Getting Zero values for CPU_Utlization metrics for IBM-i

[varun-wadhwani]

Hello,

I have configured the prometheus-exporter-jdbc on the IBM-i systems and I am able to pull-off the data on Grafana dashboards as wel. I am able to get the data for all the metrics but not for the following,

1. AVERAGE_CPU_RATE
2. AVERAGE_CPU_UTILIZATION
3. AVERAGE_CPU_RATE_REAL
4. AVERAGE_CPU_UTILIZATION_REAL
5. MINIMUM_CPU_UTILIZATION
6. MAXIMUM_CPU_UTILIZATION
7. SQL_CPU_UTILIZATION

Though, we are able to get values when we are running the SQL commands locally on the IBM-i machines.

Following are some screenshots of Prometheus metric data that we are receiving from the IBM-i machines,

Could you please help us in this regard?

Thanks
Varun Wadhwani

[mcouillard]

We ran into something similar recently. Likely relates to the View change IBM did via PTFs:

• QSYS2.SYSTEM_STATUS_INFO

• from SYSTEM_STATUS_INFO SYSTEM_STATUS_INFO.AVERAGE_CPU_UTILIZATION

• to SYSTEM_ACTIVITY_INFO SYSTEM_ACTIVITY_INFO.AVERAGE_CPU_UTILIZATION

Should be resolvable by updating the queries ran.

[ThePrez]

With the latest version, check the metrics with SYSACT in the name