@leverege/build-tools 2.99.1 → 2.100.0

package/src/dashboard/json/standard/gpu-autoscaling-dashboard-ui-import.json

@@ -0,0 +1,319 @@
+{
+  "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": 0,
+  "id": 12,
+  "links": [],
+  "panels": [
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic"
+          },
+          "custom": {
+            "axisBorderShow": false,
+            "axisCenteredZero": false,
+            "axisColorMode": "text",
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "barWidthFactor": 0.6,
+            "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"
+              },
+              {
+                "color": "yellow",
+                "value": 70
+              },
+              {
+                "color": "red",
+                "value": 90
+              }
+            ]
+          },
+          "unit": "percent"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 24,
+        "x": 0,
+        "y": 0
+      },
+      "id": 1,
+      "options": {
+        "legend": {
+          "calcs": [],
+          "displayMode": "table",
+          "placement": "bottom",
+          "showLegend": true
+        },
+        "tooltip": {
+          "hideZeros": false,
+          "mode": "single",
+          "sort": "none"
+        }
+      },
+      "pluginVersion": "12.0.0",
+      "targets": [
+        {
+          "expr": "gpu_duty_cycle_percent{namespace=\"default\"}",
+          "interval": "",
+          "legendFormat": "{{pod}}",
+          "refId": "A"
+        }
+      ],
+      "title": "GPU Duty Cycle by Pod",
+      "type": "timeseries"
+    },
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic"
+          },
+          "custom": {
+            "axisBorderShow": false,
+            "axisCenteredZero": false,
+            "axisColorMode": "text",
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "barWidthFactor": 0.6,
+            "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"
+              },
+              {
+                "color": "red",
+                "value": 80
+              }
+            ]
+          },
+          "unit": "none"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 7,
+        "w": 24,
+        "x": 0,
+        "y": 8
+      },
+      "id": 2,
+      "options": {
+        "legend": {
+          "calcs": [],
+          "displayMode": "table",
+          "placement": "bottom",
+          "showLegend": true
+        },
+        "tooltip": {
+          "hideZeros": false,
+          "mode": "multi",
+          "sort": "none"
+        }
+      },
+      "pluginVersion": "12.0.0",
+      "targets": [
+        {
+          "expr": "kube_horizontalpodautoscaler_status_desired_replicas{horizontalpodautoscaler=\"tpi-defect-detector-hpa\"}",
+          "legendFormat": "Desired Replicas",
+          "refId": "A"
+        },
+        {
+          "expr": "kube_horizontalpodautoscaler_status_current_replicas{horizontalpodautoscaler=\"tpi-defect-detector-hpa\"}",
+          "legendFormat": "Current Replicas",
+          "refId": "B"
+        }
+      ],
+      "title": "HPA Desired vs Current Replicas",
+      "type": "timeseries"
+    },
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic"
+          },
+          "custom": {
+            "axisBorderShow": false,
+            "axisCenteredZero": false,
+            "axisColorMode": "text",
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "barWidthFactor": 0.6,
+            "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"
+              },
+              {
+                "color": "red",
+                "value": 80
+              }
+            ]
+          },
+          "unit": "none"
+        },
+        "overrides": []
+      },
+      "gridPos": {
+        "h": 7,
+        "w": 24,
+        "x": 0,
+        "y": 15
+      },
+      "id": 3,
+      "options": {
+        "legend": {
+          "calcs": [],
+          "displayMode": "table",
+          "placement": "bottom",
+          "showLegend": true
+        },
+        "tooltip": {
+          "hideZeros": false,
+          "mode": "single",
+          "sort": "none"
+        }
+      },
+      "pluginVersion": "12.0.0",
+      "targets": [
+        {
+          "expr": "sum(kube_pod_status_ready{namespace=\"default\", condition=\"true\", pod=~\"tpi-defect-detector.*\"})",
+          "legendFormat": "Ready Pods",
+          "refId": "A"
+        }
+      ],
+      "title": "Ready Pods Matching Pattern",
+      "type": "timeseries"
+    }
+  ],
+  "preload": false,
+  "refresh": "1m",
+  "schemaVersion": 41,
+  "tags": [
+    "gpu",
+    "autoscaling",
+    "hpa"
+  ],
+  "templating": {
+    "list": []
+  },
+  "time": {
+    "from": "now-3h",
+    "to": "now"
+  },
+  "timepicker": {},
+  "timezone": "browser",
+  "title": "GPU Autoscaling - tpi-defect-detector",
+  "uid": "feh52q72w0b28a",
+  "version": 3
+}

