{
  "annotations": {
    "list": [
      {
        "builtIn": 1,
        "datasource": {
          "type": "grafana",
          "uid": "-- Grafana --"
        },
        "enable": true,
        "hide": true,
        "iconColor": "rgba(0, 211, 255, 1)",
        "name": "Annotations & Alerts",
        "type": "dashboard"
      }
    ]
  },
  "description": "Oshun node-exporter dashboard: CPU, memory, disk I/O, filesystem usage, network throughput, load average. Uses the Prometheus node_exporter schema.",
  "editable": true,
  "fiscalYearStartMonth": 0,
  "graphTooltip": 1,
  "id": null,
  "liveNow": false,
  "panels": [
    {
      "datasource": { "type": "prometheus", "uid": "${datasource}" },
      "description": "CPU cores reported by node_exporter on the selected instance.",
      "fieldConfig": {
        "defaults": {
          "color": { "mode": "thresholds" },
          "mappings": [],
          "thresholds": { "mode": "absolute", "steps": [{ "color": "blue", "value": null }] },
          "unit": "short"
        }
      },
      "gridPos": { "h": 4, "w": 6, "x": 0, "y": 0 },
      "id": 1,
      "options": {
        "colorMode": "background",
        "graphMode": "none",
        "justifyMode": "auto",
        "orientation": "auto",
        "reduceOptions": { "calcs": ["lastNotNull"], "fields": "", "values": false },
        "textMode": "value_and_name"
      },
      "pluginVersion": "10.2.0",
      "targets": [
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "count(count by (cpu) (node_cpu_seconds_total{instance=~\"$instance\"}))",
          "legendFormat": "cores",
          "range": true,
          "refId": "A"
        }
      ],
      "title": "CPU Cores",
      "type": "stat"
    },
    {
      "datasource": { "type": "prometheus", "uid": "${datasource}" },
      "description": "Total physical memory on the selected instance.",
      "fieldConfig": {
        "defaults": {
          "color": { "mode": "thresholds" },
          "mappings": [],
          "thresholds": { "mode": "absolute", "steps": [{ "color": "blue", "value": null }] },
          "unit": "bytes"
        }
      },
      "gridPos": { "h": 4, "w": 6, "x": 6, "y": 0 },
      "id": 2,
      "options": {
        "colorMode": "background",
        "graphMode": "none",
        "justifyMode": "auto",
        "orientation": "auto",
        "reduceOptions": { "calcs": ["lastNotNull"], "fields": "", "values": false },
        "textMode": "value_and_name"
      },
      "pluginVersion": "10.2.0",
      "targets": [
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "max(node_memory_MemTotal_bytes{instance=~\"$instance\"})",
          "legendFormat": "total",
          "range": true,
          "refId": "A"
        }
      ],
      "title": "Memory Total",
      "type": "stat"
    },
    {
      "datasource": { "type": "prometheus", "uid": "${datasource}" },
      "description": "System uptime on the selected instance.",
      "fieldConfig": {
        "defaults": {
          "color": { "mode": "thresholds" },
          "mappings": [],
          "thresholds": { "mode": "absolute", "steps": [{ "color": "blue", "value": null }] },
          "unit": "s"
        }
      },
      "gridPos": { "h": 4, "w": 6, "x": 12, "y": 0 },
      "id": 3,
      "options": {
        "colorMode": "background",
        "graphMode": "none",
        "justifyMode": "auto",
        "orientation": "auto",
        "reduceOptions": { "calcs": ["lastNotNull"], "fields": "", "values": false },
        "textMode": "value_and_name"
      },
      "pluginVersion": "10.2.0",
      "targets": [
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "max(time() - node_boot_time_seconds{instance=~\"$instance\"})",
          "legendFormat": "uptime",
          "range": true,
          "refId": "A"
        }
      ],
      "title": "Uptime",
      "type": "stat"
    },
    {
      "datasource": { "type": "prometheus", "uid": "${datasource}" },
      "description": "1-minute load average. Divide by cores to get a normalized load.",
      "fieldConfig": {
        "defaults": {
          "color": { "mode": "thresholds" },
          "mappings": [],
          "thresholds": {
            "mode": "absolute",
            "steps": [
              { "color": "green", "value": null },
              { "color": "orange", "value": 2 },
              { "color": "red", "value": 8 }
            ]
          },
          "unit": "short"
        }
      },
      "gridPos": { "h": 4, "w": 6, "x": 18, "y": 0 },
      "id": 4,
      "options": {
        "colorMode": "background",
        "graphMode": "area",
        "justifyMode": "auto",
        "orientation": "auto",
        "reduceOptions": { "calcs": ["lastNotNull"], "fields": "", "values": false },
        "textMode": "value_and_name"
      },
      "pluginVersion": "10.2.0",
      "targets": [
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "max(node_load1{instance=~\"$instance\"})",
          "legendFormat": "load1",
          "range": true,
          "refId": "A"
        }
      ],
      "title": "Load (1m)",
      "type": "stat"
    },
    {
      "datasource": { "type": "prometheus", "uid": "${datasource}" },
      "description": "Per-instance CPU usage by mode.",
      "fieldConfig": {
        "defaults": {
          "color": { "mode": "palette-classic" },
          "custom": {
            "drawStyle": "line",
            "fillOpacity": 25,
            "gradientMode": "opacity",
            "lineInterpolation": "smooth",
            "lineWidth": 2,
            "pointSize": 5,
            "showPoints": "never",
            "spanNulls": true,
            "stacking": { "group": "A", "mode": "percent" }
          },
          "mappings": [],
          "max": 1,
          "min": 0,
          "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] },
          "unit": "percentunit"
        }
      },
      "gridPos": { "h": 8, "w": 12, "x": 0, "y": 4 },
      "id": 5,
      "options": {
        "legend": {
          "calcs": ["mean"],
          "displayMode": "table",
          "placement": "right",
          "showLegend": true
        },
        "tooltip": { "mode": "multi", "sort": "desc" }
      },
      "targets": [
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "avg by (mode) (rate(node_cpu_seconds_total{instance=~\"$instance\",mode!=\"idle\"}[5m]))",
          "legendFormat": "{{mode}}",
          "range": true,
          "refId": "A"
        }
      ],
      "title": "CPU Utilization by Mode",
      "type": "timeseries"
    },
    {
      "datasource": { "type": "prometheus", "uid": "${datasource}" },
      "description": "Memory breakdown: used vs available.",
      "fieldConfig": {
        "defaults": {
          "color": { "mode": "palette-classic" },
          "custom": {
            "drawStyle": "line",
            "fillOpacity": 25,
            "gradientMode": "opacity",
            "lineInterpolation": "smooth",
            "lineWidth": 2,
            "pointSize": 5,
            "showPoints": "never",
            "spanNulls": true,
            "stacking": { "group": "A", "mode": "normal" }
          },
          "mappings": [],
          "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] },
          "unit": "bytes"
        }
      },
      "gridPos": { "h": 8, "w": 12, "x": 12, "y": 4 },
      "id": 6,
      "options": {
        "legend": {
          "calcs": ["mean", "max"],
          "displayMode": "table",
          "placement": "right",
          "showLegend": true
        },
        "tooltip": { "mode": "multi", "sort": "desc" }
      },
      "targets": [
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "avg(node_memory_MemTotal_bytes{instance=~\"$instance\"}) - avg(node_memory_MemAvailable_bytes{instance=~\"$instance\"})",
          "legendFormat": "used",
          "range": true,
          "refId": "A"
        },
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "avg(node_memory_Buffers_bytes{instance=~\"$instance\"})",
          "legendFormat": "buffers",
          "range": true,
          "refId": "B"
        },
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "avg(node_memory_Cached_bytes{instance=~\"$instance\"})",
          "legendFormat": "cached",
          "range": true,
          "refId": "C"
        },
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "avg(node_memory_MemFree_bytes{instance=~\"$instance\"})",
          "legendFormat": "free",
          "range": true,
          "refId": "D"
        }
      ],
      "title": "Memory Breakdown",
      "type": "timeseries"
    },
    {
      "datasource": { "type": "prometheus", "uid": "${datasource}" },
      "description": "Disk I/O throughput per device (read + write).",
      "fieldConfig": {
        "defaults": {
          "color": { "mode": "palette-classic" },
          "custom": {
            "drawStyle": "line",
            "fillOpacity": 15,
            "gradientMode": "opacity",
            "lineInterpolation": "smooth",
            "lineWidth": 2,
            "pointSize": 5,
            "showPoints": "never",
            "spanNulls": true
          },
          "mappings": [],
          "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] },
          "unit": "Bps"
        }
      },
      "gridPos": { "h": 8, "w": 12, "x": 0, "y": 12 },
      "id": 7,
      "options": {
        "legend": {
          "calcs": ["mean", "max"],
          "displayMode": "table",
          "placement": "right",
          "showLegend": true
        },
        "tooltip": { "mode": "multi", "sort": "desc" }
      },
      "targets": [
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "sum by (device) (rate(node_disk_read_bytes_total{instance=~\"$instance\",device!~\"loop.*|ram.*\"}[5m]))",
          "legendFormat": "{{device}} read",
          "range": true,
          "refId": "A"
        },
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "sum by (device) (rate(node_disk_written_bytes_total{instance=~\"$instance\",device!~\"loop.*|ram.*\"}[5m]))",
          "legendFormat": "{{device}} write",
          "range": true,
          "refId": "B"
        }
      ],
      "title": "Disk I/O",
      "type": "timeseries"
    },
    {
      "datasource": { "type": "prometheus", "uid": "${datasource}" },
      "description": "Network throughput (rx + tx) per interface.",
      "fieldConfig": {
        "defaults": {
          "color": { "mode": "palette-classic" },
          "custom": {
            "drawStyle": "line",
            "fillOpacity": 15,
            "gradientMode": "opacity",
            "lineInterpolation": "smooth",
            "lineWidth": 2,
            "pointSize": 5,
            "showPoints": "never",
            "spanNulls": true
          },
          "mappings": [],
          "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] },
          "unit": "bps"
        }
      },
      "gridPos": { "h": 8, "w": 12, "x": 12, "y": 12 },
      "id": 8,
      "options": {
        "legend": {
          "calcs": ["mean", "max"],
          "displayMode": "table",
          "placement": "right",
          "showLegend": true
        },
        "tooltip": { "mode": "multi", "sort": "desc" }
      },
      "targets": [
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "sum by (device) (rate(node_network_receive_bytes_total{instance=~\"$instance\",device!~\"lo|veth.*|docker.*\"}[5m]) * 8)",
          "legendFormat": "{{device}} rx",
          "range": true,
          "refId": "A"
        },
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "sum by (device) (rate(node_network_transmit_bytes_total{instance=~\"$instance\",device!~\"lo|veth.*|docker.*\"}[5m]) * 8)",
          "legendFormat": "{{device}} tx",
          "range": true,
          "refId": "B"
        }
      ],
      "title": "Network Throughput",
      "type": "timeseries"
    },
    {
      "datasource": { "type": "prometheus", "uid": "${datasource}" },
      "description": "Filesystem usage percentage per mountpoint. Alert threshold at 85%.",
      "fieldConfig": {
        "defaults": {
          "color": { "mode": "thresholds" },
          "custom": {
            "drawStyle": "line",
            "fillOpacity": 15,
            "gradientMode": "opacity",
            "lineInterpolation": "smooth",
            "lineWidth": 2,
            "pointSize": 5,
            "showPoints": "never",
            "spanNulls": true
          },
          "mappings": [],
          "max": 1,
          "min": 0,
