diff --git a/ee/scripts/helmcharts/manifests/dashboards/nginx-ingress.yaml b/ee/scripts/helmcharts/manifests/dashboards/nginx-ingress.yaml deleted file mode 100644 index 76a64aa43..000000000 --- a/ee/scripts/helmcharts/manifests/dashboards/nginx-ingress.yaml +++ /dev/null @@ -1,2458 +0,0 @@ -apiVersion: v1 -data: - nginx-ingress.json: |- - { - "__inputs": [ - { - "name": "DS_PROMETHEUS", - "label": "Prometheus", - "description": "", - "type": "datasource", - "pluginId": "prometheus", - "pluginName": "Prometheus" - } - ], - "__requires": [ - { - "type": "grafana", - "id": "grafana", - "name": "Grafana", - "version": "5.2.1" - }, - { - "type": "datasource", - "id": "prometheus", - "name": "Prometheus", - "version": "5.0.0" - }, - { - "type": "panel", - "id": "singlestat", - "name": "Singlestat", - "version": "5.0.0" - } - ], - "annotations": { - "list": [ - { - "builtIn": 1, - "datasource": "-- Grafana --", - "enable": true, - "hide": true, - "iconColor": "rgba(0, 211, 255, 1)", - "name": "Annotations & Alerts", - "type": "dashboard" - }, - { - "datasource": "${DS_PROMETHEUS}", - "enable": true, - "expr": "sum(changes(nginx_ingress_controller_config_last_reload_successful_timestamp_seconds{instance!=\"unknown\",controller_class=~\"$controller_class\",namespace=~\"$namespace\"}[30s])) by (controller_class)", - "hide": false, - "iconColor": "rgba(255, 96, 96, 1)", - "limit": 100, - "name": "Config Reloads", - "showIn": 0, - "step": "30s", - "tagKeys": "controller_class", - "tags": [], - "titleFormat": "Config Reloaded", - "type": "tags" - } - ] - }, - "editable": true, - "gnetId": null, - "graphTooltip": 0, - "iteration": 1534359654832, - "links": [], - "panels": [ - { - "cacheTimeout": null, - "colorBackground": false, - "colorValue": false, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "datasource": "${DS_PROMETHEUS}", - "format": "ops", - "gauge": { - "maxValue": 100, - "minValue": 0, - "show": false, - "thresholdLabels": false, - "thresholdMarkers": true - }, - "gridPos": { - "h": 3, - "w": 6, - "x": 0, - "y": 0 - }, - "id": 20, - "interval": null, - "links": [], - "mappingType": 1, - "mappingTypes": [ - { - "name": "value to text", - "value": 1 - }, - { - "name": "range to text", - "value": 2 - } - ], - "maxDataPoints": 100, - "nullPointMode": "connected", - "nullText": null, - "postfix": "", - "postfixFontSize": "50%", - "prefix": "", - "prefixFontSize": "50%", - "rangeMaps": [ - { - "from": "null", - "text": "N/A", - "to": "null" - } - ], - "sparkline": { - "fillColor": "rgba(31, 118, 189, 0.18)", - "full": true, - "lineColor": "rgb(31, 120, 193)", - "show": true - }, - "tableColumn": "", - "targets": [ - { - "expr": "round(sum(irate(nginx_ingress_controller_requests{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",namespace=~\"$namespace\"}[2m])), 0.001)", - "format": "time_series", - "intervalFactor": 1, - "refId": "A", - "step": 4 - } - ], - "thresholds": "", - "title": "Controller Request Volume", - "transparent": false, - "type": "singlestat", - "valueFontSize": "80%", - "valueMaps": [ - { - "op": "=", - "text": "N/A", - "value": "null" - } - ], - "valueName": "avg" - }, - { - "cacheTimeout": null, - "colorBackground": false, - "colorValue": false, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "datasource": "${DS_PROMETHEUS}", - "format": "none", - "gauge": { - "maxValue": 100, - "minValue": 0, - "show": false, - "thresholdLabels": false, - "thresholdMarkers": true - }, - "gridPos": { - "h": 3, - "w": 6, - "x": 6, - "y": 0 - }, - "id": 82, - "interval": null, - "links": [], - "mappingType": 1, - "mappingTypes": [ - { - "name": "value to text", - "value": 1 - }, - { - "name": "range to text", - "value": 2 - } - ], - "maxDataPoints": 100, - "nullPointMode": "connected", - "nullText": null, - "postfix": "", - "postfixFontSize": "50%", - "prefix": "", - "prefixFontSize": "50%", - "rangeMaps": [ - { - "from": "null", - "text": "N/A", - "to": "null" - } - ], - "sparkline": { - "fillColor": "rgba(31, 118, 189, 0.18)", - "full": true, - "lineColor": "rgb(31, 120, 193)", - "show": true - }, - "tableColumn": "", - "targets": [ - { - "expr": "sum(avg_over_time(nginx_ingress_controller_nginx_process_connections{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\",state=\"active\"}[2m]))", - "format": "time_series", - "instant": false, - "intervalFactor": 1, - "refId": "A", - "step": 4 - } - ], - "thresholds": "", - "title": "Controller Connections", - "transparent": false, - "type": "singlestat", - "valueFontSize": "80%", - "valueMaps": [ - { - "op": "=", - "text": "N/A", - "value": "null" - } - ], - "valueName": "avg" - }, - { - "cacheTimeout": null, - "colorBackground": false, - "colorValue": false, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "datasource": "${DS_PROMETHEUS}", - "format": "percentunit", - "gauge": { - "maxValue": 100, - "minValue": 80, - "show": false, - "thresholdLabels": false, - "thresholdMarkers": false - }, - "gridPos": { - "h": 3, - "w": 6, - "x": 12, - "y": 0 - }, - "id": 21, - "interval": null, - "links": [], - "mappingType": 1, - "mappingTypes": [ - { - "name": "value to text", - "value": 1 - }, - { - "name": "range to text", - "value": 2 - } - ], - "maxDataPoints": 100, - "nullPointMode": "connected", - "nullText": null, - "postfix": "", - "postfixFontSize": "50%", - "prefix": "", - "prefixFontSize": "50%", - "rangeMaps": [ - { - "from": "null", - "text": "N/A", - "to": "null" - } - ], - "sparkline": { - "fillColor": "rgba(31, 118, 189, 0.18)", - "full": true, - "lineColor": "rgb(31, 120, 193)", - "show": true - }, - "tableColumn": "", - "targets": [ - { - "expr": "sum(rate(nginx_ingress_controller_requests{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",namespace=~\"$namespace\",status!~\"[4-5].*\"}[2m])) / sum(rate(nginx_ingress_controller_requests{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",namespace=~\"$namespace\"}[2m]))", - "format": "time_series", - "intervalFactor": 1, - "refId": "A", - "step": 4 - } - ], - "thresholds": "95, 99, 99.5", - "title": "Controller Success Rate (non-4|5xx responses)", - "transparent": false, - "type": "singlestat", - "valueFontSize": "80%", - "valueMaps": [ - { - "op": "=", - "text": "N/A", - "value": "null" - } - ], - "valueName": "avg" - }, - { - "cacheTimeout": null, - "colorBackground": false, - "colorValue": false, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "datasource": "${DS_PROMETHEUS}", - "decimals": 0, - "format": "none", - "gauge": { - "maxValue": 100, - "minValue": 0, - "show": false, - "thresholdLabels": false, - "thresholdMarkers": true - }, - "gridPos": { - "h": 3, - "w": 3, - "x": 18, - "y": 0 - }, - "id": 81, - "interval": null, - "links": [], - "mappingType": 1, - "mappingTypes": [ - { - "name": "value to text", - "value": 1 - }, - { - "name": "range to text", - "value": 2 - } - ], - "maxDataPoints": 100, - "nullPointMode": "connected", - "nullText": null, - "postfix": "", - "postfixFontSize": "50%", - "prefix": "", - "prefixFontSize": "50%", - "rangeMaps": [ - { - "from": "null", - "text": "N/A", - "to": "null" - } - ], - "sparkline": { - "fillColor": "rgba(31, 118, 189, 0.18)", - "full": true, - "lineColor": "rgb(31, 120, 193)", - "show": true - }, - "tableColumn": "", - "targets": [ - { - "expr": "avg(irate(nginx_ingress_controller_success{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\"}[1m])) * 60", - "format": "time_series", - "instant": false, - "intervalFactor": 1, - "refId": "A", - "step": 4 - } - ], - "thresholds": "", - "title": "Config Reloads", - "transparent": false, - "type": "singlestat", - "valueFontSize": "80%", - "valueMaps": [ - { - "op": "=", - "text": "N/A", - "value": "null" - } - ], - "valueName": "total" - }, - { - "cacheTimeout": null, - "colorBackground": false, - "colorValue": false, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "datasource": "${DS_PROMETHEUS}", - "decimals": 0, - "format": "none", - "gauge": { - "maxValue": 100, - "minValue": 0, - "show": false, - "thresholdLabels": false, - "thresholdMarkers": true - }, - "gridPos": { - "h": 3, - "w": 3, - "x": 21, - "y": 0 - }, - "id": 83, - "interval": null, - "links": [], - "mappingType": 1, - "mappingTypes": [ - { - "name": "value to text", - "value": 1 - }, - { - "name": "range to text", - "value": 2 - } - ], - "maxDataPoints": 100, - "nullPointMode": "connected", - "nullText": null, - "postfix": "", - "postfixFontSize": "50%", - "prefix": "", - "prefixFontSize": "50%", - "rangeMaps": [ - { - "from": "null", - "text": "N/A", - "to": "null" - } - ], - "sparkline": { - "fillColor": "rgba(31, 118, 189, 0.18)", - "full": true, - "lineColor": "rgb(31, 120, 193)", - "show": true - }, - "tableColumn": "", - "targets": [ - { - "expr": "count(nginx_ingress_controller_config_last_reload_successful{controller_pod=~\"$controller\",controller_namespace=~\"$namespace\"} == 0)", - "format": "time_series", - "instant": true, - "intervalFactor": 1, - "refId": "A", - "step": 4 - } - ], - "thresholds": "", - "title": "Last Config Failed", - "transparent": false, - "type": "singlestat", - "valueFontSize": "80%", - "valueMaps": [ - { - "op": "=", - "text": "N/A", - "value": "null" - } - ], - "valueName": "avg" - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "decimals": 2, - "editable": true, - "error": false, - "fill": 1, - "grid": {}, - "gridPos": { - "h": 7, - "w": 12, - "x": 0, - "y": 3 - }, - "height": "200px", - "id": 86, - "isNew": true, - "legend": { - "alignAsTable": true, - "avg": true, - "current": false, - "hideEmpty": false, - "hideZero": true, - "max": false, - "min": false, - "rightSide": true, - "show": true, - "sideWidth": 300, - "sort": "current", - "sortDesc": true, - "total": false, - "values": true - }, - "lines": true, - "linewidth": 2, - "links": [], - "nullPointMode": "connected", - "percentage": false, - "pointradius": 5, - "points": false, - "renderer": "flot", - "repeat": null, - "repeatDirection": "h", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "round(sum(irate(nginx_ingress_controller_requests{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\",ingress=~\"$ingress\"}[2m])) by (ingress), 0.001)", - "format": "time_series", - "hide": false, - "instant": false, - "interval": "", - "intervalFactor": 1, - "legendFormat": "{{ ingress }}", - "metric": "network", - "refId": "A", - "step": 10 - } - ], - "thresholds": [], - "timeFrom": null, - "timeShift": null, - "title": "Ingress Request Volume", - "tooltip": { - "msResolution": false, - "shared": true, - "sort": 2, - "value_type": "cumulative" - }, - "transparent": false, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "reqps", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "Bps", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": false - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": { - "max - istio-proxy": "#890f02", - "max - master": "#bf1b00", - "max - prometheus": "#bf1b00" - }, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "decimals": 2, - "editable": false, - "error": false, - "fill": 0, - "grid": {}, - "gridPos": { - "h": 7, - "w": 12, - "x": 12, - "y": 3 - }, - "id": 87, - "isNew": true, - "legend": { - "alignAsTable": true, - "avg": true, - "current": false, - "hideEmpty": true, - "hideZero": false, - "max": false, - "min": false, - "rightSide": true, - "show": true, - "sideWidth": 300, - "sort": "avg", - "sortDesc": true, - "total": false, - "values": true - }, - "lines": true, - "linewidth": 2, - "links": [], - "nullPointMode": "connected", - "percentage": false, - "pointradius": 5, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "sum(rate(nginx_ingress_controller_requests{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",namespace=~\"$namespace\",ingress=~\"$ingress\",status!~\"[4-5].