mirror of
https://github.com/storytold/terraform-aws-observability-accelerator.git
synced 2026-10-09 00:09:43 +00:00
Rodrigue Koffi
626 lines
16 KiB
JSON
626 lines
16 KiB
JSON
{
|
|
"annotations": {
|
|
"list": [
|
|
{
|
|
"builtIn": 1,
|
|
"datasource": {
|
|
"type": "grafana",
|
|
"uid": "-- Grafana --"
|
|
},
|
|
"enable": true,
|
|
"hide": true,
|
|
"iconColor": "rgba(0, 211, 255, 1)",
|
|
"name": "Annotations & Alerts",
|
|
"target": {
|
|
"limit": 100,
|
|
"matchAny": false,
|
|
"tags": [],
|
|
"type": "dashboard"
|
|
},
|
|
"type": "dashboard"
|
|
}
|
|
]
|
|
},
|
|
"editable": false,
|
|
"fiscalYearStartMonth": 0,
|
|
"graphTooltip": 1,
|
|
"id": 22,
|
|
"iteration": 1656210900773,
|
|
"links": [],
|
|
"liveNow": false,
|
|
"panels": [
|
|
{
|
|
"collapsed": false,
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"gridPos": {
|
|
"h": 1,
|
|
"w": 24,
|
|
"x": 0,
|
|
"y": 0
|
|
},
|
|
"id": 6,
|
|
"panels": [],
|
|
"targets": [
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "Alerts",
|
|
"type": "row"
|
|
},
|
|
{
|
|
"aliasColors": {},
|
|
"bars": false,
|
|
"dashLength": 10,
|
|
"dashes": false,
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"fill": 1,
|
|
"fillGradient": 0,
|
|
"gridPos": {
|
|
"h": 7,
|
|
"w": 12,
|
|
"x": 0,
|
|
"y": 1
|
|
},
|
|
"hiddenSeries": false,
|
|
"id": 2,
|
|
"legend": {
|
|
"alignAsTable": false,
|
|
"avg": false,
|
|
"current": false,
|
|
"max": false,
|
|
"min": false,
|
|
"rightSide": false,
|
|
"show": false,
|
|
"total": false,
|
|
"values": false
|
|
},
|
|
"lines": true,
|
|
"linewidth": 1,
|
|
"links": [],
|
|
"nullPointMode": "null",
|
|
"options": {
|
|
"alertThreshold": true
|
|
},
|
|
"percentage": false,
|
|
"pluginVersion": "9.0.1",
|
|
"pointradius": 5,
|
|
"points": false,
|
|
"renderer": "flot",
|
|
"seriesOverrides": [],
|
|
"spaceLength": 10,
|
|
"stack": true,
|
|
"steppedLine": false,
|
|
"targets": [
|
|
{
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"expr": "sum(alertmanager_alerts{namespace=~\"$namespace\",service=~\"$service\"}) by (namespace,service,instance)",
|
|
"format": "time_series",
|
|
"intervalFactor": 2,
|
|
"legendFormat": "{{instance}}",
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"thresholds": [],
|
|
"timeRegions": [],
|
|
"title": "Alerts",
|
|
"tooltip": {
|
|
"shared": true,
|
|
"sort": 0,
|
|
"value_type": "individual"
|
|
},
|
|
"type": "graph",
|
|
"xaxis": {
|
|
"mode": "time",
|
|
"show": true,
|
|
"values": []
|
|
},
|
|
"yaxes": [
|
|
{
|
|
"format": "none",
|
|
"logBase": 1,
|
|
"show": true
|
|
},
|
|
{
|
|
"format": "none",
|
|
"logBase": 1,
|
|
"show": true
|
|
}
|
|
],
|
|
"yaxis": {
|
|
"align": false
|
|
}
|
|
},
|
|
{
|
|
"aliasColors": {},
|
|
"bars": false,
|
|
"dashLength": 10,
|
|
"dashes": false,
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"fill": 1,
|
|
"fillGradient": 0,
|
|
"gridPos": {
|
|
"h": 7,
|
|
"w": 12,
|
|
"x": 12,
|
|
"y": 1
|
|
},
|
|
"hiddenSeries": false,
|
|
"id": 3,
|
|
"legend": {
|
|
"alignAsTable": false,
|
|
"avg": false,
|
|
"current": false,
|
|
"max": false,
|
|
"min": false,
|
|
"rightSide": false,
|
|
"show": false,
|
|
"total": false,
|
|
"values": false
|
|
},
|
|
"lines": true,
|
|
"linewidth": 1,
