Developer story

As a Developer, I want to remove the deprecated metric rules in windows_exporter metrics so that I can benefit from the improved and more granular collectors for CPU, memory, and OS information, ensuring better observability of my Windows nodes.

Description

The cs collector in windows_exporter has been deprecated and replaced by the more specific cpu, memory, and os collectors. This change requires updating our Prometheus recording rules to use the new metrics provided by these collectors. The goal is to align with the upstream windows_exporter changes, ensuring continued support and accuracy of Windows node monitoring.

Specifically, the windows_cs_physical_memory_bytes metric, which is currently used to record node_memory_MemTotal_bytes, needs to be replaced.

The cs collector is deprecated and will be removed in a future release. Logical processors has been moved to cpu_info
                     collector. Physical memory has been moved to memory collector. Hostname has been moved to os collector."'

Engineering Details

The required changes are outlined in the upstream windows_exporter pull request:

PR #1596: Deprecate cs collector https://github.com/prometheus-community/windows_exporter/pull/1596

This story involves updating the Prometheus rules in config/windows-exporter/prometheusRule.yaml.

https://github.com/openshift/windows-machine-config-operator/blob/d9a739209fe4566ec6cedd5a8e62c2bba27ca570/config/windows-exporter/prometheusRule.yaml#L37

Enhancements

https://github.com/openshift/enhancements/blob/6d1d50226273ff6790760369805078c0fb3570ab/enhancements/windows-containers/monitoring-windows-nodes.md?plain=1#L163 

Acceptance Criteria

The prometheusRule.yaml for the windows-exporter is updated to use metrics from the new cpu, memory, and os collectors instead of the deprecated cs collector.
The recording rule for node_memory_MemTotal_bytes is updated to use a metric from the new collectors.
All existing recording rules for Windows nodes continue to function correctly after the changes.
The changes are validated in a cluster environment.