*\"}[2m])) by (ingress) / sum(rate(nginx_ingress_controller_requests{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",namespace=~\"$namespace\",ingress=~\"$ingress\"}[2m])) by (ingress)", - "format": "time_series", - "instant": false, - "interval": "10s", - "intervalFactor": 1, - "legendFormat": "{{ ingress }}", - "metric": "container_memory_usage:sort_desc", - "refId": "A", - "step": 10 - } - ], - "thresholds": [], - "timeFrom": null, - "timeShift": null, - "title": "Ingress Success Rate (non-4|5xx responses)", - "tooltip": { - "msResolution": false, - "shared": true, - "sort": 1, - "value_type": "cumulative" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "percentunit", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": false - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "decimals": 2, - "editable": true, - "error": false, - "fill": 1, - "grid": {}, - "gridPos": { - "h": 6, - "w": 8, - "x": 0, - "y": 10 - }, - "height": "200px", - "id": 32, - "isNew": true, - "legend": { - "alignAsTable": false, - "avg": true, - "current": true, - "max": false, - "min": false, - "rightSide": false, - "show": false, - "sideWidth": 200, - "sort": "current", - "sortDesc": true, - "total": false, - "values": true - }, - "lines": true, - "linewidth": 2, - "links": [], - "nullPointMode": "connected", - "percentage": false, - "pointradius": 5, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "sum (irate (nginx_ingress_controller_request_size_sum{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\"}[2m]))", - "format": "time_series", - "instant": false, - "interval": "10s", - "intervalFactor": 1, - "legendFormat": "Received", - "metric": "network", - "refId": "A", - "step": 10 - }, - { - "expr": "- sum (irate (nginx_ingress_controller_response_size_sum{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\"}[2m]))", - "format": "time_series", - "hide": false, - "interval": "10s", - "intervalFactor": 1, - "legendFormat": "Sent", - "metric": "network", - "refId": "B", - "step": 10 - } - ], - "thresholds": [], - "timeFrom": null, - "timeShift": null, - "title": "Network I/O pressure", - "tooltip": { - "msResolution": false, - "shared": true, - "sort": 0, - "value_type": "cumulative" - }, - "transparent": false, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "Bps", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "Bps", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": false - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": { - "max - istio-proxy": "#890f02", - "max - master": "#bf1b00", - "max - prometheus": "#bf1b00" - }, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "decimals": 2, - "editable": false, - "error": false, - "fill": 0, - "grid": {}, - "gridPos": { - "h": 6, - "w": 8, - "x": 8, - "y": 10 - }, - "id": 77, - "isNew": true, - "legend": { - "alignAsTable": true, - "avg": true, - "current": true, - "max": false, - "min": false, - "rightSide": false, - "show": false, - "sideWidth": 200, - "sort": "current", - "sortDesc": true, - "total": false, - "values": true - }, - "lines": true, - "linewidth": 2, - "links": [], - "nullPointMode": "connected", - "percentage": false, - "pointradius": 5, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "avg(nginx_ingress_controller_nginx_process_resident_memory_bytes{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\"}) ", - "format": "time_series", - "instant": false, - "interval": "10s", - "intervalFactor": 1, - "legendFormat": "nginx", - "metric": "container_memory_usage:sort_desc", - "refId": "A", - "step": 10 - } - ], - "thresholds": [], - "timeFrom": null, - "timeShift": null, - "title": "Average Memory Usage", - "tooltip": { - "msResolution": false, - "shared": true, - "sort": 2, - "value_type": "cumulative" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "bytes", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": false - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": { - "max - istio-proxy": "#890f02", - "max - master": "#bf1b00" - }, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "decimals": 3, - "editable": false, - "error": false, - "fill": 0, - "grid": {}, - "gridPos": { - "h": 6, - "w": 8, - "x": 16, - "y": 10 - }, - "height": "", - "id": 79, - "isNew": true, - "legend": { - "alignAsTable": true, - "avg": true, - "current": true, - "max": false, - "min": false, - "rightSide": false, - "show": false, - "sort": null, - "sortDesc": null, - "total": false, - "values": true - }, - "lines": true, - "linewidth": 2, - "links": [], - "nullPointMode": "connected", - "percentage": false, - "pointradius": 5, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "avg (rate (nginx_ingress_controller_nginx_process_cpu_seconds_total{controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\"}[2m])) ", - "format": "time_series", - "interval": "10s", - "intervalFactor": 1, - "legendFormat": "nginx", - "metric": "container_cpu", - "refId": "A", - "step": 10 - } - ], - "thresholds": [ - { - "colorMode": "critical", - "fill": true, - "line": true, - "op": "gt" - } - ], - "timeFrom": null, - "timeShift": null, - "title": "Average CPU Usage", - "tooltip": { - "msResolution": true, - "shared": true, - "sort": 2, - "value_type": "cumulative" - }, - "transparent": false, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "none", - "label": "cores", - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "columns": [], - "datasource": "${DS_PROMETHEUS}", - "fontSize": "100%", - "gridPos": { - "h": 8, - "w": 24, - "x": 0, - "y": 16 - }, - "hideTimeOverride": false, - "id": 75, - "links": [], - "pageSize": 7, - "repeat": null, - "repeatDirection": "h", - "scroll": true, - "showHeader": true, - "sort": { - "col": 1, - "desc": true - }, - "styles": [ - { - "alias": "Ingress", - "colorMode": null, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "dateFormat": "YYYY-MM-DD HH:mm:ss", - "decimals": 2, - "pattern": "ingress", - "preserveFormat": false, - "sanitize": false, - "thresholds": [], - "type": "string", - "unit": "short" - }, - { - "alias": "Requests", - "colorMode": null, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "dateFormat": "YYYY-MM-DD HH:mm:ss", - "decimals": 2, - "pattern": "Value #A", - "thresholds": [ - "" - ], - "type": "number", - "unit": "ops" - }, - { - "alias": "Errors", - "colorMode": null, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "dateFormat": "YYYY-MM-DD HH:mm:ss", - "decimals": 2, - "pattern": "Value #B", - "thresholds": [], - "type": "number", - "unit": "ops" - }, - { - "alias": "P50 Latency", - "colorMode": null, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "dateFormat": "YYYY-MM-DD HH:mm:ss", - "decimals": 0, - "link": false, - "pattern": "Value #C", - "thresholds": [], - "type": "number", - "unit": "dtdurations" - }, - { - "alias": "P90 Latency", - "colorMode": null, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "dateFormat": "YYYY-MM-DD HH:mm:ss", - "decimals": 0, - "pattern": "Value #D", - "thresholds": [], - "type": "number", - "unit": "dtdurations" - }, - { - "alias": "P99 Latency", - "colorMode": null, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "dateFormat": "YYYY-MM-DD HH:mm:ss", - "decimals": 0, - "pattern": "Value #E", - "thresholds": [], - "type": "number", - "unit": "dtdurations" - }, - { - "alias": "IN", - "colorMode": null, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "dateFormat": "YYYY-MM-DD HH:mm:ss", - "decimals": 2, - "pattern": "Value #F", - "thresholds": [ - "" - ], - "type": "number", - "unit": "Bps" - }, - { - "alias": "", - "colorMode": null, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "dateFormat": "YYYY-MM-DD HH:mm:ss", - "decimals": 2, - "pattern": "Time", - "thresholds": [], - "type": "hidden", - "unit": "short" - }, - { - "alias": "OUT", - "colorMode": null, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "dateFormat": "YYYY-MM-DD HH:mm:ss", - "decimals": 2, - "mappingType": 1, - "pattern": "Value #G", - "thresholds": [], - "type": "number", - "unit": "Bps" - } - ], - "targets": [ - { - "expr": "histogram_quantile(0.50, sum(rate(nginx_ingress_controller_request_duration_seconds_bucket{ingress!=\"\",controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\",ingress=~\"$ingress\"}[2m])) by (le, ingress))", - "format": "table", - "hide": false, - "instant": true, - "intervalFactor": 1, - "legendFormat": "{{ ingress }}", - "refId": "C" - }, - { - "expr": "histogram_quantile(0.90, sum(rate(nginx_ingress_controller_request_duration_seconds_bucket{ingress!=\"\",controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\",ingress=~\"$ingress\"}[2m])) by (le, ingress))", - "format": "table", - "hide": false, - "instant": true, - "intervalFactor": 1, - "legendFormat": "{{ ingress }}", - "refId": "D" - }, - { - "expr": "histogram_quantile(0.99, sum(rate(nginx_ingress_controller_request_duration_seconds_bucket{ingress!=\"\",controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\",ingress=~\"$ingress\"}[2m])) by (le, ingress))", - "format": "table", - "hide": false, - "instant": true, - "intervalFactor": 1, - "legendFormat": "{{ destination_service }}", - "refId": "E" - }, - { - "expr": "sum(irate(nginx_ingress_controller_request_size_sum{ingress!=\"\",controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\",ingress=~\"$ingress\"}[2m])) by (ingress)", - "format": "table", - "hide": false, - "instant": true, - "interval": "", - "intervalFactor": 1, - "legendFormat": "{{ ingress }}", - "refId": "F" - }, - { - "expr": "sum(irate(nginx_ingress_controller_response_size_sum{ingress!=\"\",controller_pod=~\"$controller\",controller_class=~\"$controller_class\",controller_namespace=~\"$namespace\",ingress=~\"$ingress\"}[2m])) by (ingress)", - "format": "table", - "instant": true, - "intervalFactor": 1, - "legendFormat": "{{ ingress }}", - "refId": "G" - } - ], - "timeFrom": null, - "title": "Ingress Percentile Response Times and Transfer Rates", - "transform": "table", - "transparent": false, - "type": "table" - }, - { - "columns": [ - { - "text": "Current", - "value": "current" - } - ], - "datasource": "${DS_PROMETHEUS}", - "fontSize": "100%", - "gridPos": { - "h": 8, - "w": 24, - "x": 0, - "y": 24 - }, - "height": "1024", - "id": 85, - "links": [], - "pageSize": 7, - "scroll": true, - "showHeader": true, - "sort": { - "col": 1, - "desc": false - }, - "styles": [ - { - "alias": "Time", - "dateFormat": "YYYY-MM-DD HH:mm:ss", - "pattern": "Time", - "type": "date" - }, - { - "alias": "TTL", - "colorMode": "cell", - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "dateFormat": "YYYY-MM-DD HH:mm:ss", - "decimals": 0, - "pattern": "Current", - "thresholds": [ - "0", - "691200" - ], - "type": "number", - "unit": "s" - }, - { - "alias": "", - "colorMode": null, - "colors": [ - "rgba(245, 54, 54, 0.9)", - "rgba(237, 129, 40, 0.89)", - "rgba(50, 172, 45, 0.97)" - ], - "decimals": 2, - "pattern": "/.*/", - "thresholds": [], - "type": "number", - "unit": "short" - } - ], - "targets": [ - { - "expr": "avg(nginx_ingress_controller_ssl_expire_time_seconds{kubernetes_pod_name=~\"$controller\",namespace=~\"$namespace\",ingress=~\"$ingress\"}) by (host) - time()", - "format": "time_series", - "intervalFactor": 1, - "legendFormat": "{{ host }}", - "metric": "gke_letsencrypt_cert_expiration", - "refId": "A", - "step": 1 - } - ], - "title": "Ingress Certificate Expiry", - "transform": "timeseries_aggregations", - "type": "table" - } - ], - "refresh": "5s", - "schemaVersion": 16, - "style": "dark", - "tags": [ - "nginx" - ], - "templating": { - "list": [ - { - "hide": 0, - "label": "datasource", - "name": "DS_PROMETHEUS", - "options": [], - "query": "prometheus", - "refresh": 1, - "regex": "", - "type": "datasource" - }, - { - "allValue": ".*", - "current": { - "text": "All", - "value": "$__all" - }, - "datasource": "${DS_PROMETHEUS}", - "hide": 0, - "includeAll": true, - "label": "Namespace", - "multi": false, - "name": "namespace", - "options": [], - "query": "label_values(nginx_ingress_controller_config_hash, controller_namespace)", - "refresh": 1, - "regex": "", - "sort": 0, - "tagValuesQuery": "", - "tags": [], - "tagsQuery": "", - "type": "query", - "useTags": false - }, - { - "allValue": ".*", - "current": { - "text": "All", - "value": "$__all" - }, - "datasource": "${DS_PROMETHEUS}", - "hide": 0, - "includeAll": true, - "label": "Controller Class", - "multi": false, - "name": "controller_class", - "options": [], - "query": "label_values(nginx_ingress_controller_config_hash{namespace=~\"$namespace\"}, controller_class) ", - "refresh": 1, - "regex": "", - "sort": 0, - "tagValuesQuery": "", - "tags": [], - "tagsQuery": "", - "type": "query", - "useTags": false - }, - { - "allValue": ".*", - "current": { - "text": "All", - "value": "$__all" - }, - "datasource": "${DS_PROMETHEUS}", - "hide": 0, - "includeAll": true, - "label": "Controller", - "multi": false, - "name": "controller", - "options": [], - "query": "label_values(nginx_ingress_controller_config_hash{namespace=~\"$namespace\",controller_class=~\"$controller_class\"}, controller_pod) ", - "refresh": 1, - "regex": "", - "sort": 0, - "tagValuesQuery": "", - "tags": [], - "tagsQuery": "", - "type": "query", - "useTags": false - }, - { - "allValue": ".*", - "current": { - "tags": [], - "text": "All", - "value": "$__all" - }, - "datasource": "${DS_PROMETHEUS}", - "hide": 0, - "includeAll": true, - "label": "Ingress", - "multi": false, - "name": "ingress", - "options": [], - "query": "label_values(nginx_ingress_controller_requests{namespace=~\"$namespace\",controller_class=~\"$controller_class\",controller_pod=~\"$controller\"}, ingress) ", - "refresh": 1, - "regex": "", - "sort": 2, - "tagValuesQuery": "", - "tags": [], - "tagsQuery": "", - "type": "query", - "useTags": false - } - ] - }, - "time": { - "from": "now-1h", - "to": "now" - }, - "timepicker": { - "refresh_intervals": [ - "5s", - "10s", - "30s", - "2m", - "5m", - "15m", - "30m", - "1h", - "2h", - "1d" - ], - "time_options": [ - "5m", - "15m", - "1h", - "6h", - "12h", - "24h", - "2d", - "7d", - "30d" - ] - }, - "timezone": "browser", - "title": "NGINX Ingress controller", - "uid": "nginx", - "version": 1 - } - nginx-performace.json: |- - { - "__inputs": [ - { - "name": "DS_PROMETHEUS", - "label": "Prometheus", - "description": "", - "type": "datasource", - "pluginId": "prometheus", - "pluginName": "Prometheus" - } - ], - "__requires": [ - { - "type": "grafana", - "id": "grafana", - "name": "Grafana", - "version": "6.6.0" - }, - { - "type": "panel", - "id": "graph", - "name": "Graph", - "version": "" - }, - { - "type": "datasource", - "id": "prometheus", - "name": "Prometheus", - "version": "1.0.0" - } - ], - "annotations": { - "list": [ - { - "builtIn": 1, - "datasource": "-- Grafana --", - "enable": true, - "hide": true, - "iconColor": "rgba(0, 211, 255, 1)", - "name": "Annotations & Alerts", - "type": "dashboard" - } - ] - }, - "description": "", - "editable": true, - "gnetId": 9614, - "graphTooltip": 1, - "id": null, - "iteration": 1582146566338, - "links": [], - "panels": [ - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "description": "Total time taken for nginx and upstream servers to process a request and send a response", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 8, - "w": 12, - "x": 0, - "y": 0 - }, - "hiddenSeries": false, - "id": 91, - "legend": { - "avg": false, - "current": false, - "max": false, - "min": false, - "show": true, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null", - "options": { - "dataLinks": [] - }, - "percentage": false, - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "histogram_quantile(\n 0.5,\n sum by (le)(\n rate(\n nginx_ingress_controller_request_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", - "interval": "", - "legendFormat": ".5", - "refId": "D" - }, - { - "expr": "histogram_quantile(\n 0.95,\n sum by (le)(\n rate(\n nginx_ingress_controller_request_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", - "interval": "", - "legendFormat": ".95", - "refId": "B" - }, - { - "expr": "histogram_quantile(\n 0.99,\n sum by (le)(\n rate(\n nginx_ingress_controller_request_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", - "interval": "", - "legendFormat": ".99", - "refId": "A" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Total request handling time", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "s", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "description": "The time spent on receiving the response from the upstream server", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 8, - "w": 12, - "x": 12, - "y": 0 - }, - "hiddenSeries": false, - "id": 94, - "legend": { - "avg": false, - "current": false, - "max": false, - "min": false, - "show": true, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null", - "options": { - "dataLinks": [] - }, - "percentage": false, - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "histogram_quantile(\n 0.5,\n sum by (le)(\n rate(\n nginx_ingress_controller_response_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", - "instant": false, - "interval": "", - "intervalFactor": 1, - "legendFormat": ".5", - "refId": "D" - }, - { - "expr": "histogram_quantile(\n 0.95,\n sum by (le)(\n rate(\n nginx_ingress_controller_response_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", - "interval": "", - "legendFormat": ".95", - "refId": "B" - }, - { - "expr": "histogram_quantile(\n 0.99,\n sum by (le)(\n rate(\n nginx_ingress_controller_response_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", - "interval": "", - "legendFormat": ".99", - "refId": "A" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Upstream response time", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "s", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 8, - "w": 12, - "x": 0, - "y": 8 - }, - "hiddenSeries": false, - "id": 93, - "legend": { - "alignAsTable": true, - "avg": false, - "current": false, - "max": false, - "min": false, - "rightSide": true, - "show": true, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null", - "options": { - "dataLinks": [] - }, - "percentage": false, - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": " sum by (path)(\n rate(\n nginx_ingress_controller_request_duration_seconds_count{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n", - "interval": "", - "intervalFactor": 1, - "legendFormat": "{{ path }}", - "refId": "A" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Request volume by Path", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "reqps", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "description": "For each path observed, its median upstream response time", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 8, - "w": 12, - "x": 12, - "y": 8 - }, - "hiddenSeries": false, - "id": 98, - "legend": { - "alignAsTable": true, - "avg": false, - "current": false, - "max": false, - "min": false, - "rightSide": true, - "show": true, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null", - "options": { - "dataLinks": [] - }, - "percentage": false, - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "histogram_quantile(\n .5,\n sum by (le, path)(\n rate(\n nginx_ingress_controller_response_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", - "interval": "", - "intervalFactor": 1, - "legendFormat": "{{ path }}", - "refId": "A" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Median upstream response time by Path", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "s", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "description": "Percentage of 4xx and 5xx responses among all responses.", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 8, - "w": 12, - "x": 0, - "y": 16 - }, - "hiddenSeries": false, - "id": 100, - "legend": { - "alignAsTable": true, - "avg": false, - "current": false, - "max": false, - "min": false, - "rightSide": true, - "show": true, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null as zero", - "options": { - "dataLinks": [] - }, - "percentage": false, - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "sum by (path) (rate(nginx_ingress_controller_request_duration_seconds_count{\n ingress =~ \"$ingress\",\n status =~ \"[4-5].*\"\n}[1m])) / sum by (path) (rate(nginx_ingress_controller_request_duration_seconds_count{\n ingress =~ \"$ingress\",\n}[1m]))", - "interval": "", - "intervalFactor": 1, - "legendFormat": "{{ path }}", - "refId": "A" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Response error rate by Path", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "percentunit", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "description": "For each path observed, the sum of upstream request time", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 8, - "w": 12, - "x": 12, - "y": 16 - }, - "hiddenSeries": false, - "id": 102, - "legend": { - "alignAsTable": true, - "avg": false, - "current": false, - "max": false, - "min": false, - "rightSide": true, - "show": true, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null", - "options": { - "dataLinks": [] - }, - "percentage": false, - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "sum by (path) (rate(nginx_ingress_controller_response_duration_seconds_sum{ingress =~ \"$ingress\"}[1m]))", - "interval": "", - "intervalFactor": 1, - "legendFormat": "{{ path }}", - "refId": "A" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Upstream time consumed by Path", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "s", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 8, - "w": 12, - "x": 0, - "y": 24 - }, - "hiddenSeries": false, - "id": 101, - "legend": { - "alignAsTable": true, - "avg": false, - "current": false, - "max": false, - "min": false, - "rightSide": true, - "show": true, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null", - "options": { - "dataLinks": [] - }, - "percentage": false, - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": " sum (\n rate(\n nginx_ingress_controller_request_duration_seconds_count{\n ingress =~ \"$ingress\",\n status =~\"[4-5].*\",\n }[1m]\n )\n ) by(path, status)\n", - "interval": "", - "intervalFactor": 1, - "legendFormat": "{{ path }} {{ status }}", - "refId": "A" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Response error volume by Path", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "reqps", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 8, - "w": 12, - "x": 12, - "y": 24 - }, - "hiddenSeries": false, - "id": 99, - "legend": { - "alignAsTable": true, - "avg": false, - "current": false, - "max": false, - "min": false, - "rightSide": true, - "show": true, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null", - "options": { - "dataLinks": [] - }, - "percentage": false, - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "sum (\n rate (\n nginx_ingress_controller_response_size_sum {\n ingress =~ \"$ingress\",\n }[1m]\n )\n) by (path) / sum (\n rate(\n nginx_ingress_controller_response_size_count {\n ingress =~ \"$ingress\",\n }[1m]\n )\n) by (path)\n", - "hide": false, - "instant": false, - "interval": "", - "intervalFactor": 1, - "legendFormat": "{{ path }}", - "refId": "D" - }, - { - "expr": " sum (rate(nginx_ingress_controller_response_size_bucket{\n ingress =~ \"$ingress\",\n }[1m])) by (le)\n", - "hide": true, - "legendFormat": "{{le}}", - "refId": "A" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Average response size by Path", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "decbytes", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": "${DS_PROMETHEUS}", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 8, - "w": 12, - "x": 0, - "y": 32 - }, - "hiddenSeries": false, - "id": 96, - "legend": { - "avg": false, - "current": false, - "max": false, - "min": false, - "show": true, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null", - "options": { - "dataLinks": [] - }, - "percentage": false, - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "expr": "sum (\n rate(\n nginx_ingress_controller_ingress_upstream_latency_seconds_sum {\n ingress =~ \"$ingress\",\n }[1m]\n)) / sum (\n rate(\n nginx_ingress_controller_ingress_upstream_latency_seconds_count {\n ingress =~ \"$ingress\",\n }[1m]\n )\n)\n", - "hide": false, - "instant": false, - "interval": "", - "intervalFactor": 1, - "legendFormat": "average", - "refId": "B" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Upstream service latency", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "s", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - } - ], - "refresh": "30s", - "schemaVersion": 22, - "style": "dark", - "tags": [ - "nginx" - ], - "templating": { - "list": [ - { - "hide": 0, - "label": "datasource", - "name": "DS_PROMETHEUS", - "options": [], - "query": "prometheus", - "refresh": 1, - "regex": "", - "type": "datasource" - }, - { - "allValue": ".*", - "current": {}, - "datasource": "${DS_PROMETHEUS}", - "definition": "label_values(nginx_ingress_controller_requests, ingress) ", - "hide": 0, - "includeAll": true, - "label": "Service Ingress", - "multi": false, - "name": "ingress", - "options": [], - "query": "label_values(nginx_ingress_controller_requests, ingress) ", - "refresh": 1, - "regex": "", - "skipUrlSync": false, - "sort": 2, - "tagValuesQuery": "", - "tags": [], - "tagsQuery": "", - "type": "query", - "useTags": false - } - ] - }, - "time": { - "from": "now-15m", - "to": "now" - }, - "timepicker": { - "refresh_intervals": [ - "5s", - "10s", - "30s", - "2m", - "5m", - "15m", - "30m", - "1h", - "2h", - "1d" - ], - "time_options": [ - "5m", - "15m", - "1h", - "6h", - "12h", - "24h", - "2d", - "7d", - "30d" - ] - }, - "timezone": "browser", - "title": "Request Handling Performance", - "uid": "4GFbkOsZk", - "version": 1 - } - -kind: ConfigMap -metadata: - labels: - app: kube-prometheus-stack-grafana - grafana_dashboard: "1" - release: monitoring - name: nginx-igress - namespace: monitoring - diff --git a/ee/scripts/helmcharts/manifests/dashboards/openreplay-components.yaml b/ee/scripts/helmcharts/manifests/dashboards/openreplay-components.yaml deleted file mode 100644 index d8840c4a9..000000000 --- a/ee/scripts/helmcharts/manifests/dashboards/openreplay-components.yaml +++ /dev/null @@ -1,1237 +0,0 @@ - -apiVersion: v1 -data: - openreplay-components.json: |- - { - "__inputs": [], - "__requires": [ - { - "type": "grafana", - "id": "grafana", - "name": "Grafana", - "version": "8.2.0" - }, - { - "type": "panel", - "id": "graph", - "name": "Graph (old)", - "version": "" - }, - { - "type": "panel", - "id": "heatmap", - "name": "Heatmap", - "version": "" - }, - { - "type": "panel", - "id": "stat", - "name": "Stat", - "version": "" - } - ], - "annotations": { - "list": [ - { - "builtIn": 1, - "datasource": "-- 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": true, - "fiscalYearStartMonth": 0, - "gnetId": null, - "graphTooltip": 0, - "id": null, - "iteration": 1656567623206, - "links": [], - "liveNow": false, - "panels": [ - { - "datasource": null, - "description": "Number of all requests that have been processed / service", - "fieldConfig": { - "defaults": { - "color": { - "mode": "thresholds" - }, - "mappings": [], - "thresholds": { - "mode": "absolute", - "steps": [ - { - "color": "dark-red", - "value": null - }, - { - "color": "green", - "value": 80 - } - ] - }, - "unit": "short" - }, - "overrides": [] - }, - "id": 18, - "options": { - "colorMode": "value", - "graphMode": "area", - "justifyMode": "auto", - "orientation": "auto", - "reduceOptions": { - "calcs": [ - "lastNotNull" - ], - "fields": "", - "values": false - }, - "text": {}, - "textMode": "auto" - }, - "pluginVersion": "8.2.0", - "targets": [ - { - "exemplar": true, - "expr": "sum(increase(requests_total[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "title": "Requests total / ${time}", - "type": "stat" - }, - { - "datasource": null, - "description": "Number of all messages that have been processed / service", - "fieldConfig": { - "defaults": { - "color": { - "mode": "thresholds" - }, - "mappings": [], - "thresholds": { - "mode": "absolute", - "steps": [ - { - "color": "dark-red", - "value": null - }, - { - "color": "light-green", - "value": 80 - } - ] - }, - "unit": "short" - }, - "overrides": [] - }, - "gridPos": { - "h": 3, - "w": 6, - "x": 6, - "y": 0 - }, - "id": 22, - "options": { - "colorMode": "value", - "graphMode": "area", - "justifyMode": "auto", - "orientation": "auto", - "reduceOptions": { - "calcs": [ - "lastNotNull" - ], - "fields": "", - "values": false - }, - "text": {}, - "textMode": "auto" - }, - "pluginVersion": "8.2.0", - "targets": [ - { - "exemplar": true, - "expr": "sum(increase(messages_total[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "title": "Messages processed / ${time}", - "type": "stat" - }, - { - "datasource": null, - "description": "Number