diff --git a/infrastructure/monitoring/dashboards/cluster-cpu-check.json b/infrastructure/monitoring/dashboards/cluster-cpu-check.json
new file mode 100644
index 0000000000000000000000000000000000000000..85dd81d8813e3f133f0e55d4165317bb6ddc0d19
--- /dev/null
+++ b/infrastructure/monitoring/dashboards/cluster-cpu-check.json
@@ -0,0 +1,582 @@
+{
+ "__inputs": [
+ {
+ "name": "DS_PROMETHEUS",
+ "label": "Prometheus",
+ "description": "",
+ "type": "datasource",
+ "pluginId": "prometheus",
+ "pluginName": "Prometheus"
+ }
+ ],
+ "__elements": {},
+ "__requires": [
+ {
+ "type": "panel",
+ "id": "gauge",
+ "name": "Gauge",
+ "version": ""
+ },
+ {
+ "type": "grafana",
+ "id": "grafana",
+ "name": "Grafana",
+ "version": "10.1.2"
+ },
+ {
+ "type": "datasource",
+ "id": "prometheus",
+ "name": "Prometheus",
+ "version": "1.0.0"
+ },
+ {
+ "type": "panel",
+ "id": "timeseries",
+ "name": "Time series",
+ "version": ""
+ }
+ ],
+ "annotations": {
+ "list": [
+ {
+ "builtIn": 1,
+ "datasource": {
+ "type": "grafana",
+ "uid": "-- Grafana --"
+ },
+ "enable": true,
+ "hide": true,
+ "iconColor": "rgba(0, 211, 255, 1)",
+ "name": "Annotations & Alerts",
+ "type": "dashboard"
+ }
+ ]
+ },
+ "editable": true,
+ "fiscalYearStartMonth": 0,
+ "graphTooltip": 1,
+ "id": null,
+ "links": [],
+ "liveNow": false,
+ "panels": [
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "fieldConfig": {
+ "defaults": {
+ "color": {
+ "mode": "thresholds"
+ },
+ "mappings": [],
+ "thresholds": {
+ "mode": "absolute",
+ "steps": [
+ {
+ "color": "green",
+ "value": null
+ },
+ {
+ "color": "semi-dark-yellow",
+ "value": 15
+ },
+ {
+ "color": "semi-dark-red",
+ "value": 25
+ }
+ ]
+ },
+ "unit": "watt"
+ },
+ "overrides": []
+ },
+ "gridPos": {
+ "h": 10,
+ "w": 7,
+ "x": 0,
+ "y": 0
+ },
+ "id": 4,
+ "options": {
+ "orientation": "auto",
+ "reduceOptions": {
+ "calcs": [
+ "lastNotNull"
+ ],
+ "fields": "",
+ "values": false
+ },
+ "showThresholdLabels": false,
+ "showThresholdMarkers": true
+ },
+ "pluginVersion": "10.1.2",
+ "targets": [
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "editorMode": "code",
+ "expr": "avg(avg_over_time(node_hwmon_power_average_watt[$__interval]))",
+ "instant": false,
+ "legendFormat": "__auto",
+ "range": true,
+ "refId": "A"
+ }
+ ],
+ "title": "Average per Machine Power Consumption",
+ "type": "gauge"
+ },
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "fieldConfig": {
+ "defaults": {
+ "color": {
+ "mode": "palette-classic"
+ },
+ "custom": {
+ "axisCenteredZero": false,
+ "axisColorMode": "text",
+ "axisLabel": "",
+ "axisPlacement": "auto",
+ "barAlignment": 0,
+ "drawStyle": "line",
+ "fillOpacity": 0,
+ "gradientMode": "none",
+ "hideFrom": {
+ "legend": false,
+ "tooltip": false,
+ "viz": false
+ },
+ "insertNulls": false,
+ "lineInterpolation": "smooth",
+ "lineStyle": {
+ "fill": "solid"
+ },
+ "lineWidth": 1,
+ "pointSize": 5,
+ "scaleDistribution": {
+ "type": "linear"
+ },
+ "showPoints": "auto",
+ "spanNulls": false,
+ "stacking": {
+ "group": "A",
+ "mode": "none"
+ },
+ "thresholdsStyle": {
+ "mode": "off"
+ }
+ },
+ "mappings": [],
+ "thresholds": {
+ "mode": "absolute",
+ "steps": [
+ {
+ "color": "green",
+ "value": null
+ },
+ {
+ "color": "red",
+ "value": 80
+ }
+ ]
+ },
+ "unit": "watt"
+ },
+ "overrides": []
+ },
+ "gridPos": {
+ "h": 10,
+ "w": 17,
+ "x": 7,
+ "y": 0
+ },
+ "id": 1,
+ "options": {
+ "legend": {
+ "calcs": [],
