grafana/k8s-monitoring-helm k8s-monitoring-4.1.1

on GitHub

What's changed

• Fix the pre-delete and post-install Alloy-finalizer hook templates failing to render when alloy-operator.waitForAlloyRemoval.securityContext is set to null (#2569) (@petewall)
• Update Alloy Operator to 0.5.7, Beyla to 1.16.6, and k8s-manifest-tail to 0.1.4 (@petewall)
• Fix extraLogProcessingStages (and extraLogProcessingRules on podLogsObjects) failing to render under older Helm versions (e.g. those bundled with the Terraform Helm provider) with cannot retrieve Template.Basepath from values inside tpl function, by propagating Template through to the feature module includes that call tpl (@petewall)
• Set opencost.opencost.exporter.cloudProviderApiKey to "disabled" so OpenCost does not try to write /var/configs/gcp.json on GCP/GKE and panic. Removes the placeholder GCP API key (and its trufflehog:ignore workarounds) from the chart defaults (@petewall)
• Add the ability to set scrapeNativeHistograms for the Prometheus Operator Objects feature (@lukas-unity)* Fix OpenCost crashing on GKE: point CONFIG_PATH at /tmp so the GCP provider can write its gcp.json (the default /var/configs is not writable by the non-root container user, which caused a nil-pointer panic). Set CLOUD_PROVIDER_API_KEY=disabled so the GCP provider can be constructed without a real billing API key, and remove the placeholder API key (and its trufflehog:ignore workarounds) from the chart defaults (@petewall)

Full changelog: CHANGELOG.md