of messages that have been saved to session files", - "fieldConfig": { - "defaults": { - "color": { - "mode": "thresholds" - }, - "mappings": [], - "thresholds": { - "mode": "absolute", - "steps": [ - { - "color": "dark-red", - "value": null - }, - { - "color": "light-green", - "value": 80 - } - ] - }, - "unit": "short" - }, - "overrides": [] - }, - "gridPos": { - "h": 3, - "w": 6, - "x": 12, - "y": 0 - }, - "id": 24, - "options": { - "colorMode": "value", - "graphMode": "area", - "justifyMode": "auto", - "orientation": "auto", - "reduceOptions": { - "calcs": [ - "lastNotNull" - ], - "fields": "", - "values": false - }, - "text": {}, - "textMode": "auto" - }, - "pluginVersion": "8.2.0", - "targets": [ - { - "exemplar": true, - "expr": "sum(increase(messages_saved[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "title": "Messages saved / ${time}", - "type": "stat" - }, - { - "datasource": null, - "description": "Number of all sessions that have been compressed and saved to s3", - "fieldConfig": { - "defaults": { - "color": { - "mode": "thresholds" - }, - "mappings": [], - "thresholds": { - "mode": "absolute", - "steps": [ - { - "color": "dark-red", - "value": null - }, - { - "color": "light-green", - "value": 80 - } - ] - } - }, - "overrides": [] - }, - "gridPos": { - "h": 3, - "w": 6, - "x": 18, - "y": 0 - }, - "id": 30, - "options": { - "colorMode": "value", - "graphMode": "area", - "justifyMode": "auto", - "orientation": "auto", - "reduceOptions": { - "calcs": [ - "lastNotNull" - ], - "fields": "", - "values": false - }, - "text": {}, - "textMode": "auto" - }, - "pluginVersion": "8.2.0", - "targets": [ - { - "exemplar": true, - "expr": "sum(increase(sessions_total{container_name=\"saas-storage\"}[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "title": "Sessions stored in s3 / ${time}", - "type": "stat" - }, - { - "collapsed": false, - "datasource": null, - "gridPos": { - "h": 1, - "w": 24, - "x": 0, - "y": 3 - }, - "id": 28, - "panels": [], - "title": "Storage", - "type": "row" - }, - { - "cards": { - "cardPadding": null, - "cardRound": null - }, - "color": { - "cardColor": "#FADE2A", - "colorScale": "sqrt", - "colorScheme": "interpolateGreys", - "exponent": 0.5, - "mode": "opacity" - }, - "dataFormat": "timeseries", - "datasource": null, - "description": "Sizes of saved sessions (before compression)\n", - "gridPos": { - "h": 8, - "w": 12, - "x": 0, - "y": 4 - }, - "heatmap": {}, - "hideZeroBuckets": false, - "highlightCards": true, - "id": 32, - "legend": { - "show": false - }, - "pluginVersion": "7.5.11", - "reverseYBuckets": false, - "targets": [ - { - "exemplar": true, - "expr": "sum(rate(sessions_size_sum[${time}]))/sum(rate(sessions_size_count[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "title": "Session size histogram / ${time}", - "tooltip": { - "show": true, - "showHistogram": false - }, - "type": "heatmap", - "xAxis": { - "show": true - }, - "xBucketNumber": null, - "xBucketSize": null, - "yAxis": { - "decimals": null, - "format": "decbytes", - "logBase": 1, - "max": null, - "min": null, - "show": true, - "splitFactor": null - }, - "yBucketBound": "auto", - "yBucketNumber": null, - "yBucketSize": null - }, - { - "cards": { - "cardPadding": null, - "cardRound": null - }, - "color": { - "cardColor": "#FADE2A", - "colorScale": "sqrt", - "colorScheme": "interpolateOranges", - "exponent": 0.5, - "mode": "opacity" - }, - "dataFormat": "timeseries", - "datasource": null, - "description": "Durations of archiving process for all handled sessions", - "gridPos": { - "h": 8, - "w": 12, - "x": 12, - "y": 4 - }, - "heatmap": {}, - "hideZeroBuckets": false, - "highlightCards": true, - "id": 34, - "legend": { - "show": false - }, - "pluginVersion": "7.5.11", - "reverseYBuckets": false, - "targets": [ - { - "exemplar": true, - "expr": "sum(rate(archiving_duration_sum[${time}]))/sum(rate(archiving_duration_count[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "title": "Archiving duration histogram / ${time}", - "tooltip": { - "show": true, - "showHistogram": false - }, - "type": "heatmap", - "xAxis": { - "show": true - }, - "xBucketNumber": null, - "xBucketSize": null, - "yAxis": { - "decimals": null, - "format": "ms", - "logBase": 1, - "max": null, - "min": null, - "show": true, - "splitFactor": null - }, - "yBucketBound": "auto", - "yBucketNumber": null, - "yBucketSize": null - }, - { - "collapsed": false, - "datasource": null, - "gridPos": { - "h": 1, - "w": 24, - "x": 0, - "y": 12 - }, - "id": 20, - "panels": [], - "title": "Sink", - "type": "row" - }, - { - "cards": { - "cardPadding": null, - "cardRound": null - }, - "color": { - "cardColor": "#FADE2A", - "colorScale": "sqrt", - "colorScheme": "interpolateOranges", - "exponent": 0.5, - "mode": "opacity" - }, - "dataFormat": "timeseries", - "datasource": null, - "description": "Sizes of saved messages\n", - "gridPos": { - "h": 8, - "w": 12, - "x": 0, - "y": 13 - }, - "heatmap": {}, - "hideZeroBuckets": false, - "highlightCards": true, - "id": 26, - "legend": { - "show": false - }, - "pluginVersion": "7.5.11", - "reverseYBuckets": false, - "targets": [ - { - "exemplar": true, - "expr": "sum(rate(messages_size_sum[${time}])) / sum(rate(messages_size_count[${time}]))", - "hide": false, - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "title": "Message size histogram / ${time}", - "tooltip": { - "show": true, - "showHistogram": false - }, - "type": "heatmap", - "xAxis": { - "show": true - }, - "xBucketNumber": null, - "xBucketSize": null, - "yAxis": { - "decimals": null, - "format": "decbytes", - "logBase": 1, - "max": null, - "min": null, - "show": true, - "splitFactor": null - }, - "yBucketBound": "auto", - "yBucketNumber": null, - "yBucketSize": null - }, - { - "collapsed": false, - "datasource": null, - "gridPos": { - "h": 1, - "w": 24, - "x": 0, - "y": 21 - }, - "id": 12, - "panels": [], - "title": "Http", - "type": "row" - }, - { - "cards": { - "cardPadding": null, - "cardRound": null - }, - "color": { - "cardColor": "#FADE2A", - "colorScale": "sqrt", - "colorScheme": "interpolateOranges", - "exponent": 0.5, - "mode": "opacity" - }, - "dataFormat": "timeseries", - "datasource": null, - "description": "Sizes of request’s body for start and ingest endpoints", - "gridPos": { - "h": 8, - "w": 12, - "x": 0, - "y": 22 - }, - "heatmap": {}, - "hideZeroBuckets": false, - "highlightCards": true, - "id": 14, - "legend": { - "show": false - }, - "pluginVersion": "7.5.11", - "reverseYBuckets": false, - "targets": [ - { - "exemplar": true, - "expr": "sum(rate(requests_body_size_sum[${time}]))/sum(rate(requests_body_size_count[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "timeFrom": null, - "timeShift": null, - "title": "Request body size / ${time}", - "tooltip": { - "show": true, - "showHistogram": false - }, - "type": "heatmap", - "xAxis": { - "show": true - }, - "xBucketNumber": null, - "xBucketSize": null, - "yAxis": { - "decimals": null, - "format": "decbytes", - "logBase": 1, - "max": null, - "min": null, - "show": true, - "splitFactor": null - }, - "yBucketBound": "auto", - "yBucketNumber": null, - "yBucketSize": null - }, - { - "cards": { - "cardPadding": null, - "cardRound": null - }, - "color": { - "cardColor": "#FADE2A", - "colorScale": "sqrt", - "colorScheme": "interpolateOranges", - "exponent": 0.5, - "mode": "opacity" - }, - "dataFormat": "timeseries", - "datasource": null, - "description": "Request processing duration for all endpoints", - "gridPos": { - "h": 8, - "w": 12, - "x": 12, - "y": 22 - }, - "heatmap": {}, - "hideZeroBuckets": false, - "highlightCards": true, - "id": 16, - "legend": { - "show": false - }, - "pluginVersion": "7.5.11", - "reverseYBuckets": false, - "targets": [ - { - "exemplar": true, - "expr": "sum(rate(requests_duration_sum[${time}]))/sum(rate(requests_duration_count[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "title": "Request duration / ${time}", - "tooltip": { - "show": true, - "showHistogram": false - }, - "type": "heatmap", - "xAxis": { - "show": true - }, - "xBucketNumber": null, - "xBucketSize": null, - "yAxis": { - "decimals": null, - "format": "ms", - "logBase": 1, - "max": null, - "min": null, - "show": true, - "splitFactor": null - }, - "yBucketBound": "auto", - "yBucketNumber": null, - "yBucketSize": null - }, - { - "collapsed": false, - "datasource": null, - "gridPos": { - "h": 1, - "w": 24, - "x": 0, - "y": 30 - }, - "id": 6, - "panels": [], - "title": "ender", - "type": "row" - }, - { - "aliasColors": { - "Value": "green" - }, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": null, - "description": "Number of active sessions (unfinished sessions)\n", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 8, - "w": 12, - "x": 0, - "y": 31 - }, - "hiddenSeries": false, - "id": 8, - "legend": { - "avg": false, - "current": false, - "max": false, - "min": false, - "show": false, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null", - "options": { - "alertThreshold": true - }, - "percentage": false, - "pluginVersion": "8.2.0", - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "exemplar": true, - "expr": "sum(sessions_active)", - "format": "time_series", - "interval": "", - "legendFormat": "{{ pod }}", - "refId": "A" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Active sessions", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": null, - "description": "Number of all sessions that have been processed / service", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 8, - "w": 12, - "x": 12, - "y": 31 - }, - "hiddenSeries": false, - "id": 10, - "legend": { - "avg": false, - "current": false, - "max": false, - "min": false, - "show": false, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null", - "options": { - "alertThreshold": true - }, - "percentage": false, - "pluginVersion": "8.2.0", - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "exemplar": true, - "expr": "sum(increase(sessions_total{container_name=\"saas-ender\"}[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Sessions processed / ${time}", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "collapsed": false, - "datasource": null, - "gridPos": { - "h": 1, - "w": 24, - "x": 0, - "y": 39 - }, - "id": 4, - "panels": [], - "title": "Assets", - "type": "row" - }, - { - "aliasColors": {}, - "bars": false, - "dashLength": 10, - "dashes": false, - "datasource": null, - "description": "Number of all assets that have been processed / service", - "fill": 1, - "fillGradient": 0, - "gridPos": { - "h": 9, - "w": 12, - "x": 0, - "y": 40 - }, - "hiddenSeries": false, - "id": 2, - "legend": { - "avg": false, - "current": false, - "max": false, - "min": false, - "show": false, - "total": false, - "values": false - }, - "lines": true, - "linewidth": 1, - "nullPointMode": "null", - "options": { - "alertThreshold": true - }, - "percentage": false, - "pluginVersion": "8.2.0", - "pointradius": 2, - "points": false, - "renderer": "flot", - "seriesOverrides": [], - "spaceLength": 10, - "stack": false, - "steppedLine": false, - "targets": [ - { - "exemplar": true, - "expr": "sum(increase(assets_total[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "thresholds": [], - "timeFrom": null, - "timeRegions": [], - "timeShift": null, - "title": "Assets processed / ${time}", - "tooltip": { - "shared": true, - "sort": 0, - "value_type": "individual" - }, - "type": "graph", - "xaxis": { - "buckets": null, - "mode": "time", - "name": null, - "show": true, - "values": [] - }, - "yaxes": [ - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - }, - { - "format": "short", - "label": null, - "logBase": 1, - "max": null, - "min": null, - "show": true - } - ], - "yaxis": { - "align": false, - "alignLevel": null - } - }, - { - "collapsed": false, - "datasource": null, - "gridPos": { - "h": 1, - "w": 24, - "x": 0, - "y": 49 - }, - "id": 36, - "panels": [], - "title": "Db", - "type": "row" - }, - { - "cards": { - "cardPadding": null, - "cardRound": null - }, - "color": { - "cardColor": "#FADE2A", - "colorScale": "sqrt", - "colorScheme": "interpolateOranges", - "exponent": 0.5, - "mode": "opacity" - }, - "dataFormat": "timeseries", - "datasource": null, - "description": "Size of batch in bytes before execution in PG", - "gridPos": { - "h": 8, - "w": 12, - "x": 0, - "y": 50 - }, - "heatmap": {}, - "hideZeroBuckets": false, - "highlightCards": true, - "id": 38, - "legend": { - "show": false - }, - "reverseYBuckets": false, - "targets": [ - { - "exemplar": true, - "expr": "sum(rate(batch_size_bytes_sum[${time}]))/sum(rate(batch_size_bytes_count[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "title": "Batch size / ${time}", - "tooltip": { - "show": true, - "showHistogram": false - }, - "type": "heatmap", - "xAxis": { - "show": true - }, - "xBucketNumber": null, - "xBucketSize": null, - "yAxis": { - "decimals": null, - "format": "decbytes", - "logBase": 1, - "max": null, - "min": null, - "show": true, - "splitFactor": null - }, - "yBucketBound": "auto", - "yBucketNumber": null, - "yBucketSize": null - }, - { - "cards": { - "cardPadding": null, - "cardRound": null - }, - "color": { - "cardColor": "#FADE2A", - "colorScale": "sqrt", - "colorScheme": "interpolateOranges", - "exponent": 0.5, - "mode": "opacity" - }, - "dataFormat": "timeseries", - "datasource": null, - "description": "Number of sql requests before execution in PG", - "gridPos": { - "h": 8, - "w": 12, - "x": 12, - "y": 50 - }, - "heatmap": {}, - "hideZeroBuckets": false, - "highlightCards": true, - "id": 39, - "legend": { - "show": false - }, - "reverseYBuckets": false, - "targets": [ - { - "exemplar": true, - "expr": "sum(rate(batch_size_lines_sum[${time}]))/sum(rate(batch_size_lines_count[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "title": "Number of sql requests before PG execution / ${time}", - "tooltip": { - "show": true, - "showHistogram": false - }, - "type": "heatmap", - "xAxis": { - "show": true - }, - "xBucketNumber": null, - "xBucketSize": null, - "yAxis": { - "decimals": null, - "format": "short", - "logBase": 1, - "max": null, - "min": null, - "show": true, - "splitFactor": null - }, - "yBucketBound": "auto", - "yBucketNumber": null, - "yBucketSize": null - }, - { - "cards": { - "cardPadding": null, - "cardRound": null - }, - "color": { - "cardColor": "#FADE2A", - "colorScale": "sqrt", - "colorScheme": "interpolateOranges", - "exponent": 0.5, - "mode": "opacity" - }, - "dataFormat": "timeseries", - "datasource": null, - "description": "Duration of SQL request", - "gridPos": { - "h": 8, - "w": 12, - "x": 0, - "y": 58 - }, - "heatmap": {}, - "hideZeroBuckets": false, - "highlightCards": true, - "id": 40, - "legend": { - "show": false - }, - "reverseYBuckets": false, - "targets": [ - { - "exemplar": true, - "expr": "sum(rate(sql_request_time_sum[${time}]))/sum(rate(sql_request_time_count[${time}]))", - "interval": "", - "legendFormat": "", - "refId": "A" - } - ], - "title": "Duration of sql request / ${time}", - "tooltip": { - "show": true, - "showHistogram": false - }, - "type": "heatmap", - "xAxis": { - "show": true - }, - "xBucketNumber": null, - "xBucketSize": null, - "yAxis": { - "decimals": null, - "format": "ms", - "logBase": 1, - "max": null, - "min": null, - "show": true, - "splitFactor": null - }, - "yBucketBound": "auto", - "yBucketNumber": null, - "yBucketSize": null - } - ], - "refresh": "", - "schemaVersion": 31, - "style": "dark", - "tags": [], - "templating": { - "list": [ - { - "auto": true, - "auto_count": 30, - "auto_min": "1m", - "current": { - "selected": false, - "text": "auto", - "value": "$__auto_interval_time" - }, - "description": null, - "error": null, - "hide": 1, - "label": "Data sampling time", - "name": "time", - "options": [ - { - "selected": true, - "text": "auto", - "value": "$__auto_interval_time" - }, - { - "selected": false, - "text": "1m", - "value": "1m" - }, - { - "selected": false, - "text": "2m", - "value": "2m" - }, - { - "selected": false, - "text": "5m", - "value": "5m" - }, - { - "selected": false, - "text": "10m", - "value": "10m" - } - ], - "query": "1m,2m,5m,10m", - "queryValue": "", - "refresh": 2, - "skipUrlSync": false, - "type": "interval" - } - ] - }, - "time": { - "from": "now-6h", - "to": "now" - }, - "timepicker": {}, - "timezone": "", - "title": "Openreplay Components Dashboard", - "uid": "aq_tupq7k", - "version": 31 - } - -kind: ConfigMap -metadata: - labels: - app: kube-prometheus-stack-grafana - grafana_dashboard: "1" - release: monitoring - name: openreplay-components - namespace: monitoring - diff --git a/ee/scripts/helmcharts/manifests/monitoring.yaml b/ee/scripts/helmcharts/manifests/monitoring.yaml deleted file mode 100644 index ce5640aa6..000000000 --- a/ee/scripts/helmcharts/manifests/monitoring.yaml +++ /dev/null @@ -1,168 +0,0 @@ -################################################## -## Update the following values -## -## For example, -# domainName: &domainName openreplay.supercompany.com -################################################## - -domainName: &domainName "changeme.mycorp.org" -grafanaAdminPassword: &adminpass "changeNeGrafanaAdminPassword" -slackWebhookUrl: &slackwebhook "https://hooks.slack.com/services/xxxx/xxxx/xxxxx" # Slack webhook token for sending alerts -slackChannel: &slackchannel "changeMeAlertsChannel" # Name of the channel, to alerts to be delivered to. - - -######################################################## -## Custom configuration for Monitoring and logging stack -######################################################## - -fullnameOverride: "openreplay" -grafana: - adminPassword: *adminpass - env: - GF_SERVER_ROOT_URL: http://grafana.local.com/grafana - additionalDataSources: - - name: loki - editable: true - type: loki - url: http://loki.logging:3100 - plugins: - - grafana-piechart-panel - - vertamedia-clickhouse-datasource - - digrich-bubblechart-panel - - grafana-clock-panel - ingress: - enabled: true - ingressClassName: openreplay - hosts: - - *domainName - annotations: - nginx.ingress.kubernetes.io/rewrite-target: /$1 - path: /grafana/(.*) - tls: - - hosts: - - *domainName - secretName: openreplay-ssl - -prometheus: - prometheusSpec: - storageSpec: - volumeClaimTemplate: - spec: - accessModes: ["ReadWriteOnce"] - resources: - requests: - storage: 200Gi -alertmanager: - config: - global: - resolve_timeout: 5m - slack_api_url: *slackwebhook - route: - # group_by: ['job'] - group_by: ['alertname','container'] - group_wait: 30s - group_interval: 5m - repeat_interval: 12h - receiver: 'slack' - routes: - - match: - alertname: Watchdog - receiver: 'slack' - receivers: - - name: slack - slack_configs: - - channel: *slackchannel - color: '{{ template "slack.color" . }}' - title: '{{ template "slack.title" . }}' - text: '{{ template "slack.text" . }}' - send_resolved: true - actions: - - type: button - text: 'Runbook :green_book:' - url: '{{ (index .Alerts 0).Annotations.runbook_url }}' - # - type: button - # text: 'Query :mag:' - # url: '{{ (index .Alerts 0).GeneratorURL }}' - # - type: button - # text: 'Dashboard :chart_with_upwards_trend:' - # url: '{{ (index .Alerts 0).Annotations.dashboard_url }}' - # - type: button - # text: 'Silence :no_bell:' - # url: '{{ template "__alert_silence_link" . }}' - templates: - - /etc/alertmanager/config/*.tmpl - - templateFiles: - template_1.tmpl: |- - {{/* Alertmanager Silence link */}} - {{ define "__alert_silence_link" -}} - {{ .ExternalURL }}/#/silences/new?filter=%7B - {{- range .CommonLabels.SortedPairs -}} - {{- if ne .Name "alertname" -}} - {{- .Name }}%3D"{{- .Value -}}"%2C%20 - {{- end -}} - {{- end -}} - alertname%3D"{{- .CommonLabels.alertname -}}"%7D - {{- end }} - - {{/* Severity of the alert */}} - {{ define "__alert_severity" -}} - {{- if eq .CommonLabels.severity "critical" -}} - *Severity:* `Critical` - {{- else if eq .CommonLabels.severity "warning" -}} - *Severity:* `Warning` - {{- else if eq .CommonLabels.severity "info" -}} - *Severity:* `Info` - {{- else -}} - *Severity:* :question: {{ .CommonLabels.severity }} - {{- end }} - {{- end }} - - {{/* Title of the Slack alert */}} - {{ define "slack.title" -}} - [{{ .Status | toUpper -}} - {{ if eq .Status "firing" }}:{{ .Alerts.Firing | len }}{{- end -}} - ] {{ .CommonLabels.alertname }} - {{- end }} - - - {{/* Color of Slack attachment (appears as line next to alert )*/}} - {{ define "slack.color" -}} - {{ if eq .Status "firing" -}} - {{ if eq .CommonLabels.severity "warning" -}} - warning - {{- else if eq .CommonLabels.severity "critical" -}} - danger - {{- else -}} - #439FE0 - {{- end -}} - {{ else -}} - good - {{- end }} - {{- end }} - - {{/* The text to display in the alert */}} - {{ define "slack.text" -}} - - {{ template "__alert_severity" . }} - {{- if (index .Alerts 0).Annotations.summary }} - {{- "\n" -}} - *Summary:* {{ (index .Alerts 0).Annotations.summary }} - {{- end }} - - {{ range .Alerts }} - - {{- if .Annotations.description }} - {{- "\n" -}} - {{ .Annotations.description }} - {{- "\n" -}} - {{- end }} - {{- if .Annotations.message }} - {{- "\n" -}} - {{ .Annotations.message }} - {{- "\n" -}} - {{- end }} - - {{- end }} - - {{- end }} diff --git a/ee/scripts/helmcharts/manifests/observability-values.yaml b/ee/scripts/helmcharts/manifests/observability-values.yaml new file mode 100644 index 000000000..f03a4f65c --- /dev/null +++ b/ee/scripts/helmcharts/manifests/observability-values.yaml @@ -0,0 +1,300 @@ +# vim: set fdm=indent: +################################################## +## Update the following values +## +## For example, +# domainName: &domainName openreplay.supercompany.com +################################################## + +domainName: &domainName "changeme.mycorp.org" +grafanaAdminPassword: &adminpass "changeNeGrafanaAdminPassword" +slackWebhookUrl: &slackwebhook "https://hooks.slack.com/services/xxxx/xxxx/xxxxx" # Slack webhook token for sending alerts +slackChannel: &slackchannel "changeMeAlertsChannel" # Name of the channel, to alerts to be delivered to. + + +######################################################## +## Custom configuration for Monitoring and logging stack +######################################################## + +observability: + kube-prometheus-stack: + fullnameOverride: "openreplay" + grafana: + adminPassword: *adminpass + env: + GF_SERVER_ROOT_URL: http://grafana.local.com/grafana + additionalDataSources: + - name: loki + editable: true + type: loki + url: http://loki.observability:3100 + plugins: + - grafana-piechart-panel + - vertamedia-clickhouse-datasource + - digrich-bubblechart-panel + - grafana-clock-panel + ingress: + enabled: true + ingressClassName: openreplay + hosts: + - *domainName + annotations: + nginx.ingress.kubernetes.io/rewrite-target: /$1 + path: /grafana/(.*) + tls: + - hosts: + - *domainName + secretName: openreplay-ssl + + prometheus: + prometheusSpec: + storageSpec: + volumeClaimTemplate: + spec: + accessModes: ["ReadWriteOnce"] + resources: + requests: + storage: 200Gi + alertmanager: + config: + global: + resolve_timeout: 5m + slack_api_url: *slackwebhook + route: + # group_by: ['job'] + group_by: ['alertname','container'] + group_wait: 30s + group_interval: 5m + repeat_interval: 12h + receiver: 'slack' + routes: + - match: + alertname: Watchdog + receiver: 'slack' + receivers: + - name: slack + slack_configs: + - channel: *slackchannel + color: '{{ template "slack.color" . }}' + title: '{{ template "slack.title" . }}' + text: '{{ template "slack.text" . }}' + send_resolved: true + actions: + - type: button + text: 'Runbook :green_book:' + url: '{{ (index .Alerts 0).Annotations.runbook_url }}' + # - type: button + # text: 'Query :mag:' + # url: '{{ (index .Alerts 0).GeneratorURL }}' + # - type: button + # text: 'Dashboard :chart_with_upwards_trend:' + # url: '{{ (index .Alerts 0).Annotations.dashboard_url }}' + # - type: button + # text: 'Silence :no_bell:' + # url: '{{ template "__alert_silence_link" . }}' + templates: + - /etc/alertmanager/config/*.tmpl + + templateFiles: + template_1.tmpl: |- + {{/* Alertmanager Silence link */}} + {{ define "__alert_silence_link" -}} + {{ .ExternalURL }}/#/silences/new?filter=%7B + {{- range .CommonLabels.SortedPairs -}} + {{- if ne .Name "alertname" -}} + {{- .Name }}%3D"{{- .Value -}}"%2C%20 + {{- end -}} + {{- end -}} + alertname%3D"{{- .CommonLabels.alertname -}}"%7D + {{- end }} + + {{/* Severity of the alert */}} + {{ define "__alert_severity" -}} + {{- if eq .CommonLabels.severity "critical" -}} + *Severity:* `Critical` + {{- else if eq .CommonLabels.severity "warning" -}} + *Severity:* `Warning` + {{- else if eq .CommonLabels.severity "info" -}} + *Severity:* `Info` + {{- else -}} + *Severity:* :question: {{ .CommonLabels.severity }} + {{- end }} + {{- end }} + + {{/* Title of the Slack alert */}} + {{ define "slack.title" -}} + [{{ .Status | toUpper -}} + {{ if eq .Status "firing" }}:{{ .Alerts.Firing | len }}{{- end -}} + ] {{ .CommonLabels.alertname }} + {{- end }} + + + {{/* Color of Slack attachment (appears as line next to alert )*/}} + {{ define "slack.color" -}} + {{ if eq .Status "firing" -}} + {{ if eq .CommonLabels.severity "warning" -}} + warning + {{- else if eq .CommonLabels.severity "critical" -}} + danger + {{- else -}} + #439FE0 + {{- end -}} + {{ else -}} + good + {{- end }} + {{- end }} + + {{/* The text to display in the alert */}} + {{ define "slack.text" -}} + + {{ template "__alert_severity" . }} + {{- if (index .Alerts 0).Annotations.summary }} + {{- "\n" -}} + *Summary:* {{ (index .Alerts 0).Annotations.summary }} + {{- end }} + + {{ range .Alerts }} + + {{- if .Annotations.description }} + {{- "\n" -}} + {{ .Annotations.description }} + {{- "\n" -}} + {{- end }} + {{- if .Annotations.message }} + {{- "\n" -}} + {{ .Annotations.message }} + {{- "\n" -}} + {{- end }} + + {{- end }} + + {{- end }} + loki: + config: + # existingSecret: + auth_enabled: false + ingester: + chunk_idle_period: 3m + chunk_block_size: 262144 + chunk_retain_period: 1m + max_transfer_retries: 0 + wal: + dir: /data/loki/wal + lifecycler: + ring: + kvstore: + store: inmemory + replication_factor: 1 + + limits_config: + enforce_metric_name: false + reject_old_samples: true + reject_old_samples_max_age: 168h + schema_config: + configs: + - from: 2020-10-24 + store: boltdb-shipper + object_store: filesystem + schema: v11 + index: + prefix: index_ + period: 24h + server: + http_listen_port: 3100 + storage_config: + boltdb_shipper: + active_index_directory: /data/loki/boltdb-shipper-active + cache_location: /data/loki/boltdb-shipper-cache + cache_ttl: 24h # Can be increased for faster performance over longer query periods, uses more disk space + shared_store: filesystem + filesystem: + directory: /data/loki/chunks + chunk_store_config: + max_look_back_period: 0s + table_manager: + retention_deletes_enabled: false + retention_period: 0s + compactor: + working_directory: /data/loki/boltdb-shipper-compactor + shared_store: filesystem + retention_enabled: true + # Needed for Alerting: https://grafana.com/docs/loki/latest/rules/ + # This is just a simple example, for more details: https://grafana.com/docs/loki/latest/configuration/#ruler_config + ruler: + storage: + type: local + local: + directory: /rules + rule_path: /tmp/scratch + alertmanager_url: http://openreplay-alertmanager.observability.svc.cluster.local:9093 + ring: + kvstore: + store: inmemory + enable_api: true + + persistence: + enabled: true + accessModes: + - ReadWriteOnce + size: 100Gi + + serviceMonitor: + enabled: true + interval: "" + additionalLabels: + release: monitoring + annotations: {} + # scrapeTimeout: 10s + # path: /metrics + prometheusRule: + enabled: true + additionalLabels: + release: monitoring + rules: + - alert: LokiProcessTooManyRestarts + expr: changes(process_start_time_seconds{job=~"loki"}[15m]) > 2 + for: 0m + labels: + severity: warning + annotations: + summary: Loki process too many restarts (instance {{ $labels.instance }}) + description: "A loki process had too many restarts (target {{ $labels.instance }})\n VALUE = {{ $value }}\n LABELS = {{ $labels }}" + - alert: LokiRequestErrors + expr: 100 * sum(rate(loki_request_duration_seconds_count{status_code=~"5.."}[1m])) by (namespace, job, route) / sum(rate(loki_request_duration_seconds_count[1m])) by (namespace, job, route) > 10 + for: 15m + labels: + severity: critical + annotations: + summary: Loki request errors (instance {{ $labels.instance }}) + description: "The {{ $labels.job }} and {{ $labels.route }} are experiencing errors\n VALUE = {{ $value }}\n LABELS = {{ $labels }}" + - alert: LokiRequestPanic + expr: sum(increase(loki_panic_total[10m])) by (namespace, job) > 0 + for: 5m + labels: + severity: critical + annotations: + summary: Loki request panic (instance {{ $labels.instance }}) + description: "The {{ $labels.job }} is experiencing {{ printf \"%.2f\" $value }}% increase of panics\n VALUE = {{ $value }}\n LABELS = {{ $labels }}" + - alert: LokiRequestLatency + expr: (histogram_quantile(0.99, sum(rate(loki_request_duration_seconds_bucket{route!~"(?i).*tail.*"}[5m])) by (le))) > 1 + for: 5m + labels: + severity: critical + annotations: + summary: Loki request latency (instance {{ $labels.instance }}) + description: "The {{ $labels.job }} {{ $labels.route }} is experiencing {{ printf \"%.2f\" $value }}s 99th percentile latency\n VALUE = {{ $value }}\n LABELS = {{ $labels }}" + + # Specify Loki Alerting rules based on this documentation: https://grafana.com/docs/loki/latest/rules/ + # When specified, you also need to add a ruler config section above. An example is shown in the alerting docs. + alerting_groups: + - name: dbZombie + rules: + - alert: dbZombie + expr: absent_over_time({namespace="app", app="db"} |~ "Queue Statistics"[5m]) == 1 + for: 10m + labels: + severity: warning + promtail: + config: + clients: + - url: http://loki.observability.svc.cluster.local:3100/loki/api/v1/push diff --git a/ee/scripts/helmcharts/manifests/observability/.helmignore b/ee/scripts/helmcharts/manifests/observability/.helmignore new file mode 100644 index 000000000..0e8a0eb36 --- /dev/null +++ b/ee/scripts/helmcharts/manifests/observability/.helmignore @@ -0,0 +1,23 @@ +# Patterns to ignore when building packages. +# This supports shell glob matching, relative path matching, and +# negation (prefixed with !). Only one pattern per line. +.DS_Store +# Common VCS dirs +.git/ +.gitignore +.bzr/ +.bzrignore +.hg/ +.hgignore +.svn/ +# Common backup files +*.swp +*.bak +*.tmp +*.orig +*~ +# Various IDEs +.project +.idea/ +*.tmproj +.vscode/ diff --git a/ee/scripts/helmcharts/manifests/observability/Chart.lock b/ee/scripts/helmcharts/manifests/observability/Chart.lock new file mode 100644 index 000000000..cbd64b62e --- /dev/null +++ b/ee/scripts/helmcharts/manifests/observability/Chart.lock @@ -0,0 +1,12 @@ +dependencies: +- name: kube-prometheus-stack + repository: https://prometheus-community.github.io/helm-charts + version: 36.2.0 +- name: loki + repository: https://grafana.github.io/helm-charts + version: 2.12.2 +- name: promtail + repository: https://grafana.github.io/helm-charts + version: 6.0.0 +digest: sha256:7f28353afa3626a7aaf59159a877be84f29bfac004e325d12f2ce43f31bd3e90 +generated: "2022-06-30T11:23:29.753889457+02:00" diff --git a/ee/scripts/helmcharts/manifests/observability/Chart.yaml b/ee/scripts/helmcharts/manifests/observability/Chart.yaml new file mode 100644 index 000000000..b6baa807d --- /dev/null +++ b/ee/scripts/helmcharts/manifests/observability/Chart.yaml @@ -0,0 +1,34 @@ +apiVersion: v2 +name: observability +description: A Helm chart for Kubernetes + +# A chart can be either an 'application' or a 'library' chart. +# +# Application charts are a collection of templates that can be packaged into versioned archives +# to be deployed. +# +# Library charts provide useful utilities or functions for the chart developer. They're included as +# a dependency of application charts to inject those utilities and functions into the rendering +# pipeline. Library charts do not define any templates and therefore cannot be deployed. +type: application + +# This is the chart version. This version number should be incremented each time you make changes +# to the chart and its templates, including the app version. +# Versions are expected to follow Semantic Versioning (https://semver.org/) +version: 0.1.0 + +# This is the version number of the application being deployed. This version number should be +# incremented each time you make changes to the application. Versions are not expected to +# follow Semantic Versioning. They should reflect the version the application is using. +# It is recommended to use it with quotes. +appVersion: "1.0.0" +dependencies: + - name: kube-prometheus-stack + repository: https://prometheus-community.github.io/helm-charts + version: 36.2.* + - name: loki + repository: https://grafana.github.io/helm-charts + version: 2.12.* + - name: promtail + repository: https://grafana.github.io/helm-charts + version: 6.0.* diff --git a/ee/scripts/helmcharts/manifests/observability/dashboards/nginx-performance.json b/ee/scripts/helmcharts/manifests/observability/dashboards/nginx-performance.json new file mode 100644 index 000000000..f81cb8429 --- /dev/null +++ b/ee/scripts/helmcharts/manifests/observability/dashboards/nginx-performance.json @@ -0,0 +1,982 @@ +{ + "__inputs": [ + { + "name": "DS_PROMETHEUS", + "label": "Prometheus", + "description": "", + "type": "datasource", + "pluginId": "prometheus", + "pluginName": "Prometheus" + } + ], + "__requires": [ + { + "type": "grafana", + "id": "grafana", + "name": "Grafana", + "version": "6.6.0" + }, + { + "type": "panel", + "id": "graph", + "name": "Graph", + "version": "" + }, + { + "type": "datasource", + "id": "prometheus", + "name": "Prometheus", + "version": "1.0.0" + } + ], + "annotations": { + "list": [ + { + "builtIn": 1, + "datasource": "-- Grafana --", + "enable": true, + "hide": true, + "iconColor": "rgba(0, 211, 255, 1)", + "name": "Annotations & Alerts", + "type": "dashboard" + } + ] + }, + "description": "", + "editable": true, + "gnetId": 9614, + "graphTooltip": 1, + "id": null, + "iteration": 1582146566338, + "links": [], + "panels": [ + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "${DS_PROMETHEUS}", + "description": "Total time taken for nginx and upstream servers to process a request and send a response", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 8, + "w": 12, + "x": 0, + "y": 0 + }, + "hiddenSeries": false, + "id": 91, + "legend": { + "avg": false, + "current": false, + "max": false, + "min": false, + "show": true, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null", + "options": { + "dataLinks": [] + }, + "percentage": false, + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": "histogram_quantile(\n 0.5,\n sum by (le)(\n rate(\n nginx_ingress_controller_request_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", + "interval": "", + "legendFormat": ".5", + "refId": "D" + }, + { + "expr": "histogram_quantile(\n 0.95,\n sum by (le)(\n rate(\n nginx_ingress_controller_request_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", + "interval": "", + "legendFormat": ".95", + "refId": "B" + }, + { + "expr": "histogram_quantile(\n 0.99,\n sum by (le)(\n rate(\n nginx_ingress_controller_request_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", + "interval": "", + "legendFormat": ".99", + "refId": "A" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Total request handling time", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "s", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "${DS_PROMETHEUS}", + "description": "The time spent on receiving the response from the upstream server", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 8, + "w": 12, + "x": 12, + "y": 0 + }, + "hiddenSeries": false, + "id": 94, + "legend": { + "avg": false, + "current": false, + "max": false, + "min": false, + "show": true, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null", + "options": { + "dataLinks": [] + }, + "percentage": false, + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": "histogram_quantile(\n 0.5,\n sum by (le)(\n rate(\n nginx_ingress_controller_response_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", + "instant": false, + "interval": "", + "intervalFactor": 1, + "legendFormat": ".5", + "refId": "D" + }, + { + "expr": "histogram_quantile(\n 0.95,\n sum by (le)(\n rate(\n nginx_ingress_controller_response_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", + "interval": "", + "legendFormat": ".95", + "refId": "B" + }, + { + "expr": "histogram_quantile(\n 0.99,\n sum by (le)(\n rate(\n nginx_ingress_controller_response_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", + "interval": "", + "legendFormat": ".99", + "refId": "A" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Upstream response time", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "s", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "${DS_PROMETHEUS}", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 8, + "w": 12, + "x": 0, + "y": 8 + }, + "hiddenSeries": false, + "id": 93, + "legend": { + "alignAsTable": true, + "avg": false, + "current": false, + "max": false, + "min": false, + "rightSide": true, + "show": true, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null", + "options": { + "dataLinks": [] + }, + "percentage": false, + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": " sum by (path)(\n rate(\n nginx_ingress_controller_request_duration_seconds_count{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n", + "interval": "", + "intervalFactor": 1, + "legendFormat": "{{ path }}", + "refId": "A" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Request volume by Path", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "reqps", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "${DS_PROMETHEUS}", + "description": "For each path observed, its median upstream response time", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 8, + "w": 12, + "x": 12, + "y": 8 + }, + "hiddenSeries": false, + "id": 98, + "legend": { + "alignAsTable": true, + "avg": false, + "current": false, + "max": false, + "min": false, + "rightSide": true, + "show": true, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null", + "options": { + "dataLinks": [] + }, + "percentage": false, + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": "histogram_quantile(\n .5,\n sum by (le, path)(\n rate(\n nginx_ingress_controller_response_duration_seconds_bucket{\n ingress =~ \"$ingress\"\n }[1m]\n )\n )\n)", + "interval": "", + "intervalFactor": 1, + "legendFormat": "{{ path }}", + "refId": "A" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Median upstream response time by Path", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "s", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "${DS_PROMETHEUS}", + "description": "Percentage of 4xx and 5xx responses among all responses.", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 8, + "w": 12, + "x": 0, + "y": 16 + }, + "hiddenSeries": false, + "id": 100, + "legend": { + "alignAsTable": true, + "avg": false, + "current": false, + "max": false, + "min": false, + "rightSide": true, + "show": true, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null as zero", + "options": { + "dataLinks": [] + }, + "percentage": false, + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": "sum by (path) (rate(nginx_ingress_controller_request_duration_seconds_count{\n ingress =~ \"$ingress\",\n status =~ \"[4-5].*\"\n}[1m])) / sum by (path) (rate(nginx_ingress_controller_request_duration_seconds_count{\n ingress =~ \"$ingress\",\n}[1m]))", + "interval": "", + "intervalFactor": 1, + "legendFormat": "{{ path }}", + "refId": "A" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Response error rate by Path", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "percentunit", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "${DS_PROMETHEUS}", + "description": "For each path observed, the sum of upstream request time", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 8, + "w": 12, + "x": 12, + "y": 16 + }, + "hiddenSeries": false, + "id": 102, + "legend": { + "alignAsTable": true, + "avg": false, + "current": false, + "max": false, + "min": false, + "rightSide": true, + "show": true, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null", + "options": { + "dataLinks": [] + }, + "percentage": false, + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": "sum by (path) (rate(nginx_ingress_controller_response_duration_seconds_sum{ingress =~ \"$ingress\"}[1m]))", + "interval": "", + "intervalFactor": 1, + "legendFormat": "{{ path }}", + "refId": "A" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Upstream time consumed by Path", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "s", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "${DS_PROMETHEUS}", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 8, + "w": 12, + "x": 0, + "y": 24 + }, + "hiddenSeries": false, + "id": 101, + "legend": { + "alignAsTable": true, + "avg": false, + "current": false, + "max": false, + "min": false, + "rightSide": true, + "show": true, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null", + "options": { + "dataLinks": [] + }, + "percentage": false, + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": " sum (\n rate(\n nginx_ingress_controller_request_duration_seconds_count{\n ingress =~ \"$ingress\",\n status =~\"[4-5].