+ "displayMode": "list",
+ "placement": "bottom",
+ "showLegend": true
+ },
+ "tooltip": {
+ "mode": "single",
+ "sort": "none"
+ }
+ },
+ "targets": [
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "editorMode": "code",
+ "expr": "max by (node) (label_replace(max(node_hwmon_power_average_watt) by (instance), \"internal_ip\", \"$1\", \"instance\", \"(.*):9100\") * on (internal_ip) group_left(node) kube_node_info)",
+ "hide": true,
+ "instant": false,
+ "legendFormat": "__auto",
+ "range": true,
+ "refId": "A"
+ },
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "editorMode": "code",
+ "exemplar": false,
+ "expr": "avg(avg_over_time(node_hwmon_power_average_watt[$__rate_interval]))",
+ "hide": false,
+ "instant": false,
+ "interval": "60",
+ "legendFormat": "Average",
+ "range": true,
+ "refId": "B"
+ }
+ ],
+ "title": "Avg per Node Watt",
+ "type": "timeseries"
+ },
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "description": "",
+ "fieldConfig": {
+ "defaults": {
+ "color": {
+ "mode": "palette-classic"
+ },
+ "custom": {
+ "axisCenteredZero": false,
+ "axisColorMode": "text",
+ "axisLabel": "",
+ "axisPlacement": "auto",
+ "barAlignment": 0,
+ "drawStyle": "line",
+ "fillOpacity": 0,
+ "gradientMode": "none",
+ "hideFrom": {
+ "legend": false,
+ "tooltip": false,
+ "viz": false
+ },
+ "insertNulls": false,
+ "lineInterpolation": "smooth",
+ "lineWidth": 1,
+ "pointSize": 5,
+ "scaleDistribution": {
+ "type": "linear"
+ },
+ "showPoints": "auto",
+ "spanNulls": false,
+ "stacking": {
+ "group": "A",
+ "mode": "none"
+ },
+ "thresholdsStyle": {
+ "mode": "off"
+ }
+ },
+ "mappings": [],
+ "thresholds": {
+ "mode": "absolute",
+ "steps": [
+ {
+ "color": "green",
+ "value": null
+ },
+ {
+ "color": "red",
+ "value": 80
+ }
+ ]
+ },
+ "unit": "celsius"
+ },
+ "overrides": [
+ {
+ "__systemRef": "hideSeriesFrom",
+ "matcher": {
+ "id": "byNames",
+ "options": {
+ "mode": "exclude",
+ "names": [
+ "Average"
+ ],
+ "prefix": "All except:",
+ "readOnly": true
+ }
+ },
+ "properties": [
+ {
+ "id": "custom.hideFrom",
+ "value": {
+ "legend": false,
+ "tooltip": false,
+ "viz": true
+ }
+ }
+ ]
+ }
+ ]
+ },
+ "gridPos": {
+ "h": 13,
+ "w": 24,
+ "x": 0,
+ "y": 10
+ },
+ "id": 2,
+ "options": {
+ "legend": {
+ "calcs": [],
+ "displayMode": "list",
+ "placement": "bottom",
+ "showLegend": true
+ },
+ "tooltip": {
+ "mode": "single",
+ "sort": "none"
+ }
+ },
+ "targets": [
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "editorMode": "code",
+ "expr": "avg(node_hwmon_temp_celsius{chip=\"0000:00:08_1_0000:04:00_0\"})",
+ "instant": false,
+ "legendFormat": "Average",
+ "range": true,
+ "refId": "A"
+ },
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "editorMode": "code",
+ "expr": "max by (node) (label_replace(max(node_hwmon_temp_celsius{chip=\"0000:00:08_1_0000:04:00_0\"}) by (instance), \"internal_ip\", \"$1\", \"instance\", \"(.*):9100\") * on (internal_ip) group_left(node) kube_node_info)",
+ "hide": false,
+ "instant": false,
+ "legendFormat": "__auto",
+ "range": true,
+ "refId": "B"
+ }
+ ],
+ "title": "Avg Cluster Temperature",
+ "type": "timeseries"
+ },
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "fieldConfig": {
+ "defaults": {
+ "color": {
+ "mode": "thresholds"
+ },
+ "mappings": [],
+ "max": 40,
+ "thresholds": {
+ "mode": "absolute",
+ "steps": [
+ {
+ "color": "green",
+ "value": null
+ },
+ {
+ "color": "#EAB839",
+ "value": 18
+ },
