Grafana Dashboards & Visualization

Time Series Panel

{
  "type": "timeseries",
  "title": "Request Rate",
  "gridPos": { "x": 0, "y": 0, "w": 12, "h": 8 },
  "datasource": { "type": "prometheus", "uid": "prometheus-main" },
  "targets": [{
    "expr": "sum(rate(http_requests_total{job=\"$job\"}[5m])) by (status_code)",
    "legendFormat": "{{status_code}}"
  }],
  "fieldConfig": {
    "defaults": {
      "unit": "reqps",
      "custom": { "drawStyle": "line", "fillOpacity": 10, "lineWidth": 2 }
    },
    "overrides": [{
      "matcher": { "id": "byName", "options": "500" },
      "properties": [{ "id": "color", "value": { "fixedColor": "red", "mode": "fixed" } }]
    }]
  }
}

Stat Panel

{
  "type": "stat",
  "title": "Total Requests",
  "targets": [{ "expr": "sum(increase(http_requests_total{job=\"$job\"}[24h]))" }],
  "fieldConfig": {
    "defaults": {
      "unit": "short",
      "thresholds": {
        "mode": "absolute",
        "steps": [
          { "color": "green", "value": null },
          { "color": "yellow", "value": 10000 },
          { "color": "red", "value": 50000 }
        ]
      }
    }
  },
  "options": { "reduceOptions": { "calcs": ["lastNotNull"] }, "colorMode": "background" }
}

Gauge Panel

Set options.showThresholdLabels: true and fieldConfig.defaults.min/max for proper scaling. Use unit field for display (percent, bytes, etc.).

Table Panel

{
  "type": "table",
  "title": "Top Endpoints",
  "targets": [{
    "expr": "topk(10, sum(rate(http_requests_total[5m])) by (handler))",
    "format": "table", "instant": true
  }],
  "transformations": [
    { "id": "organize", "options": { "excludeByName": { "Time": true } } }
  ],
  "fieldConfig": {
    "overrides": [{
      "matcher": { "id": "byName", "options": "Value" },
      "properties": [{ "id": "custom.cellOptions", "value": { "type": "color-background" } }]
    }]
  }
}

Logs Panel

{
  "type": "logs",
  "title": "Application Logs",
  "datasource": { "type": "loki", "uid": "loki-main" },
  "targets": [{
    "expr": "{namespace=\"$namespace\", app=\"$app\"} |= \"$search\" | logfmt | level=~\"$level\"",
    "refId": "A"
  }],
  "options": { "showTime": true, "sortOrder": "Descending", "wrapLogMessage": true }
}

Heatmap, Bar Chart, Geomap Panels

• Heatmap: "type": "heatmap". Set options.calculate: true for raw data, false for pre-bucketed. Configure options.yAxis.unit and color.scheme.
• Bar Chart: "type": "barchart". Set options.orientation ("horizontal"/"vertical"), options.showValue. Pair with instant queries and table format.
• Geomap: "type": "geomap". Requires latitude/longitude fields or field mappings. Set options.view.id for map type.

Queries

PromQL Patterns for Dashboards

# Rate of counter
rate(http_requests_total{job="$job"}[$__rate_interval])

# Error ratio
sum(rate(http_requests_total{status=~"5.."}[5m])) / sum(rate(http_requests_total[5m]))

# Histogram quantile (p99 latency)
histogram_quantile(0.99, sum(rate(http_request_duration_seconds_bucket{job="$job"}[5m])) by (le))

# Aggregation with variable
sum by ($groupby) (rate(metric_name[$__rate_interval]))

Use $__rate_interval instead of hardcoded [5m] — it auto-adjusts to scrape interval. Use $__interval for non-rate aggregations.

LogQL Patterns for Dashboards

# Filter and parse
{namespace="production", app="api"} |= "error" | json | line_format "{{.message}}"

# Metric query from logs (rate of errors)
sum(rate({app="api"} |= "error" [5m])) by (level)

# Pattern-based extraction
{job="nginx"} | pattern "<ip> - - <_> \"<method> <uri> <_>\" <status> <size>" | status >= 400

Variables (Templating)

Define in templating.list. Variable types:

Query Variable (dynamic from datasource)

{
  "name": "namespace",
  "type": "query",
  "datasource": { "type": "prometheus", "uid": "prometheus-main" },
  "query": "label_values(kube_pod_info, namespace)",
  "refresh": 2,
  "sort": 1,
  "multi": true,
  "includeAll": true,
  "allValue": ".*"
}

refresh: 1=on dashboard load, 2=on time range change. sort: 0=disabled, 1=alpha asc, 2=alpha desc, 3=numeric asc.

Chained Variables

Create dependency chains: namespace → service → pod.

[
  { "name": "namespace", "query": "label_values(kube_pod_info, namespace)" },
  { "name": "service", "query": "label_values(kube_pod_info{namespace=\"$namespace\"}, service)" },
  { "name": "pod", "query": "label_values(kube_pod_info{namespace=\"$namespace\", service=\"$service\"}, pod)" }
]

Custom Variable

{ "name": "env", "type": "custom", "query": "production,staging,development", "current": { "text": "production", "value": "production" } }

Interval Variable

{ "name": "smoothing", "type": "interval", "query": "1m,5m,15m,30m,1h", "auto": true, "auto_count": 30, "auto_min": "10s" }

Datasource Variable

{ "name": "ds_prometheus", "type": "datasource", "query": "prometheus" }

Reference in panels: "datasource": { "type": "prometheus", "uid": "$ds_prometheus" }.

Variable Syntax

Use $var for simple substitution, ${var} mid-string, ${var:regex} for regex-escaped, ${var:pipe} for pipe-delimited multi-values, ${var:csv} for comma-separated.

Annotations

{
  "annotations": {
    "list": [{
      "name": "Deployments",
      "datasource": { "type": "prometheus", "uid": "prometheus-main" },
      "expr": "changes(deploy_timestamp{app=\"$app\"}[1m]) > 0",
      "tagKeys": "app,version",
      "textFormat": "Deployed {{version}}",
      "iconColor": "blue",
      "enable": true
    }]
  }
}

Dashboard Links

{
  "links": [
    { "title": "Logs", "url": "/d/logs-dash/logs?var-app=$app", "type": "link", "icon": "doc", "targetBlank": false },
    { "title": "Related", "type": "dashboards", "tags": ["production"], "asDropdown": true }
  ]
}

Provisioning

Dashboard Provider (YAML)

Place in provisioning/dashboards/. Set foldersFromFilesStructure: true to mirror subdirectory layout as Grafana folders. Set allowUiUpdates: false to enforce file-based source of truth.

apiVersion: 1