Observability

  • Changed

    • Upgrade grafana chart: 11.3.6 => 11.3.7
    • Upgrade pg-cluster-recovery-test subchart: v0.4.1 => v0.5.0

    Fixed

    • Make sure crossplane cannot delete Crossplane azure storage accounts/containers and s3 buckets.
  • Fixed

    • Make sure crossplane cannot delete Crossplane azure storage accounts/containers and s3 buckets.
  • Fixed

    • Make sure crossplane cannot delete Crossplane azure storage accounts/containers and s3 buckets.
  • Changed

    • Upgrade Tempo Vulture chart from 0.12.7 to 0.12.9
    • Upgrade Tempo chart from to 2.6.2 to 2.14.3
      • Upgrades Tempo from 2.10.2 to 2.10.4

    Fixed

    • Make sure crossplane cannot delete Crossplane azure storage accounts/containers and s3 buckets.
  • Changed

    • Update envoy-gateway-loadtesting dashboard with added kong graphs
  • Fixed

    • Disabled per-component serviceaccounts so we rely on a central serviceaccount that has proper IRSA annotation.
  • Changed

    • Updated Loki Helm chart dependency to use grafana-community/helm-charts repository.
    • Upgrade upstream loki helm chart from v6 (6.53.0) to v11 (11.6.4), Loki app version 3.6.5 → 3.7.1.
    • Remove custom ScaledObject wrapper templates (backend, gateway, read, write); upstream v11 ships native ScaledObject support, keeping both caused duplicate resources.
    • Remove kedaAutoscaling value blocks consumed only by the deleted wrapper templates.
    • Remove selfMonitoring dead code; removed from upstream prior to v11.
    • Remove redundant podSecurityContext and containerSecurityContext overrides for loki and gateway components; they are now identical to upstream v11 defaults.
    • Rename sample_configs/ directory to examples/.
  • Added

    • Add envoy-gateway-loadtesting dashboard

    Changed

    • Reorganize Grafana Cloud directories: dashboards/grafana-cloud/sources/, grafana-cloud/grafana-cloud/backup/, GC API scripts → grafana-cloud/scripts/
    • Update backup workflow and README to reflect new paths
  • Added

    • Add monthly GitHub Actions workflow to auto-update Tempo, Mimir, Loki, and Alloy mixin dashboards
    • Add Monitoring Landscape / Customer Audit dashboard to audit monitoring tools across workload clusters and compare resource consumption with the GiantSwarm observability platform

    Changed

    • Move Tempo dashboards from private_dashboards_mz to team_atlas under Giant Swarm/Observability/Tempo
    • Move Loki dashboards from private_dashboards_al to team_atlas under Giant Swarm/Observability/Loki
    • Refresh Loki dashboards from latest upstream mixin
    • Move Mimir dashboards from private_dashboards_mz to team_atlas under Giant Swarm/Observability/Mimir
    • Refresh Mimir mixin dashboards from upstream mimir-2.17.6
    • Move Alloy dashboards from private_dashboards_al to team_atlas under Giant Swarm/Observability/Alloy
    • Refresh Alloy mixin dashboards from upstream v1.15.0
      • Adds alloy-loki and alloy-otel-engine-overview dashboards
  • Fixed

    • Update recovery-test AWS IAM to support unique names