+ {
+ "color": "red",
+ "value": 36
+ }
+ ]
+ }
+ },
+ "overrides": []
+ },
+ "gridPos": {
+ "h": 8,
+ "w": 6,
+ "x": 0,
+ "y": 23
+ },
+ "id": 6,
+ "options": {
+ "orientation": "auto",
+ "reduceOptions": {
+ "calcs": [
+ "lastNotNull"
+ ],
+ "fields": "",
+ "values": false
+ },
+ "showThresholdLabels": false,
+ "showThresholdMarkers": true
+ },
+ "pluginVersion": "10.1.2",
+ "targets": [
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "editorMode": "code",
+ "exemplar": false,
+ "expr": "sum(avg_over_time(node_load15[$__range]))",
+ "instant": false,
+ "legendFormat": "__auto",
+ "range": true,
+ "refId": "A"
+ }
+ ],
+ "title": "Cluster Load15",
+ "type": "gauge"
+ },
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "fieldConfig": {
+ "defaults": {
+ "color": {
+ "mode": "palette-classic"
+ },
+ "custom": {
+ "axisCenteredZero": false,
+ "axisColorMode": "text",
+ "axisLabel": "",
+ "axisPlacement": "auto",
+ "barAlignment": 0,
+ "drawStyle": "line",
+ "fillOpacity": 0,
+ "gradientMode": "none",
+ "hideFrom": {
+ "legend": false,
+ "tooltip": false,
+ "viz": false
+ },
+ "insertNulls": false,
+ "lineInterpolation": "linear",
+ "lineWidth": 1,
+ "pointSize": 5,
+ "scaleDistribution": {
+ "type": "linear"
+ },
+ "showPoints": "auto",
+ "spanNulls": false,
+ "stacking": {
+ "group": "A",
+ "mode": "none"
+ },
+ "thresholdsStyle": {
+ "mode": "off"
+ }
+ },
+ "mappings": [],
+ "thresholds": {
+ "mode": "absolute",
+ "steps": [
+ {
+ "color": "green",
+ "value": null
+ },
+ {
+ "color": "red",
+ "value": 80
+ }
+ ]
+ }
+ },
+ "overrides": []
+ },
+ "gridPos": {
+ "h": 8,
+ "w": 18,
+ "x": 6,
+ "y": 23
+ },
+ "id": 5,
+ "options": {
+ "legend": {
+ "calcs": [],
+ "displayMode": "list",
+ "placement": "bottom",
+ "showLegend": true
+ },
+ "tooltip": {
+ "mode": "single",
+ "sort": "none"
+ }
+ },
+ "targets": [
+ {
+ "datasource": {
+ "type": "prometheus",
+ "uid": "${DS_PROMETHEUS}"
+ },
+ "editorMode": "code",
+ "expr": "avg(node_load15)",
+ "instant": false,
+ "legendFormat": "__auto",
+ "range": true,
+ "refId": "A"
+ }
+ ],
+ "title": "Avg LoadAvg15",
+ "type": "timeseries"
+ }
+ ],
+ "refresh": false,
+ "schemaVersion": 38,
+ "style": "dark",
+ "tags": ["node-exporter"],
+ "templating": {
+ "list": [
+ {
+ "current": {
+ "selected": true,
+ "text": "Prometheus",
+ "value": "Prometheus"
+ },
+ "error": null,
+ "hide": 0,
+ "includeAll": false,
+ "label": "Datasource",
+ "multi": false,
+ "name": "DS_PROMETHEUS",
+ "options": [],
+ "query": "prometheus",
+ "queryValue": "",
+ "refresh": 1,
+ "regex": "",
+ "skipUrlSync": false,
+ "type": "datasource"
+ }
+ ]
+ },
+ "time": {
+ "from": "now-7d",
+ "to": "now"
+ },
+ "timepicker": {},
+ "timezone": "",
+ "title": "Cluster CPU check",
+ "uid": "ab765508-9634-48d4-93b8-bd6f2e24ac05",
+ "version": 5,
+ "weekStart": ""
+ }
\ No newline at end of file
diff --git a/infrastructure/monitoring/kustomization.yaml b/infrastructure/monitoring/kustomization.yaml
index 23a6e8c3bc3040bac594754a319bd3147817bef8..369f64a2cf1ed925172bc900171879604151acfa 100644
--- a/infrastructure/monitoring/kustomization.yaml
+++ b/infrastructure/monitoring/kustomization.yaml
@@ -15,6 +15,7 @@ patchesStrategicMerge:
configMapGenerator:
- name: monitoring-grafana-dashboards
files:
+ - ./dashboards/cluster-cpu-check.json
- ./dashboards/kubernetes-persistent-volumes-cluster.json
- ./dashboards/sloth-details.json
- ./dashboards/sloth-high-level.json