1%, but output through policy is showing 99%

Output of sar Command :

[root@hpovtsmfb instrumentation]# sar 1 10
Linux 2.6.9-22.ELsmp (hpovtsmfb) 12/01/2009

04:31:56 PM CPU %user %nice %system %iowait %idle
04:31:57 PM all 0.00 0.00 0.00 0.00 100.00
04:31:58 PM all 0.00 0.00 0.00 0.00 100.00
04:31:59 PM all 0.00 0.00 0.12 0.00 99.88
04:32:00 PM all 0.12 0.00 0.12 0.00 99.75
04:32:01 PM all 0.87 0.00 2.50 0.12 96.50

04:32:01 PM CPU %user %nice %system %iowait %idle
04:32:02 PM all 10.62 0.00 22.00 0.00 67.38
04:32:03 PM all 11.61 0.00 24.09 0.00 64.29
04:32:04 PM all 6.88 0.00 12.25 0.00 80.88
04:32:05 PM all 4.99 0.00 7.99 0.00 87.02
04:32:06 PM all 4.24 0.00 7.23 0.25 88.28
Average: all 3.93 0.00 7.63 0.04 88.40

Kindly let me know what could be reason for this Major difference/Floating point exception.

[Steven E. Protter]

Shalom,

Areas to investigate:

1) Bad OVPA software. You may need a patch
2) Os problem. This looks like RHEL 4 update 1 or 2. Patching/updating the OS might resolve floating point exception

As far as differences go, sar collects data its way, and OVPA/measureware has a totally different appraoch. Expect discrepencies.

SEP

[jhwong]

Hi All

I have the same issue also. I found out that when I issue sar command in my Linux box, the result is different from what ovo use opclxcputil. Below are the result of the both command running together at the same time

[root@hcsms1x1 ~]# /opt/OV/bin/OpC/utils/opclxcputil 5 5
CPU states: %user %system %nice %idle
57 42 0 0
50 50 0 0
28 71 0 0
33 66 0 0
50 50 0 0
Average: 43 56 0 0

[root@hcsms1x1 ~]# sar -u 5 5
Linux 2.6.9-67.ELsmp (hcsms1x1) 12/11/2009

10:50:58 AM CPU %user %nice %system %iowait %idle
10:51:03 AM all 0.10 0.00 0.07 0.00 99.83
10:51:08 AM all 0.07 0.00 0.12 0.00 99.80
10:51:13 AM all 0.05 0.00 0.10 0.00 99.85

10:51:13 AM CPU %user %nice %system %iowait %idle
10:51:18 AM all 0.05 0.00 0.12 0.00 99.83
10:51:23 AM all 0.10 0.00 0.12 0.00 99.78
Average: all 0.07 0.00 0.11 0.00 99.82


Anyone can advice me on this?

Thanks
Jun Hao

[Giirish]

I received the hot fix from the HP-Support.