          "thresholds": {
            "mode": "absolute",
            "steps": [
              { "color": "green", "value": null },
              { "color": "orange", "value": 0.75 },
              { "color": "red", "value": 0.9 }
            ]
          },
          "unit": "percentunit"
        }
      },
      "gridPos": { "h": 8, "w": 12, "x": 0, "y": 20 },
      "id": 9,
      "options": {
        "legend": {
          "calcs": ["lastNotNull", "max"],
          "displayMode": "table",
          "placement": "right",
          "showLegend": true
        },
        "tooltip": { "mode": "multi", "sort": "desc" }
      },
      "targets": [
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "1 - (node_filesystem_avail_bytes{instance=~\"$instance\",fstype!~\"tmpfs|overlay|nsfs|squashfs\"} / node_filesystem_size_bytes{instance=~\"$instance\",fstype!~\"tmpfs|overlay|nsfs|squashfs\"})",
          "legendFormat": "{{mountpoint}}",
          "range": true,
          "refId": "A"
        }
      ],
      "title": "Filesystem Usage",
      "type": "timeseries"
    },
    {
      "datasource": { "type": "prometheus", "uid": "${datasource}" },
      "description": "Context switches and interrupts per second. Sudden spikes often correlate with runaway workloads.",
      "fieldConfig": {
        "defaults": {
          "color": { "mode": "palette-classic" },
          "custom": {
            "drawStyle": "line",
            "fillOpacity": 10,
            "gradientMode": "opacity",
            "lineInterpolation": "smooth",
            "lineWidth": 2,
            "pointSize": 5,
            "showPoints": "never",
            "spanNulls": true
          },
          "mappings": [],
          "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] },
          "unit": "ops"
        }
      },
      "gridPos": { "h": 8, "w": 12, "x": 12, "y": 20 },
      "id": 10,
      "options": {
        "legend": {
          "calcs": ["mean", "max"],
          "displayMode": "table",
          "placement": "right",
          "showLegend": true
        },
        "tooltip": { "mode": "multi", "sort": "desc" }
      },
      "targets": [
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "sum(rate(node_context_switches_total{instance=~\"$instance\"}[5m]))",
          "legendFormat": "context switches",
          "range": true,
          "refId": "A"
        },
        {
          "datasource": { "type": "prometheus", "uid": "${datasource}" },
          "editorMode": "code",
          "expr": "sum(rate(node_intr_total{instance=~\"$instance\"}[5m]))",
          "legendFormat": "interrupts",
          "range": true,
          "refId": "B"
        }
      ],
      "title": "Context Switches & Interrupts",
      "type": "timeseries"
    }
  ],
  "refresh": "30s",
  "schemaVersion": 39,
  "tags": ["oshun", "infrastructure", "node", "node-exporter"],
  "templating": {
    "list": [
      {
        "current": { "selected": false, "text": "Prometheus", "value": "prometheus" },
        "hide": 0,
        "includeAll": false,
        "label": "Datasource",
        "multi": false,
        "name": "datasource",
        "options": [],
        "query": "prometheus",
        "refresh": 1,
        "regex": "",
        "skipUrlSync": false,
        "type": "datasource"
      },
      {
        "allValue": ".+",
        "current": { "selected": true, "text": ["All"], "value": ["$__all"] },
        "datasource": { "type": "prometheus", "uid": "${datasource}" },
        "definition": "label_values(node_cpu_seconds_total, instance)",
        "hide": 0,
        "includeAll": true,
        "label": "Instance",
        "multi": true,
        "name": "instance",
        "options": [],
        "query": {
          "query": "label_values(node_cpu_seconds_total, instance)",
          "refId": "StandardVariableQuery"
        },
        "refresh": 2,
        "regex": "",
        "skipUrlSync": false,
        "sort": 1,
        "type": "query"
      }
    ]
  },
  "time": { "from": "now-3h", "to": "now" },
  "timepicker": {},
  "timezone": "",
  "title": "Infrastructure — Node",
  "uid": "infra-node",
  "version": 1,
  "weekStart": ""
}