|
|
"links": [],
|
|
"nullPointMode": "null",
|
|
"options": {
|
|
"alertThreshold": true
|
|
},
|
|
"percentage": false,
|
|
"pluginVersion": "9.0.1",
|
|
"pointradius": 5,
|
|
"points": false,
|
|
"renderer": "flot",
|
|
"seriesOverrides": [],
|
|
"spaceLength": 10,
|
|
"stack": true,
|
|
"steppedLine": false,
|
|
"targets": [
|
|
{
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"expr": "sum(rate(alertmanager_alerts_received_total{namespace=~\"$namespace\",service=~\"$service\"}[$__rate_interval])) by (namespace,service,instance)",
|
|
"format": "time_series",
|
|
"intervalFactor": 2,
|
|
"legendFormat": "{{instance}} Received",
|
|
"refId": "A"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"expr": "sum(rate(alertmanager_alerts_invalid_total{namespace=~\"$namespace\",service=~\"$service\"}[$__rate_interval])) by (namespace,service,instance)",
|
|
"format": "time_series",
|
|
"intervalFactor": 2,
|
|
"legendFormat": "{{instance}} Invalid",
|
|
"refId": "B"
|
|
}
|
|
],
|
|
"thresholds": [],
|
|
"timeRegions": [],
|
|
"title": "Alerts receive rate",
|
|
"tooltip": {
|
|
"shared": true,
|
|
"sort": 0,
|
|
"value_type": "individual"
|
|
},
|
|
"type": "graph",
|
|
"xaxis": {
|
|
"mode": "time",
|
|
"show": true,
|
|
"values": []
|
|
},
|
|
"yaxes": [
|
|
{
|
|
"format": "ops",
|
|
"logBase": 1,
|
|
"show": true
|
|
},
|
|
{
|
|
"format": "ops",
|
|
"logBase": 1,
|
|
"show": true
|
|
}
|
|
],
|
|
"yaxis": {
|
|
"align": false
|
|
}
|
|
},
|
|
{
|
|
"collapsed": false,
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"gridPos": {
|
|
"h": 1,
|
|
"w": 24,
|
|
"x": 0,
|
|
"y": 8
|
|
},
|
|
"id": 7,
|
|
"panels": [],
|
|
"targets": [
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "Notifications",
|
|
"type": "row"
|
|
},
|
|
{
|
|
"aliasColors": {},
|
|
"bars": false,
|
|
"dashLength": 10,
|
|
"dashes": false,
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"fill": 1,
|
|
"fillGradient": 0,
|
|
"gridPos": {
|
|
"h": 7,
|
|
"w": 6,
|
|
"x": 0,
|
|
"y": 9
|
|
},
|
|
"hiddenSeries": false,
|
|
"id": 4,
|
|
"legend": {
|
|
"alignAsTable": false,
|
|
"avg": false,
|
|
"current": false,
|
|
"max": false,
|
|
"min": false,
|
|
"rightSide": false,
|
|
"show": false,
|
|
"total": false,
|
|
"values": false
|
|
},
|
|
"lines": true,
|
|
"linewidth": 1,
|
|
"links": [],
|
|
"nullPointMode": "null",
|
|
"options": {
|
|
"alertThreshold": true
|
|
},
|
|
"percentage": false,
|
|
"pluginVersion": "9.0.1",
|
|
"pointradius": 5,
|
|
"points": false,
|
|
"renderer": "flot",
|
|
"repeat": "integration",
|
|
"seriesOverrides": [],
|
|
"spaceLength": 10,
|
|
"stack": true,
|
|
"steppedLine": false,
|
|
"targets": [
|
|
{
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"expr": "sum(rate(alertmanager_notifications_total{namespace=~\"$namespace\",service=~\"$service\", integration=\"$integration\"}[$__rate_interval])) by (integration,namespace,service,instance)",
|
|
"format": "time_series",
|
|
"intervalFactor": 2,
|
|
"legendFormat": "{{instance}} Total",
|
|
"refId": "A"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"expr": "sum(rate(alertmanager_notifications_failed_total{namespace=~\"$namespace\",service=~\"$service\", integration=\"$integration\"}[$__rate_interval])) by (integration,namespace,service,instance)",
|
|
"format": "time_series",
|
|
"intervalFactor": 2,
|
|
"legendFormat": "{{instance}} Failed",
|
|
"refId": "B"
|
|
}
|
|
],
|
|
"thresholds": [],
|
|
"timeRegions": [],
|
|
"title": "$integration: Notifications Send Rate",
|
|
"tooltip": {
|
|
"shared": true,
|
|
"sort": 0,
|
|
"value_type": "individual"
|
|
},
|
|
"type": "graph",
|
|
"xaxis": {
|
|
"mode": "time",
|
|
"show": true,
|
|
"values": []
|
|
},
|
|
"yaxes": [
|
|
{
|
|
"format": "ops",
|
|
"logBase": 1,
|
|
"show": true
|
|
},
|
|
{
|
|
"format": "ops",
|
|
"logBase": 1,
|
|
"show": true
|
|
}
|
|
],
|
|
"yaxis": {
|
|
"align": false
|
|
}
|
|
},
|
|
{
|
|
"aliasColors": {},
|
|
"bars": false,
|
|
"dashLength": 10,
|
|
"dashes": false,
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"fill": 1,
|
|
"fillGradient": 0,
|
|
"gridPos": {
|
|
"h": 7,
|
|
"w": 6,
|
|
"x": 0,
|
|
"y": 16
|
|
},
|
|
"hiddenSeries": false,
|
|
"id": 5,
|
|
"legend": {
|
|
"alignAsTable": false,
|
|
"avg": false,
|
|
"current": false,
|
|
"max": false,
|
|
"min": false,
|
|
"rightSide": false,
|
|
"show": false,
|
|
"total": false,
|
|
"values": false
|
|
},
|
|
"lines": true,
|
|
"linewidth": 1,
|
|
"links": [],
|
|
"nullPointMode": "null",
|
|
"options": {
|
|
"alertThreshold": true
|
|
},
|
|
"percentage": false,
|
|
"pluginVersion": "9.0.1",
|
|
"pointradius": 5,
|
|
"points": false,
|
|
"renderer": "flot",
|
|
"repeat": "integration",
|
|
"seriesOverrides": [],
|
|
"spaceLength": 10,
|
|
"stack": false,
|
|
"steppedLine": false,
|
|
"targets": [
|
|
{
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"expr": "histogram_quantile(0.99,\n sum(rate(alertmanager_notification_latency_seconds_bucket{namespace=~\"$namespace\",service=~\"$service\", integration=\"$integration\"}[$__rate_interval])) by (le,namespace,service,instance)\n) \n",
|
|
"format": "time_series",
|
|
"intervalFactor": 2,
|
|
"legendFormat": "{{instance}} 99th Percentile",
|
|
"refId": "A"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"expr": "histogram_quantile(0.50,\n sum(rate(alertmanager_notification_latency_seconds_bucket{namespace=~\"$namespace\",service=~\"$service\", integration=\"$integration\"}[$__rate_interval])) by (le,namespace,service,instance)\n) \n",
|
|
"format": "time_series",
|
|
"intervalFactor": 2,
|
|
"legendFormat": "{{instance}} Median",
|
|
"refId": "B"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"uid": "$datasource"
|
|
},
|
|
"expr": "sum(rate(alertmanager_notification_latency_seconds_sum{namespace=~\"$namespace\",service=~\"$service\", integration=\"$integration\"}[$__rate_interval])) by (namespace,service,instance)\n/\nsum(rate(alertmanager_notification_latency_seconds_count{namespace=~\"$namespace\",service=~\"$service\", integration=\"$integration\"}[$__rate_interval])) by (namespace,service,instance)\n",
|
|
"format": "time_series",
|
|
"intervalFactor": 2,
|
|
"legendFormat": "{{instance}} Average",
|
|
"refId": "C"
|
|
}
|
|
],
|
|
"thresholds": [],
|
|
"timeRegions": [],
|
|
"title": "$integration: Notification Duration",
|
|
"tooltip": {
|
|
"shared": true,
|
|
"sort": 0,
|
|
"value_type": "individual"
|
|
},
|
|
"type": "graph",
|
|
"xaxis": {
