FIX(opentelemetry-operator): remove cpu null values

- Remove cpu: null (not allowed in new chart schema)
- Keep only memory limits

goldilocks/helm-values.yaml

@@ -60,15 +60,6 @@ controller:
   # Set to false to only monitor namespaces with the label: goldilocks.fairwinds.com/enabled=true
   enableCostRecommendations: true
 
-  # Schedule on control-plane node
-  nodeSelector:
-    node-role.kubernetes.io/control-plane: "true"
-
-  tolerations:
-    - key: node-role.kubernetes.io/control-plane
-      operator: Exists
-      effect: NoSchedule
-
 # VPA configuration (should already be installed)
 vpa:
   # Set to false since we're installing VPA separately

grafana/dashboards/main.json

@@ -1987,14 +1987,14 @@
           },
           "editorMode": "code",
           "exemplar": true,
-          "expr": "sum(rate(container_cpu_cfs_throttled_seconds_total{image!=\"\", cluster=\"$cluster\"}[$__rate_interval])) by (namespace) > 0 or vector(0)",
+          "expr": "sum(rate(container_pressure_cpu_waiting_seconds_total{pod!=\"\", cluster=\"$cluster\"}[$__rate_interval])) by (namespace) > 0.1",
           "interval": "$resolution",
           "legendFormat": "{{ namespace }}",
           "range": true,
           "refId": "A"
         }
       ],
-      "title": "CPU Throttled seconds by namespace",
+      "title": "CPU Pressure (waiting) by namespace",
       "type": "timeseries"
     },
     {
@@ -2099,14 +2099,14 @@
           },
           "editorMode": "code",
           "exemplar": true,
-          "expr": "sum(rate(node_cpu_core_throttles_total{cluster=\"$cluster\", job=\"$job\"}[$__rate_interval])) by (instance) or vector(0)",
+          "expr": "sum(rate(node_pressure_cpu_waiting_seconds_total{cluster=\"$cluster\", job=\"$job\"}[$__rate_interval])) by (instance) > 0.1",
           "interval": "$resolution",
           "legendFormat": "{{ instance }}",
           "range": true,
           "refId": "A"
         }
       ],
-      "title": "CPU Core Throttled by instance",
+      "title": "Node CPU Pressure (waiting) by instance",
       "type": "timeseries"
     },
     {

grafana/dashboards/minio.json

@@ -2242,8 +2242,9 @@
               }
             ]
           },
-          "unit": "s",
+          "unit": "percent",
-          "unitScale": true
+          "min": 0,
+          "max": 100
         },
         "overrides": []
       },
@@ -2278,7 +2279,7 @@
           },
           "editorMode": "code",
           "exemplar": true,
-          "expr": "rate(minio_node_process_cpu_total_seconds{job=~\"$scrape_jobs\"}[5m])",
+          "expr": "rate(minio_node_process_cpu_total_seconds{job=\"minio\"}[5m]) * 100",
           "interval": "",
           "legendFormat": "{{server}}",
           "range": true,

kube-state-metrics/helm-values.yaml

@@ -14,15 +14,6 @@ resources:
   limits:
     memory: 105Mi
 
-# Schedule on control-plane node
-nodeSelector:
-  node-role.kubernetes.io/control-plane: "true"
-
-tolerations:
-  - key: node-role.kubernetes.io/control-plane
-    operator: Exists
-    effect: NoSchedule
-
 service:
   type: ClusterIP
   clusterIP: None

opentelemetry-collector/helm-values.yaml

@@ -32,14 +32,6 @@ resources:
   limits:
     memory: 512Mi
 
-# =============================================================================
-# Tolerations (run on all nodes including master)
-# =============================================================================
-tolerations:
-  - key: node-role.kubernetes.io/control-plane
-    operator: Exists
-    effect: NoSchedule
-
 # =============================================================================
 # Extra Volumes for Log Collection
 # =============================================================================

opentelemetry-collector/manifests/collector.yaml

@@ -27,11 +27,6 @@ spec:
     limits:
       memory: 512Mi
 
-  tolerations:
-    - key: node-role.kubernetes.io/control-plane
-      operator: Exists
-      effect: NoSchedule
-
   volumeMounts:
     - name: varlogpods
       mountPath: /var/log/pods

opentelemetry-operator/argocd.yaml

@@ -12,7 +12,7 @@ spec:
   sources:
   - repoURL: https://open-telemetry.github.io/opentelemetry-helm-charts
     chart: opentelemetry-operator
-    targetRevision: 0.74.0
+    targetRevision: 0.102.0
     helm:
       valueFiles:
       - $values/opentelemetry-operator/helm-values.yaml

opentelemetry-operator/helm-values.yaml

@@ -20,7 +20,6 @@ manager:
       repository: ghcr.io/open-telemetry/opentelemetry-operator/autoinstrumentation-go
   resources:
     limits:
-      cpu: null  # Disable chart default CPU limits
       memory: 256Mi
     requests:
       cpu: 10m
@@ -39,17 +38,8 @@ kubeRBACProxy:
   enabled: true
   resources:
     limits:
-      cpu: null  # Disable chart default CPU limits
       memory: 64Mi
     requests:
       cpu: 5m
       memory: 64Mi
 
-# Schedule on master node
-tolerations:
-  - key: node-role.kubernetes.io/control-plane
-    operator: Exists
-    effect: NoSchedule
-
-nodeSelector:
-  node-role.kubernetes.io/control-plane: "true"

promtail/helm-values.yaml

@@ -37,12 +37,6 @@ resources:
   limits:
     memory: 182Mi
 
-# Tolerations to run on all nodes including control-plane
-tolerations:
-  - key: node-role.kubernetes.io/control-plane
-    operator: Exists
-    effect: NoSchedule
-
 # ServiceMonitor disabled
 serviceMonitor:
   enabled: false

vpa/helm-values.yaml

@@ -13,15 +13,6 @@ recommender:
     limits:
       memory: 100Mi
 
-  # Schedule on control-plane node
-  nodeSelector:
-    node-role.kubernetes.io/control-plane: "true"
-
-  tolerations:
-    - key: node-role.kubernetes.io/control-plane
-      operator: Exists
-      effect: NoSchedule
-
 # Updater - applies recommended resource requests to pods
 # Disabled because we're using updateMode: Off (recommendations only)
 updater: