specweave 0.18.0 → 0.20.0

package/plugins/specweave-kafka/monitoring/grafana/dashboards/kafka-cluster-overview.json

@@ -0,0 +1,564 @@
+{
+  "annotations": {
+    "list": [
+      {
+        "builtIn": 1,
+        "datasource": "-- Grafana --",
+        "enable": true,
+        "hide": true,
+        "iconColor": "rgba(0, 211, 255, 1)",
+        "name": "Annotations & Alerts",
+        "type": "dashboard"
+      }
+    ]
+  },
+  "editable": true,
+  "gnetId": null,
+  "graphTooltip": 1,
+  "id": null,
+  "links": [],
+  "panels": [
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "thresholds"
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "red",
+                "value": null
+              },
+              {
+                "color": "green",
+                "value": 1
+              },
+              {
+                "color": "red",
+                "value": 2
+              }
+            ]
+          },
+          "unit": "short"
+        }
+      },
+      "gridPos": {
+        "h": 4,
+        "w": 4,
+        "x": 0,
+        "y": 0
+      },
+      "id": 1,
+      "options": {
+        "colorMode": "background",
+        "graphMode": "none",
+        "justifyMode": "auto",
+        "orientation": "auto",
+        "reduceOptions": {
+          "calcs": ["lastNotNull"],
+          "fields": "",
+          "values": false
+        },
+        "textMode": "auto"
+      },
+      "pluginVersion": "9.0.0",
+      "targets": [
+        {
+          "expr": "kafka_controller_active_controller_count",
+          "refId": "A"
+        }
+      ],
+      "title": "Active Controller",
+      "type": "stat",
+      "description": "Number of active controllers (should be exactly 1)"
+    },
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "thresholds"
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              },
+              {
+                "color": "yellow",
+                "value": 1
+              },
+              {
+                "color": "red",
+                "value": 10
+              }
+            ]
+          },
+          "unit": "short"
+        }
+      },
+      "gridPos": {
+        "h": 4,
+        "w": 4,
+        "x": 4,
+        "y": 0
+      },
+      "id": 2,
+      "options": {
+        "colorMode": "background",
+        "graphMode": "area",
+        "justifyMode": "auto",
+        "orientation": "auto",
+        "reduceOptions": {
+          "calcs": ["lastNotNull"],
+          "fields": "",
+          "values": false
+        },
+        "textMode": "auto"
+      },
+      "pluginVersion": "9.0.0",
+      "targets": [
+        {
+          "expr": "sum(kafka_server_replica_manager_under_replicated_partitions)",
+          "refId": "A"
+        }
+      ],
+      "title": "Under-Replicated Partitions",
+      "type": "stat",
+      "description": "CRITICAL: Number of under-replicated partitions (should be 0)"
+    },
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "thresholds"
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              },
+              {
+                "color": "red",
+                "value": 1
+              }
+            ]
+          },
+          "unit": "short"
+        }
+      },
+      "gridPos": {
+        "h": 4,
+        "w": 4,
+        "x": 8,
+        "y": 0
+      },
+      "id": 3,
+      "options": {
+        "colorMode": "background",
+        "graphMode": "area",
+        "justifyMode": "auto",
+        "orientation": "auto",
+        "reduceOptions": {
+          "calcs": ["lastNotNull"],
+          "fields": "",
+          "values": false
+        },
+        "textMode": "auto"
+      },
+      "pluginVersion": "9.0.0",
+      "targets": [
+        {
+          "expr": "kafka_controller_offline_partitions_count",
+          "refId": "A"
+        }
+      ],
+      "title": "Offline Partitions",
+      "type": "stat",
+      "description": "CRITICAL: Number of offline partitions (should be 0)"
+    },
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "thresholds"
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              },
+              {
+                "color": "yellow",
+                "value": 1
+              },
+              {
+                "color": "red",
+                "value": 10
+              }
+            ]
+          },
+          "unit": "short"
+        }
+      },
+      "gridPos": {
+        "h": 4,
+        "w": 4,
+        "x": 12,
+        "y": 0
+      },
+      "id": 4,
+      "options": {
+        "colorMode": "value",
+        "graphMode": "area",
+        "justifyMode": "auto",
+        "orientation": "auto",
+        "reduceOptions": {
+          "calcs": ["lastNotNull"],
+          "fields": "",
+          "values": false
+        },
+        "textMode": "auto"
+      },
+      "pluginVersion": "9.0.0",
+      "targets": [
+        {
+          "expr": "sum(kafka_controller_unclean_leader_elections_total)",
+          "refId": "A"
+        }
+      ],
+      "title": "Unclean Leader Elections",
+      "type": "stat",
+      "description": "SHOULD BE ZERO! Unclean leader elections indicate data loss risk"
+    },
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic"
+          },
+          "custom": {
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "drawStyle": "line",
+            "fillOpacity": 10,
+            "gradientMode": "none",
+            "hideFrom": {
+              "tooltip": false,
+              "viz": false,
+              "legend": false
+            },
+            "lineInterpolation": "linear",
+            "lineWidth": 2,
+            "pointSize": 5,
+            "scaleDistribution": {
+              "type": "linear"
+            },
+            "showPoints": "never",
+            "spanNulls": true
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              }
+            ]
+          },
+          "unit": "Bps"
+        }
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 12,
+        "x": 0,
+        "y": 4
+      },
+      "id": 5,
+      "options": {
+        "legend": {
+          "calcs": ["mean", "lastNotNull", "max"],
+          "displayMode": "table",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "pluginVersion": "9.0.0",
+      "targets": [
+        {
+          "expr": "sum(rate(kafka_server_broker_topic_metrics_bytesin_total[5m]))",
+          "legendFormat": "Bytes In",
+          "refId": "A"
+        },
+        {
+          "expr": "sum(rate(kafka_server_broker_topic_metrics_bytesout_total[5m]))",
+          "legendFormat": "Bytes Out",
+          "refId": "B"
+        }
+      ],
+      "title": "Cluster Throughput (Bytes/sec)",
+      "type": "timeseries"
+    },
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic"
+          },
+          "custom": {
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "drawStyle": "line",
+            "fillOpacity": 10,
+            "gradientMode": "none",
+            "hideFrom": {
+              "tooltip": false,
+              "viz": false,
+              "legend": false
+            },
+            "lineInterpolation": "linear",
+            "lineWidth": 2,
+            "pointSize": 5,
+            "scaleDistribution": {
+              "type": "linear"
+            },
+            "showPoints": "never",
+            "spanNulls": true
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              }
+            ]
+          },
+          "unit": "ops"
+        }
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 12,
+        "x": 12,
+        "y": 4
+      },
+      "id": 6,
+      "options": {
+        "legend": {
+          "calcs": ["mean", "lastNotNull", "max"],
+          "displayMode": "table",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "pluginVersion": "9.0.0",
+      "targets": [
+        {
+          "expr": "sum(rate(kafka_server_broker_topic_metrics_messagesin_total[5m]))",
+          "legendFormat": "Messages In/sec",
+          "refId": "A"
+        },
+        {
+          "expr": "sum(rate(kafka_network_request_metrics_totaltime_total{request=\"Produce\"}[5m]))",
+          "legendFormat": "Produce Requests/sec",
+          "refId": "B"
+        },
+        {
+          "expr": "sum(rate(kafka_network_request_metrics_totaltime_total{request=\"FetchConsumer\"}[5m]))",
+          "legendFormat": "Fetch Requests/sec",
+          "refId": "C"
+        }
+      ],
+      "title": "Cluster Request Rate",
+      "type": "timeseries"
+    },
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic"
+          },
+          "custom": {
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "drawStyle": "line",
+            "fillOpacity": 10,
+            "gradientMode": "none",
+            "hideFrom": {
+              "tooltip": false,
+              "viz": false,
+              "legend": false
+            },
+            "lineInterpolation": "linear",
+            "lineWidth": 2,
+            "pointSize": 5,
+            "scaleDistribution": {
+              "type": "linear"
+            },
+            "showPoints": "never",
+            "spanNulls": true
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              }
+            ]
+          },
+          "unit": "short"
+        }
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 12,
+        "x": 0,
+        "y": 12
+      },
+      "id": 7,
+      "options": {
+        "legend": {
+          "calcs": ["mean", "lastNotNull", "max"],
+          "displayMode": "table",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "pluginVersion": "9.0.0",
+      "targets": [
+        {
+          "expr": "sum(rate(kafka_server_replica_manager_isr_shrinks_total[5m]))",
+          "legendFormat": "ISR Shrinks/sec",
+          "refId": "A"
+        },
+        {
+          "expr": "sum(rate(kafka_server_replica_manager_isr_expands_total[5m]))",
+          "legendFormat": "ISR Expands/sec",
+          "refId": "B"
+        }
+      ],
+      "title": "ISR Changes (In-Sync Replica)",
+      "type": "timeseries",
+      "description": "High ISR shrink rate indicates replicas falling out of sync"
+    },
+    {
+      "datasource": "Prometheus",
+      "fieldConfig": {
+        "defaults": {
+          "color": {
+            "mode": "palette-classic"
+          },
+          "custom": {
+            "axisLabel": "",
+            "axisPlacement": "auto",
+            "barAlignment": 0,
+            "drawStyle": "line",
+            "fillOpacity": 10,
+            "gradientMode": "none",
+            "hideFrom": {
+              "tooltip": false,
+              "viz": false,
+              "legend": false
+            },
+            "lineInterpolation": "linear",
+            "lineWidth": 2,
+            "pointSize": 5,
+            "scaleDistribution": {
+              "type": "linear"
+            },
+            "showPoints": "never",
+            "spanNulls": true
+          },
+          "mappings": [],
+          "thresholds": {
+            "mode": "absolute",
+            "steps": [
+              {
+                "color": "green",
+                "value": null
+              }
+            ]
+          },
+          "unit": "short"
+        }
+      },
+      "gridPos": {
+        "h": 8,
+        "w": 12,
+        "x": 12,
+        "y": 12
+      },
+      "id": 8,
+      "options": {
+        "legend": {
+          "calcs": ["mean", "lastNotNull", "max"],
+          "displayMode": "table",
+          "placement": "bottom"
+        },
+        "tooltip": {
+          "mode": "multi"
+        }
+      },
+      "pluginVersion": "9.0.0",
+      "targets": [
+        {
+          "expr": "sum(rate(kafka_controller_leader_election_total[5m]))",
+          "legendFormat": "Leader Elections/sec",
+          "refId": "A"
+        }
+      ],
+      "title": "Leader Election Rate",
+      "type": "timeseries",
+      "description": "High leader election rate indicates cluster instability"
+    }
+  ],
+  "refresh": "30s",
+  "schemaVersion": 36,
+  "style": "dark",
+  "tags": ["kafka", "cluster", "overview"],
+  "templating": {
+    "list": []
+  },
+  "time": {
+    "from": "now-1h",
+    "to": "now"
+  },
+  "timepicker": {},
+  "timezone": "",
+  "title": "Kafka Cluster Overview",
+  "uid": "kafka-cluster-overview",
+  "version": 1
+}