allow prometheus-exporter to export metrics on workload level

[hebestreit]

Overview

With this change it's possible to set extra environment variables for the prometheus-exporter. This is desired to set ENABLE_WORKLOAD_METRICS=true. Additionally I've updated the image tag to the latest version.

Additional Information

The logic has been implemented in this PR:
kubescape/prometheus-exporter#24

Don't be confused by the version reduction associated with CI creating negative versions:
kubescape/prometheus-exporter#26

The Docker tag v0.2.0 has been created manually to avoid the above problem.
https://quay.io/repository/kubescape/prometheus-exporter?tab=tags

How to Test

Create a values.yaml with minimal config:

capabilities:
  prometheusExporter: enable
prometheusExporter:
  env:
    - name: ENABLE_WORKLOAD_METRICS
      value: "true"

Render the helm template and and check the occurrence:

helm template kubescape-operator kubescape-operator -f values.yaml | grep -5 "ENABLE_WORKLOAD_METRICS"
                  resource: limits.cpu
            - name: KS_LOGGER_LEVEL
              value: "info"
            - name: KS_LOGGER_NAME
              value: "zap"
            - name: ENABLE_WORKLOAD_METRICS
              value: "true"
          volumeMounts:
            - name: ks-cloud-config
              mountPath: /etc/config
              readOnly: true

[matthyx]

please also enable the new feature in snapshot_test.yaml ("all capabilities" scenario) and update unittests

[matthyx]

I would prefer to have a variable enableWorkloadMetrics that would add ENABLE_WORKLOAD_METRICS in the deployment

[hebestreit]

Sounds good to me too. Please take a look and let me know if I've missed anything.

[matthyx]

awesome, thanks @hebestreit !
I'm not sure we'll release before Christmas, but definitely in the beginning of 2025