4.54 SmartCPULoad

Use this Knowledge Script to monitor the CPU load of Linux or UNIX machines. This Knowledge Script uses either the CPU utilization or the queue length or both to determine whether the CPU is overloaded.

If you select the queue length to determine the CPU load, the Knowledge Script raises an event when both the queue length and the CPU utilization exceeds its threshold. The threshold for queue length is calculated as the maximum queue length * number of CPU cores. For example, if you specify the queue length as 2 and you are using a 2 core machine, then the threshold for queue length becomes 4. In this case, if you specify the threshold for CPU utilization as 80% and select the Use queue length in determining CPU overload? parameter, then the Knowledge Script raises an event only when the queue length exceeds 4 and CPU utilization exceeds 80%.

This Knowledge Script is supported on all UNIX platforms.

4.54.1 Resource Objects

CPU folder

4.54.2 Default Schedule

The default interval for this script is Every 5 minutes.

4.54.3 Setting Parameter Values

Set the following parameters as needed:

Description

How to Set It

General Settings

Event severity when job fails

Set the event severity level, from 1 to 40, to indicate the importance of an event reported when a job fails. The default is 5.

Event detail format

Select the format in which to view the event detail. The default is HTML Table.

Enable debugging?

Select Yes to enable debugging. The default is unselected.

Raise event if AppManager fails to get metrics?

Select Yes to raise an event if AppManager fails to retrieve the metrics.The default is Yes.

Event severity

Set the event severity level, from 1 to 40, to indicate the importance of the event that is raised when AppManager fails to get the metrics. The default is 5.

Number of seconds between samples

Specify the number of seconds, 2 to 30, to wait between samples. The default is 5 seconds.

Number of times sar should iterate before reporting an average value

Specify the number of sar iterations, 1 to 100, to report the average value. The default is 1 iterations.

Number of processes to add in detail

Specify the number of processes, 1 to 90, to add in the detail. The default is 5.

Cap processor usage values at 100 percent?

Select Yes to cap the processor usage values at 100 percent. If unselected the processor usage values might exceed 100%. The default is unselected.

Event Settings

Use queue length in determining CPU overload?

Select Yes to use the queue length to determine CPU overload. The default is Yes.

Threshold -- Maximum queue length

Specify the threshold for the maximum queue length. The default is 2 multiplied by the processor capacity.

Raise event if total CPU utilization exceeds threshold?

Select Yes to raise an event if the total CPU utilization exceeds the threshold. The default is Yes.

Threshold -- Maximum CPU utilization

Specify the threshold for the maximum CPU utilization percentage. The default is 80%.

Severity

Set the event severity level, from 1 to 40, from 1 to 40, to indicate the importance of an event reported when the maximum CPU utilization percentage is detected. The default is 5.

Data Collection

 

Collect data for CPU utilization in percent?

Select Yes to collect data for charts and reports for the total CPU utilization percentage. The default is Yes.

Collect data for %User CPU state?

Select Yes to collect data for charts and reports for the total User CPU utilization percentage. The default is unselected.

Collect data for %System CPU state?

Select Yes to collect data for charts and reports for the total System CPU utilization percentage. The default is unselected.

Collect data for %Wait CPU State?

Select Yes to collect data for charts and reports for the total Wait CPU utilization percentage. The default is unselected.

Collect data for RunQueue length

Select Yes to collect data for charts and reports for the RunQueue length. The default is unselected.