package/src/dashboard/json/standard/k8s-api-latency.json

@@ -0,0 +1,166 @@
+{
+  "title": "Kubernetes API Server Latency",
+  "tags": ["kubernetes", "apiserver", "latency"],
+  "timezone": "browser",
+  "schemaVersion": 36,
+  "version": 15,
+  "refresh": "5m",
+  "time": {
+    "from": "now-6h",
+    "to": "now"
+  },
+  "timepicker": {
+    "refresh_intervals": ["30s", "1m", "5m", "15m", "30m", "1h"],
+    "time_options": ["5m", "15m", "1h", "6h", "12h", "24h"]
+  },
+  "templating": {
+    "list": [
+      {
+        "name": "datasource",
+        "type": "datasource",
+        "query": "prometheus",
+        "label": "Prometheus",
+        "refresh": 1
+      }
+    ]
+  },
+  "panels": [
+    {
+      "type": "bargauge",
+      "title": "Top 5 Latency Offenders (99th percentile)",
+      "gridPos": { "x": 0, "y": 0, "w": 12, "h": 8 },
+      "datasource": "$datasource",
+      "targets": [
+        {
+          "expr": "topk(5, histogram_quantile(0.99, sum(rate(apiserver_request_duration_seconds_bucket{verb!~\"(?:CONNECT|WATCHLIST|WATCH|PROXY)\"}[5m])) by (le, verb, resource)))",
+          "format": "time_series",
+          "refId": "A",
+          "instant": true,
+          "legendFormat": "{{verb}} {{resource}}"
+        }
+      ],
+      "fieldConfig": {
+        "defaults": {
+          "unit": "s",
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              { "color": "green", "value": null },
+              { "color": "yellow", "value": 1 },
+              { "color": "red", "value": 2 }
+            ]
+          }
+        },
+        "overrides": []
+      }
+    },
+    {
+      "type": "timeseries",
+      "title": "99th Percentile Latency by API Resource",
+      "gridPos": { "x": 12, "y": 0, "w": 12, "h": 8 },
+      "datasource": "$datasource",
+      "targets": [
+        {
+          "expr": "histogram_quantile(0.99, sum(rate(apiserver_request_duration_seconds_bucket{verb!~\"(?:CONNECT|WATCHLIST|WATCH|PROXY)\"}[5m])) by (le, verb, resource))",
+          "format": "time_series",
+          "refId": "B",
+          "legendFormat": "{{verb}} {{resource}}"
+        }
+      ],
+      "fieldConfig": {
+        "defaults": { "unit": "s" },
+        "overrides": []
+      }
+    },
+    {
+      "type": "timeseries",
+      "title": "POST pods Latency by Namespace (99th percentile)",
+      "gridPos": { "x": 0, "y": 8, "w": 24, "h": 8 },
+      "datasource": "$datasource",
+      "targets": [
+        {
+          "expr": "histogram_quantile(0.99, sum(rate(apiserver_request_duration_seconds_bucket{verb=\"POST\", resource=\"pods\"}[5m])) by (le, namespace))",
+          "format": "time_series",
+          "refId": "E",
+          "legendFormat": "{{namespace}}"
+        }
+      ],
+      "fieldConfig": {
+        "defaults": {
+          "unit": "s"
+        },
+        "overrides": []
+      }
+    },
+    {
+      "type": "timeseries",
+      "title": "Admission Controller Latency (99th percentile) + Count",
+      "gridPos": { "x": 0, "y": 16, "w": 24, "h": 8 },
+      "datasource": "$datasource",
+      "targets": [
+        {
+          "expr": "histogram_quantile(0.99, sum(rate(apiserver_admission_controller_admission_duration_seconds_bucket[5m])) by (le, name))",
+          "format": "time_series",
+          "refId": "F",
+          "legendFormat": "Latency - {{name}}"
+        },
+        {
+          "expr": "sum(rate(apiserver_admission_controller_admission_duration_seconds_count[5m])) by (name)",
+          "format": "time_series",
+          "refId": "G",
+          "legendFormat": "Count - {{name}}"
+        }
+      ],
+      "fieldConfig": {
+        "defaults": {
+          "unit": "s"
+        },
+        "overrides": []
+      },
+      "options": {
+        "tooltip": {
+          "mode": "multi",
+          "sort": "desc"
+        },
+        "legend": {
+          "displayMode": "table",
+          "placement": "bottom"
+        }
+      }
+    },
+    {
+      "type": "table",
+      "title": "API Server Request Volume (by verb/resource)",
+      "gridPos": { "x": 0, "y": 24, "w": 24, "h": 8 },
+      "datasource": "$datasource",
+      "targets": [
+        {
+          "expr": "sum(rate(apiserver_request_total{verb!~\"(?:CONNECT|WATCHLIST|WATCH|PROXY)\"}[5m])) by (verb, resource)",
+          "format": "table",
+          "refId": "C"
+        }
+      ],
+      "fieldConfig": {
+        "defaults": { "unit": "ops" },
+        "overrides": []
+      }
+    },
+    {
+      "type": "table",
+      "title": "Active API Latency Alerts",
+      "gridPos": { "x": 0, "y": 32, "w": 24, "h": 6 },
+      "datasource": "$datasource",
+      "targets": [
+        {
+          "expr": "ALERTS{alertname=~\"KubernetesApiServerLatency.*\", alertstate=\"firing\"}",
+          "format": "table",
+          "refId": "D"
+        }
+      ],
+      "fieldConfig": {
+        "defaults": {},
+        "overrides": []
+      }
+    }
+  ]
+}