|
|
"mode": "time",
|
|
"show": true,
|
|
"values": []
|
|
},
|
|
"yaxes": [
|
|
{
|
|
"format": "s",
|
|
"logBase": 1,
|
|
"show": true
|
|
},
|
|
{
|
|
"format": "s",
|
|
"logBase": 1,
|
|
"show": true
|
|
}
|
|
],
|
|
"yaxis": {
|
|
"align": false
|
|
}
|
|
}
|
|
],
|
|
"refresh": "30s",
|
|
"schemaVersion": 36,
|
|
"style": "dark",
|
|
"tags": [
|
|
"alertmanager-mixin"
|
|
],
|
|
"templating": {
|
|
"list": [
|
|
{
|
|
"current": {
|
|
"selected": false,
|
|
"text": "Prometheus",
|
|
"value": "Prometheus"
|
|
},
|
|
"hide": 0,
|
|
"includeAll": false,
|
|
"label": "Data Source",
|
|
"multi": false,
|
|
"name": "datasource",
|
|
"options": [],
|
|
"query": "prometheus",
|
|
"refresh": 1,
|
|
"regex": "",
|
|
"skipUrlSync": false,
|
|
"type": "datasource"
|
|
},
|
|
{
|
|
"current": {
|
|
"selected": false,
|
|
"text": "default",
|
|
"value": "default"
|
|
},
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "$datasource"
|
|
},
|
|
"definition": "",
|
|
"hide": 0,
|
|
"includeAll": false,
|
|
"label": "namespace",
|
|
"multi": false,
|
|
"name": "namespace",
|
|
"options": [],
|
|
"query": {
|
|
"query": "label_values(alertmanager_alerts, namespace)",
|
|
"refId": "Prometheus-namespace-Variable-Query"
|
|
},
|
|
"refresh": 2,
|
|
"regex": "",
|
|
"skipUrlSync": false,
|
|
"sort": 1,
|
|
"tagValuesQuery": "",
|
|
"tagsQuery": "",
|
|
"type": "query",
|
|
"useTags": false
|
|
},
|
|
{
|
|
"current": {
|
|
"selected": false,
|
|
"text": "kube-prometheus-stack-alertmanager",
|
|
"value": "kube-prometheus-stack-alertmanager"
|
|
},
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "$datasource"
|
|
},
|
|
"definition": "",
|
|
"hide": 0,
|
|
"includeAll": false,
|
|
"label": "service",
|
|
"multi": false,
|
|
"name": "service",
|
|
"options": [],
|
|
"query": {
|
|
"query": "label_values(alertmanager_alerts, service)",
|
|
"refId": "Prometheus-service-Variable-Query"
|
|
},
|
|
"refresh": 2,
|
|
"regex": "",
|
|
"skipUrlSync": false,
|
|
"sort": 1,
|
|
"tagValuesQuery": "",
|
|
"tagsQuery": "",
|
|
"type": "query",
|
|
"useTags": false
|
|
},
|
|
{
|
|
"current": {
|
|
"selected": false,
|
|
"text": "All",
|
|
"value": "$__all"
|
|
},
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "$datasource"
|
|
},
|
|
"definition": "",
|
|
"hide": 2,
|
|
"includeAll": true,
|
|
"multi": false,
|
|
"name": "integration",
|
|
"options": [],
|
|
"query": {
|
|
"query": "label_values(alertmanager_notifications_total{integration=~\".*\"}, integration)",
|
|
"refId": "Prometheus-integration-Variable-Query"
|
|
},
|
|
"refresh": 2,
|
|
"regex": "",
|
|
"skipUrlSync": false,
|
|
"sort": 1,
|
|
"tagValuesQuery": "",
|
|
"tagsQuery": "",
|
|
"type": "query",
|
|
"useTags": false
|
|
}
|
|
]
|
|
},
|
|
"time": {
|
|
"from": "now-1h",
|
|
"to": "now"
|
|
},
|
|
"timepicker": {
|
|
"refresh_intervals": [
|
|
"5s",
|
|
"10s",
|
|
"30s",
|
|
"1m",
|
|
"5m",
|
|
"15m",
|
|
"30m",
|
|
"1h",
|
|
"2h",
|
|
"1d"
|
|
],
|
|
"time_options": [
|
|
"5m",
|
|
"15m",
|
|
"1h",
|
|
"6h",
|
|
"12h",
|
|
"24h",
|
|
"2d",
|
|
"7d",
|
|
"30d"
|
|
]
|
|
},
|
|
"timezone": "utc",
|
|
"title": "Alertmanager / Overview",
|
|
"uid": "alertmanager-overview",
|
|
"version": 1,
|
|
"weekStart": ""
|
|
} |