Hi Jack!
Vmstat is the wrong tool to measure CPU. You should use sar or mpstat.
I prefer "mpstat" because it's more modern and provides more data:
root@umajor:/home/mgogala# mpstat 3 3
Linux 5.3.0-24-generic (umajor) 11/17/2019 _x86_64_ (16 CPU)
08:13:40 PM CPU %usr %nice %sys %iowait %irq %soft %steal
%guest %gnice %idle
08:13:43 PM all 0.06 0.00 0.04 0.00 0.00 0.00 0.00
0.00 0.00 99.90
08:13:46 PM all 0.15 0.00 0.25 0.00 0.00 0.00 0.00
0.00 0.00 99.60
08:13:49 PM all 0.04 0.00 0.06 0.00 0.00 0.00 0.00
0.00 0.00 99.90
Average: all 0.08 0.00 0.12 0.00 0.00 0.00 0.00
0.00 0.00 99.80
However, "sar" is also good:
root@umajor:/home/mgogala# sar -u 3 3
Linux 5.3.0-24-generic (umajor) 11/17/2019 _x86_64_ (16 CPU)
08:16:15 PM CPU %user %nice %system %iowait %steal
%idle
08:16:18 PM all 0.10 0.00 0.08 0.00 0.00 99.81
08:16:21 PM all 0.33 0.00 0.17 0.00 0.00 99.50
08:16:24 PM all 0.06 0.00 0.02 0.00 0.00 99.92
Average: all 0.17 0.00 0.09 0.00 0.00 99.74
Major advantage of sar over mpstat is the fact that you can go up to a
month back in time and investigate. If you're using Red Hat or a
derivative, he files are in /var/log/sa:
[root@ora122 sa]# sar -u -f sa24
Linux 4.14.35-1902.4.8.el7uek.x86_64 (ora122) 08/24/2019 _x86_64_
(3 CPU)
10:53:20 AM LINUX RESTART
11:00:01 AM CPU %user %nice %system %iowait %steal
%idle
11:10:01 AM all 3.26 0.00 0.41 0.65 0.00 95.69
11:20:01 AM all 0.85 0.00 0.26 0.17 0.00 98.72
11:30:01 AM all 0.27 0.22 1.17 0.16 0.00 98.19
11:40:01 AM all 0.21 0.00 0.20 0.08 0.00 99.51
11:50:01 AM all 0.80 0.00 0.22 0.11 0.00 98.87
Average: all 1.08 0.04 0.45 0.23 0.00 98.20
12:34:10 PM LINUX RESTART
12:40:01 PM CPU %user %nice %system %iowait %steal
%idle
12:50:01 PM all 0.24 0.00 0.16 0.09 0.00 99.51
01:00:01 PM all 2.31 0.00 0.34 0.22 0.00 97.12
Average: all 1.27 0.00 0.25 0.15 0.00 98.32
[root@ora122 sa]#
Snapshots are taken 10 minutes apart and usually go up to a month back.
I am using virtual machines which can be inactive for several days.
Other than that, you can see CPU Wait event if you are using
DBMS_RESOURCE_MANAGER. However, if uptime and other tools, show that the
CPU is not much used, then don't pay any attention to that event.
Regards
On 11/17/19 7:53 PM, Jack van Zanen wrote:
Hi All
I am seeing on occasion that CPU WAIT in OEM graph gets very prominent
According to "https://blog.orapub.com/20150210/what-is-light-green-oracle-database-cpu-wait-time.html";
The CPU WAIT is when Oracle is waiting to get on CPU (in the queue)
I would however expect to see CPU pressure when this happens
however on OS CPU have lots of IDLE time
[oracle@cltsadm01vm01 AWR]$ vmstat 10 10
procs -----------memory---------- ---swap-- -----io---- --system-- -----cpu-----
r b swpd free buff cache si so bi bo in cs us sy id wa st
13 0 0 91456296 1246408 72531584 0 0 8 32 0 0 22 4 74 0 0
9 0 0 91441104 1246408 72532288 0 0 0 188 50024 41230 28 3 69 0 0
13 0 0 91481680 1246408 72533216 0 0 1 266 48898 41615 28 3 69 0 0
9 0 0 91487496 1246408 72536520 0 0 0 242 60376 46185 26 3 71 0 0
10 0 0 91501456 1246408 72537696 0 0 1 196 45629 37927 24 5 71 0 0
12 0 0 91464880 1246412 72541536 0 0 0 252 46247 39331 29 4 67 0 0
Jack van Zanen
-------------------------
This e-mail and any attachments may contain confidential material for the sole use of the intended recipient. If you are not the intended recipient, please be aware that any disclosure, copying, distribution or use of this e-mail or any attachment is prohibited. If you have received this e-mail in error, please contact the sender and delete all copies.
Thank you for your cooperation
