Performance reporting issues. (update)

Waldher, Travis R Travis.R.Waldher at boeing.com
Thu Oct 23 22:10:39 UTC 2008


Ok, here's the output from sar -P ALL

02:40:01 PM       all     28.47      0.00     70.92      0.61      0.00
02:40:01 PM         0      5.30      0.00     13.28      0.05     81.37
02:40:01 PM         1      1.39      0.00      4.07      0.08     94.47
02:40:01 PM         2      5.26      0.00     12.63      0.07     82.03
02:40:01 PM         3      1.31      0.00      3.06      0.09     95.54
02:50:01 PM       all     27.33      0.00     72.27      0.40      0.00
02:50:01 PM         0      3.48      0.00      9.91      0.03     86.57
02:50:01 PM         1      1.22      0.00      3.80      0.05     94.94
02:50:01 PM         2      3.10      0.00      7.58      0.02     89.31
02:50:01 PM         3      1.06      0.00      2.14      0.03     96.77

Please note, the individual CPU time and combination is not making any sense.  That is from today.  Let's go back to where it was working on October 19th.

04:40:01 AM       all      0.75      0.00      2.32      0.01     96.92
04:40:01 AM         0      0.90      0.00      4.11      0.02     94.98
04:40:01 AM         1      0.84      0.00      3.37      0.01     95.78
04:40:01 AM         2      0.75      0.00      1.15      0.02     98.08
04:40:01 AM         3      0.52      0.00      0.62      0.00     98.85
04:50:01 AM       all      2.49      0.00      6.60      0.03     90.88
04:50:01 AM         0      1.24      0.00      4.42      0.02     94.32
04:50:01 AM         1      0.76      0.00      3.31      0.00     95.92
04:50:01 AM         2      1.22      0.00      1.60      0.02     97.16
04:50:01 AM         3      0.60      0.00      0.77      0.01     98.63
05:00:01 AM       all     29.22      0.00     70.35      0.43      0.00
05:00:01 AM         0      1.50      0.00      4.32      0.02     94.16
05:00:01 AM         1      0.81      0.00      3.11      0.01     96.08
05:00:01 AM         2      1.48      0.00      2.04      0.03     96.45
05:00:01 AM         3      0.57      0.00      1.02      0.01     98.41
05:10:01 AM       all     28.77      0.00     70.90      0.33      0.00
05:10:01 AM         0      1.58      0.00      4.91      0.02     93.49
05:10:01 AM         1      0.89      0.00      3.52      0.01     95.58
05:10:01 AM         2      1.56      0.00      2.05      0.02     96.37
05:10:01 AM         3      0.61      0.00      0.94      0.01     98.44

I'm still not sure how it is figuring an average or whatever it is figuring for "ALL", but it appears to at least be more accurate up until 0500.

Sysstat version sysstat-5.0.5-1

Anyone have any advice?

From: redhat-install-list-bounces at redhat.com [mailto:redhat-install-list-bounces at redhat.com] On Behalf Of Waldher, Travis R
Sent: Thursday, October 16, 2008 10:38 AM
To: Getting started with Red Hat Linux
Subject: Performance reporting issues.

I'm having an odd issue where SAR is reporting wildly different stats than what the system is doing.

I'm not sure where sa is pulling it's numbers from but I definitely don't and haven't had the CPUs running 36% for users and 63% for system for a long time.  It's more likely the system had been idle.

Any ideas?

$ sar -H -t -u -f /var/log/sa/sa16
<hostname>;600;2008-10-16 09:10:01;-1;33.36;0.00;66.25;0.40;0.00
<hostname>;599;2008-10-16 09:20:01;-1;34.35;0.00;65.21;0.44;0.00
<hostname>;600;2008-10-16 09:30:01;-1;34.51;0.00;65.06;0.42;0.00

$ top
top - 09:38:52 up 130 days,  3:08,  2 users,  load average: 0.07, 0.03, 0.01
Tasks: 129 total,   1 running, 127 sleeping,   0 stopped,   1 zombie
Cpu0  :  0.1% us,  0.3% sy,  0.0% ni, 99.6% id,  0.0% wa,  0.0% hi,  0.0% si
Cpu1  :  0.2% us,  0.3% sy,  0.0% ni, 99.5% id,  0.0% wa,  0.0% hi,  0.0% si
Cpu2  :  0.1% us,  0.3% sy,  0.0% ni, 99.6% id,  0.0% wa,  0.0% hi,  0.0% si
Cpu3  :  0.1% us,  0.1% sy,  0.0% ni, 99.8% id,  0.0% wa,  0.0% hi,  0.0% si
Cpu4  :  0.1% us,  0.1% sy,  0.0% ni, 99.8% id,  0.0% wa,  0.0% hi,  0.0% si
Cpu5  :  0.1% us,  0.1% sy,  0.0% ni, 99.9% id,  0.0% wa,  0.0% hi,  0.0% si
Cpu6  :  0.0% us,  0.0% sy,  0.0% ni, 99.9% id,  0.0% wa,  0.0% hi,  0.0% si
Cpu7  :  0.1% us,  0.0% sy,  0.0% ni, 99.9% id,  0.0% wa,  0.0% hi,  0.0% si



-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://listman.redhat.com/archives/redhat-install-list/attachments/20081023/ded7df29/attachment.htm>


More information about the Redhat-install-list mailing list