*\",\n }[1m]\n )\n ) by(path, status)\n", + "interval": "", + "intervalFactor": 1, + "legendFormat": "{{ path }} {{ status }}", + "refId": "A" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Response error volume by Path", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "reqps", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "${DS_PROMETHEUS}", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 8, + "w": 12, + "x": 12, + "y": 24 + }, + "hiddenSeries": false, + "id": 99, + "legend": { + "alignAsTable": true, + "avg": false, + "current": false, + "max": false, + "min": false, + "rightSide": true, + "show": true, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null", + "options": { + "dataLinks": [] + }, + "percentage": false, + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": "sum (\n rate (\n nginx_ingress_controller_response_size_sum {\n ingress =~ \"$ingress\",\n }[1m]\n )\n) by (path) / sum (\n rate(\n nginx_ingress_controller_response_size_count {\n ingress =~ \"$ingress\",\n }[1m]\n )\n) by (path)\n", + "hide": false, + "instant": false, + "interval": "", + "intervalFactor": 1, + "legendFormat": "{{ path }}", + "refId": "D" + }, + { + "expr": " sum (rate(nginx_ingress_controller_response_size_bucket{\n ingress =~ \"$ingress\",\n }[1m])) by (le)\n", + "hide": true, + "legendFormat": "{{le}}", + "refId": "A" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Average response size by Path", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "decbytes", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": "${DS_PROMETHEUS}", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 8, + "w": 12, + "x": 0, + "y": 32 + }, + "hiddenSeries": false, + "id": 96, + "legend": { + "avg": false, + "current": false, + "max": false, + "min": false, + "show": true, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null", + "options": { + "dataLinks": [] + }, + "percentage": false, + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "expr": "sum (\n rate(\n nginx_ingress_controller_ingress_upstream_latency_seconds_sum {\n ingress =~ \"$ingress\",\n }[1m]\n)) / sum (\n rate(\n nginx_ingress_controller_ingress_upstream_latency_seconds_count {\n ingress =~ \"$ingress\",\n }[1m]\n )\n)\n", + "hide": false, + "instant": false, + "interval": "", + "intervalFactor": 1, + "legendFormat": "average", + "refId": "B" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Upstream service latency", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "s", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + } + ], + "refresh": "30s", + "schemaVersion": 22, + "style": "dark", + "tags": [ + "nginx" + ], + "templating": { + "list": [ + { + "hide": 0, + "label": "datasource", + "name": "DS_PROMETHEUS", + "options": [], + "query": "prometheus", + "refresh": 1, + "regex": "", + "type": "datasource" + }, + { + "allValue": ".*", + "current": {}, + "datasource": "${DS_PROMETHEUS}", + "definition": "label_values(nginx_ingress_controller_requests, ingress) ", + "hide": 0, + "includeAll": true, + "label": "Service Ingress", + "multi": false, + "name": "ingress", + "options": [], + "query": "label_values(nginx_ingress_controller_requests, ingress) ", + "refresh": 1, + "regex": "", + "skipUrlSync": false, + "sort": 2, + "tagValuesQuery": "", + "tags": [], + "tagsQuery": "", + "type": "query", + "useTags": false + } + ] + }, + "time": { + "from": "now-15m", + "to": "now" + }, + "timepicker": { + "refresh_intervals": [ + "5s", + "10s", + "30s", + "2m", + "5m", + "15m", + "30m", + "1h", + "2h", + "1d" + ], + "time_options": [ + "5m", + "15m", + "1h", + "6h", + "12h", + "24h", + "2d", + "7d", + "30d" + ] + }, + "timezone": "browser", + "title": "Request Handling Performance", + "uid": "4GFbkOsZk", + "version": 1 +} + diff --git a/ee/scripts/helmcharts/manifests/observability/dashboards/openreplay-components.json b/ee/scripts/helmcharts/manifests/observability/dashboards/openreplay-components.json new file mode 100644 index 000000000..1dc2fe7e7 --- /dev/null +++ b/ee/scripts/helmcharts/manifests/observability/dashboards/openreplay-components.json @@ -0,0 +1,1223 @@ +{ + "__inputs": [], + "__requires": [ + { + "type": "grafana", + "id": "grafana", + "name": "Grafana", + "version": "8.2.0" + }, + { + "type": "panel", + "id": "graph", + "name": "Graph (old)", + "version": "" + }, + { + "type": "panel", + "id": "heatmap", + "name": "Heatmap", + "version": "" + }, + { + "type": "panel", + "id": "stat", + "name": "Stat", + "version": "" + } + ], + "annotations": { + "list": [ + { + "builtIn": 1, + "datasource": "-- 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": true, + "fiscalYearStartMonth": 0, + "gnetId": null, + "graphTooltip": 0, + "id": null, + "iteration": 1656567623206, + "links": [], + "liveNow": false, + "panels": [ + { + "datasource": null, + "description": "Number of all requests that have been processed / service", + "fieldConfig": { + "defaults": { + "color": { + "mode": "thresholds" + }, + "mappings": [], + "thresholds": { + "mode": "absolute", + "steps": [ + { + "color": "dark-red", + "value": null + }, + { + "color": "green", + "value": 80 + } + ] + }, + "unit": "short" + }, + "overrides": [] + }, + "id": 18, + "options": { + "colorMode": "value", + "graphMode": "area", + "justifyMode": "auto", + "orientation": "auto", + "reduceOptions": { + "calcs": [ + "lastNotNull" + ], + "fields": "", + "values": false + }, + "text": {}, + "textMode": "auto" + }, + "pluginVersion": "8.2.0", + "targets": [ + { + "exemplar": true, + "expr": "sum(increase(requests_total[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "title": "Requests total / ${time}", + "type": "stat" + }, + { + "datasource": null, + "description": "Number of all messages that have been processed / service", + "fieldConfig": { + "defaults": { + "color": { + "mode": "thresholds" + }, + "mappings": [], + "thresholds": { + "mode": "absolute", + "steps": [ + { + "color": "dark-red", + "value": null + }, + { + "color": "light-green", + "value": 80 + } + ] + }, + "unit": "short" + }, + "overrides": [] + }, + "gridPos": { + "h": 3, + "w": 6, + "x": 6, + "y": 0 + }, + "id": 22, + "options": { + "colorMode": "value", + "graphMode": "area", + "justifyMode": "auto", + "orientation": "auto", + "reduceOptions": { + "calcs": [ + "lastNotNull" + ], + "fields": "", + "values": false + }, + "text": {}, + "textMode": "auto" + }, + "pluginVersion": "8.2.0", + "targets": [ + { + "exemplar": true, + "expr": "sum(increase(messages_total[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "title": "Messages processed / ${time}", + "type": "stat" + }, + { + "datasource": null, + "description": "Number of messages that have been saved to session files", + "fieldConfig": { + "defaults": { + "color": { + "mode": "thresholds" + }, + "mappings": [], + "thresholds": { + "mode": "absolute", + "steps": [ + { + "color": "dark-red", + "value": null + }, + { + "color": "light-green", + "value": 80 + } + ] + }, + "unit": "short" + }, + "overrides": [] + }, + "gridPos": { + "h": 3, + "w": 6, + "x": 12, + "y": 0 + }, + "id": 24, + "options": { + "colorMode": "value", + "graphMode": "area", + "justifyMode": "auto", + "orientation": "auto", + "reduceOptions": { + "calcs": [ + "lastNotNull" + ], + "fields": "", + "values": false + }, + "text": {}, + "textMode": "auto" + }, + "pluginVersion": "8.2.0", + "targets": [ + { + "exemplar": true, + "expr": "sum(increase(messages_saved[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "title": "Messages saved / ${time}", + "type": "stat" + }, + { + "datasource": null, + "description": "Number of all sessions that have been compressed and saved to s3", + "fieldConfig": { + "defaults": { + "color": { + "mode": "thresholds" + }, + "mappings": [], + "thresholds": { + "mode": "absolute", + "steps": [ + { + "color": "dark-red", + "value": null + }, + { + "color": "light-green", + "value": 80 + } + ] + } + }, + "overrides": [] + }, + "gridPos": { + "h": 3, + "w": 6, + "x": 18, + "y": 0 + }, + "id": 30, + "options": { + "colorMode": "value", + "graphMode": "area", + "justifyMode": "auto", + "orientation": "auto", + "reduceOptions": { + "calcs": [ + "lastNotNull" + ], + "fields": "", + "values": false + }, + "text": {}, + "textMode": "auto" + }, + "pluginVersion": "8.2.0", + "targets": [ + { + "exemplar": true, + "expr": "sum(increase(sessions_total{container_name=\"saas-storage\"}[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "title": "Sessions stored in s3 / ${time}", + "type": "stat" + }, + { + "collapsed": false, + "datasource": null, + "gridPos": { + "h": 1, + "w": 24, + "x": 0, + "y": 3 + }, + "id": 28, + "panels": [], + "title": "Storage", + "type": "row" + }, + { + "cards": { + "cardPadding": null, + "cardRound": null + }, + "color": { + "cardColor": "#FADE2A", + "colorScale": "sqrt", + "colorScheme": "interpolateGreys", + "exponent": 0.5, + "mode": "opacity" + }, + "dataFormat": "timeseries", + "datasource": null, + "description": "Sizes of saved sessions (before compression)\n", + "gridPos": { + "h": 8, + "w": 12, + "x": 0, + "y": 4 + }, + "heatmap": {}, + "hideZeroBuckets": false, + "highlightCards": true, + "id": 32, + "legend": { + "show": false + }, + "pluginVersion": "7.5.11", + "reverseYBuckets": false, + "targets": [ + { + "exemplar": true, + "expr": "sum(rate(sessions_size_sum[${time}]))/sum(rate(sessions_size_count[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "title": "Session size histogram / ${time}", + "tooltip": { + "show": true, + "showHistogram": false + }, + "type": "heatmap", + "xAxis": { + "show": true + }, + "xBucketNumber": null, + "xBucketSize": null, + "yAxis": { + "decimals": null, + "format": "decbytes", + "logBase": 1, + "max": null, + "min": null, + "show": true, + "splitFactor": null + }, + "yBucketBound": "auto", + "yBucketNumber": null, + "yBucketSize": null + }, + { + "cards": { + "cardPadding": null, + "cardRound": null + }, + "color": { + "cardColor": "#FADE2A", + "colorScale": "sqrt", + "colorScheme": "interpolateOranges", + "exponent": 0.5, + "mode": "opacity" + }, + "dataFormat": "timeseries", + "datasource": null, + "description": "Durations of archiving process for all handled sessions", + "gridPos": { + "h": 8, + "w": 12, + "x": 12, + "y": 4 + }, + "heatmap": {}, + "hideZeroBuckets": false, + "highlightCards": true, + "id": 34, + "legend": { + "show": false + }, + "pluginVersion": "7.5.11", + "reverseYBuckets": false, + "targets": [ + { + "exemplar": true, + "expr": "sum(rate(archiving_duration_sum[${time}]))/sum(rate(archiving_duration_count[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "title": "Archiving duration histogram / ${time}", + "tooltip": { + "show": true, + "showHistogram": false + }, + "type": "heatmap", + "xAxis": { + "show": true + }, + "xBucketNumber": null, + "xBucketSize": null, + "yAxis": { + "decimals": null, + "format": "ms", + "logBase": 1, + "max": null, + "min": null, + "show": true, + "splitFactor": null + }, + "yBucketBound": "auto", + "yBucketNumber": null, + "yBucketSize": null + }, + { + "collapsed": false, + "datasource": null, + "gridPos": { + "h": 1, + "w": 24, + "x": 0, + "y": 12 + }, + "id": 20, + "panels": [], + "title": "Sink", + "type": "row" + }, + { + "cards": { + "cardPadding": null, + "cardRound": null + }, + "color": { + "cardColor": "#FADE2A", + "colorScale": "sqrt", + "colorScheme": "interpolateOranges", + "exponent": 0.5, + "mode": "opacity" + }, + "dataFormat": "timeseries", + "datasource": null, + "description": "Sizes of saved messages\n", + "gridPos": { + "h": 8, + "w": 12, + "x": 0, + "y": 13 + }, + "heatmap": {}, + "hideZeroBuckets": false, + "highlightCards": true, + "id": 26, + "legend": { + "show": false + }, + "pluginVersion": "7.5.11", + "reverseYBuckets": false, + "targets": [ + { + "exemplar": true, + "expr": "sum(rate(messages_size_sum[${time}])) / sum(rate(messages_size_count[${time}]))", + "hide": false, + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "title": "Message size histogram / ${time}", + "tooltip": { + "show": true, + "showHistogram": false + }, + "type": "heatmap", + "xAxis": { + "show": true + }, + "xBucketNumber": null, + "xBucketSize": null, + "yAxis": { + "decimals": null, + "format": "decbytes", + "logBase": 1, + "max": null, + "min": null, + "show": true, + "splitFactor": null + }, + "yBucketBound": "auto", + "yBucketNumber": null, + "yBucketSize": null + }, + { + "collapsed": false, + "datasource": null, + "gridPos": { + "h": 1, + "w": 24, + "x": 0, + "y": 21 + }, + "id": 12, + "panels": [], + "title": "Http", + "type": "row" + }, + { + "cards": { + "cardPadding": null, + "cardRound": null + }, + "color": { + "cardColor": "#FADE2A", + "colorScale": "sqrt", + "colorScheme": "interpolateOranges", + "exponent": 0.5, + "mode": "opacity" + }, + "dataFormat": "timeseries", + "datasource": null, + "description": "Sizes of request’s body for start and ingest endpoints", + "gridPos": { + "h": 8, + "w": 12, + "x": 0, + "y": 22 + }, + "heatmap": {}, + "hideZeroBuckets": false, + "highlightCards": true, + "id": 14, + "legend": { + "show": false + }, + "pluginVersion": "7.5.11", + "reverseYBuckets": false, + "targets": [ + { + "exemplar": true, + "expr": "sum(rate(requests_body_size_sum[${time}]))/sum(rate(requests_body_size_count[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "timeFrom": null, + "timeShift": null, + "title": "Request body size / ${time}", + "tooltip": { + "show": true, + "showHistogram": false + }, + "type": "heatmap", + "xAxis": { + "show": true + }, + "xBucketNumber": null, + "xBucketSize": null, + "yAxis": { + "decimals": null, + "format": "decbytes", + "logBase": 1, + "max": null, + "min": null, + "show": true, + "splitFactor": null + }, + "yBucketBound": "auto", + "yBucketNumber": null, + "yBucketSize": null + }, + { + "cards": { + "cardPadding": null, + "cardRound": null + }, + "color": { + "cardColor": "#FADE2A", + "colorScale": "sqrt", + "colorScheme": "interpolateOranges", + "exponent": 0.5, + "mode": "opacity" + }, + "dataFormat": "timeseries", + "datasource": null, + "description": "Request processing duration for all endpoints", + "gridPos": { + "h": 8, + "w": 12, + "x": 12, + "y": 22 + }, + "heatmap": {}, + "hideZeroBuckets": false, + "highlightCards": true, + "id": 16, + "legend": { + "show": false + }, + "pluginVersion": "7.5.11", + "reverseYBuckets": false, + "targets": [ + { + "exemplar": true, + "expr": "sum(rate(requests_duration_sum[${time}]))/sum(rate(requests_duration_count[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "title": "Request duration / ${time}", + "tooltip": { + "show": true, + "showHistogram": false + }, + "type": "heatmap", + "xAxis": { + "show": true + }, + "xBucketNumber": null, + "xBucketSize": null, + "yAxis": { + "decimals": null, + "format": "ms", + "logBase": 1, + "max": null, + "min": null, + "show": true, + "splitFactor": null + }, + "yBucketBound": "auto", + "yBucketNumber": null, + "yBucketSize": null + }, + { + "collapsed": false, + "datasource": null, + "gridPos": { + "h": 1, + "w": 24, + "x": 0, + "y": 30 + }, + "id": 6, + "panels": [], + "title": "ender", + "type": "row" + }, + { + "aliasColors": { + "Value": "green" + }, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": null, + "description": "Number of active sessions (unfinished sessions)\n", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 8, + "w": 12, + "x": 0, + "y": 31 + }, + "hiddenSeries": false, + "id": 8, + "legend": { + "avg": false, + "current": false, + "max": false, + "min": false, + "show": false, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null", + "options": { + "alertThreshold": true + }, + "percentage": false, + "pluginVersion": "8.2.0", + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "exemplar": true, + "expr": "sum(sessions_active)", + "format": "time_series", + "interval": "", + "legendFormat": "{{ pod }}", + "refId": "A" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Active sessions", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": null, + "description": "Number of all sessions that have been processed / service", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 8, + "w": 12, + "x": 12, + "y": 31 + }, + "hiddenSeries": false, + "id": 10, + "legend": { + "avg": false, + "current": false, + "max": false, + "min": false, + "show": false, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null", + "options": { + "alertThreshold": true + }, + "percentage": false, + "pluginVersion": "8.2.0", + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "exemplar": true, + "expr": "sum(increase(sessions_total{container_name=\"saas-ender\"}[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Sessions processed / ${time}", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + }, + { + "collapsed": false, + "datasource": null, + "gridPos": { + "h": 1, + "w": 24, + "x": 0, + "y": 39 + }, + "id": 4, + "panels": [], + "title": "Assets", + "type": "row" + }, + { + "aliasColors": {}, + "bars": false, + "dashLength": 10, + "dashes": false, + "datasource": null, + "description": "Number of all assets that have been processed / service", + "fill": 1, + "fillGradient": 0, + "gridPos": { + "h": 9, + "w": 12, + "x": 0, + "y": 40 + }, + "hiddenSeries": false, + "id": 2, + "legend": { + "avg": false, + "current": false, + "max": false, + "min": false, + "show": false, + "total": false, + "values": false + }, + "lines": true, + "linewidth": 1, + "nullPointMode": "null", + "options": { + "alertThreshold": true + }, + "percentage": false, + "pluginVersion": "8.2.0", + "pointradius": 2, + "points": false, + "renderer": "flot", + "seriesOverrides": [], + "spaceLength": 10, + "stack": false, + "steppedLine": false, + "targets": [ + { + "exemplar": true, + "expr": "sum(increase(assets_total[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "thresholds": [], + "timeFrom": null, + "timeRegions": [], + "timeShift": null, + "title": "Assets processed / ${time}", + "tooltip": { + "shared": true, + "sort": 0, + "value_type": "individual" + }, + "type": "graph", + "xaxis": { + "buckets": null, + "mode": "time", + "name": null, + "show": true, + "values": [] + }, + "yaxes": [ + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + }, + { + "format": "short", + "label": null, + "logBase": 1, + "max": null, + "min": null, + "show": true + } + ], + "yaxis": { + "align": false, + "alignLevel": null + } + }, + { + "collapsed": false, + "datasource": null, + "gridPos": { + "h": 1, + "w": 24, + "x": 0, + "y": 49 + }, + "id": 36, + "panels": [], + "title": "Db", + "type": "row" + }, + { + "cards": { + "cardPadding": null, + "cardRound": null + }, + "color": { + "cardColor": "#FADE2A", + "colorScale": "sqrt", + "colorScheme": "interpolateOranges", + "exponent": 0.5, + "mode": "opacity" + }, + "dataFormat": "timeseries", + "datasource": null, + "description": "Size of batch in bytes before execution in PG", + "gridPos": { + "h": 8, + "w": 12, + "x": 0, + "y": 50 + }, + "heatmap": {}, + "hideZeroBuckets": false, + "highlightCards": true, + "id": 38, + "legend": { + "show": false + }, + "reverseYBuckets": false, + "targets": [ + { + "exemplar": true, + "expr": "sum(rate(batch_size_bytes_sum[${time}]))/sum(rate(batch_size_bytes_count[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "title": "Batch size / ${time}", + "tooltip": { + "show": true, + "showHistogram": false + }, + "type": "heatmap", + "xAxis": { + "show": true + }, + "xBucketNumber": null, + "xBucketSize": null, + "yAxis": { + "decimals": null, + "format": "decbytes", + "logBase": 1, + "max": null, + "min": null, + "show": true, + "splitFactor": null + }, + "yBucketBound": "auto", + "yBucketNumber": null, + "yBucketSize": null + }, + { + "cards": { + "cardPadding": null, + "cardRound": null + }, + "color": { + "cardColor": "#FADE2A", + "colorScale": "sqrt", + "colorScheme": "interpolateOranges", + "exponent": 0.5, + "mode": "opacity" + }, + "dataFormat": "timeseries", + "datasource": null, + "description": "Number of sql requests before execution in PG", + "gridPos": { + "h": 8, + "w": 12, + "x": 12, + "y": 50 + }, + "heatmap": {}, + "hideZeroBuckets": false, + "highlightCards": true, + "id": 39, + "legend": { + "show": false + }, + "reverseYBuckets": false, + "targets": [ + { + "exemplar": true, + "expr": "sum(rate(batch_size_lines_sum[${time}]))/sum(rate(batch_size_lines_count[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "title": "Number of sql requests before PG execution / ${time}", + "tooltip": { + "show": true, + "showHistogram": false + }, + "type": "heatmap", + "xAxis": { + "show": true + }, + "xBucketNumber": null, + "xBucketSize": null, + "yAxis": { + "decimals": null, + "format": "short", + "logBase": 1, + "max": null, + "min": null, + "show": true, + "splitFactor": null + }, + "yBucketBound": "auto", + "yBucketNumber": null, + "yBucketSize": null + }, + { + "cards": { + "cardPadding": null, + "cardRound": null + }, + "color": { + "cardColor": "#FADE2A", + "colorScale": "sqrt", + "colorScheme": "interpolateOranges", + "exponent": 0.5, + "mode": "opacity" + }, + "dataFormat": "timeseries", + "datasource": null, + "description": "Duration of SQL request", + "gridPos": { + "h": 8, + "w": 12, + "x": 0, + "y": 58 + }, + "heatmap": {}, + "hideZeroBuckets": false, + "highlightCards": true, + "id": 40, + "legend": { + "show": false + }, + "reverseYBuckets": false, + "targets": [ + { + "exemplar": true, + "expr": "sum(rate(sql_request_time_sum[${time}]))/sum(rate(sql_request_time_count[${time}]))", + "interval": "", + "legendFormat": "", + "refId": "A" + } + ], + "title": "Duration of sql request / ${time}", + "tooltip": { + "show": true, + "showHistogram": false + }, + "type": "heatmap", + "xAxis": { + "show": true + }, + "xBucketNumber": null, + "xBucketSize": null, + "yAxis": { + "decimals": null, + "format": "ms", + "logBase": 1, + "max": null, + "min": null, + "show": true, + "splitFactor": null + }, + "yBucketBound": "auto", + "yBucketNumber": null, + "yBucketSize": null + } + ], + "refresh": "", + "schemaVersion": 31, + "style": "dark", + "tags": [], + "templating": { + "list": [ + { + "auto": true, + "auto_count": 30, + "auto_min": "1m", + "current": { + "selected": false, + "text": "auto", + "value": "$__auto_interval_time" + }, + "description": null, + "error": null, + "hide": 1, + "label": "Data sampling time", + "name": "time", + "options": [ + { + "selected": true, + "text": "auto", + "value": "$__auto_interval_time" + }, + { + "selected": false, + "text": "1m", + "value": "1m" + }, + { + "selected": false, + "text": "2m", + "value": "2m" + }, + { + "selected": false, + "text": "5m", + "value": "5m" + }, + { + "selected": false, + "text": "10m", + "value": "10m" + } + ], + "query": "1m,2m,5m,10m", + "queryValue": "", + "refresh": 2, + "skipUrlSync": false, + "type": "interval" + } + ] + }, + "time": { + "from": "now-6h", + "to": "now" + }, + "timepicker": {}, + "timezone": "", + "title": "Openreplay Components Dashboard", + "uid": "aq_tupq7k", + "version": 31 +} diff --git a/ee/scripts/helmcharts/manifests/observability/templates/NOTES.txt b/ee/scripts/helmcharts/manifests/observability/templates/NOTES.txt new file mode 100644 index 000000000..0b4c60cf1 --- /dev/null +++ b/ee/scripts/helmcharts/manifests/observability/templates/NOTES.txt @@ -0,0 +1 @@ +Obeservibity stack installation complete. Please refer domain.com/grafana diff --git a/ee/scripts/helmcharts/manifests/observability/templates/_helpers.tpl b/ee/scripts/helmcharts/manifests/observability/templates/_helpers.tpl new file mode 100644 index 000000000..6e81760f5 --- /dev/null +++ b/ee/scripts/helmcharts/manifests/observability/templates/_helpers.tpl @@ -0,0 +1,62 @@ +{{/* +Expand the name of the chart. +*/}} +{{- define "observability.name" -}} +{{- default .Chart.Name .Values.nameOverride | trunc 63 | trimSuffix "-" }} +{{- end }} + +{{/* +Create a default fully qualified app name. +We truncate at 63 chars because some Kubernetes name fields are limited to this (by the DNS naming spec). +If release name contains chart name it will be used as a full name. +*/}} +{{- define "observability.fullname" -}} +{{- if .Values.fullnameOverride }} +{{- .Values.fullnameOverride | trunc 63 | trimSuffix "-" }} +{{- else }} +{{- $name := default .Chart.Name .Values.nameOverride }} +{{- if contains $name .Release.Name }} +{{- .Release.Name | trunc 63 | trimSuffix "-" }} +{{- else }} +{{- printf "%s-%s" .Release.Name $name | trunc 63 | trimSuffix "-" }} +{{- end }} +{{- end }} +{{- end }} + +{{/* +Create chart name and version as used by the chart label. +*/}} +{{- define "observability.chart" -}} +{{- printf "%s-%s" .Chart.Name .Chart.Version | replace "+" "_" | trunc 63 | trimSuffix "-" }} +{{- end }} + +{{/* +Common labels +*/}} +{{- define "observability.labels" -}} +helm.sh/chart: {{ include "observability.chart" . }} +{{ include "observability.selectorLabels" . }} +{{- if .Chart.AppVersion }} +app.kubernetes.io/version: {{ .Chart.AppVersion | quote }} +{{- end }} +app.kubernetes.io/managed-by: {{ .Release.Service }} +{{- end }} + +{{/* +Selector labels +*/}} +{{- define "observability.selectorLabels" -}} +app.kubernetes.io/name: {{ include "observability.name" . }} +app.kubernetes.io/instance: {{ .Release.Name }} +{{- end }} + +{{/* +Create the name of the service account to use +*/}} +{{- define "observability.serviceAccountName" -}} +{{- if .Values.serviceAccount.create }} +{{- default (include "observability.fullname" .) .Values.serviceAccount.name }} +{{- else }} +{{- default "default" .Values.serviceAccount.name }} +{{- end }} +{{- end }} diff --git a/ee/scripts/helmcharts/manifests/observability/templates/dashboard.yaml b/ee/scripts/helmcharts/manifests/observability/templates/dashboard.yaml new file mode 100644 index 000000000..d555a00d7 --- /dev/null +++ b/ee/scripts/helmcharts/manifests/observability/templates/dashboard.yaml @@ -0,0 +1,11 @@ +apiVersion: v1 +data: + {{- (.Files.Glob "dashboards/*").AsConfig | nindent 2 }} +kind: ConfigMap +metadata: + labels: + app: kube-prometheus-stack-grafana + grafana_dashboard: "1" + release: monitoring + name: nginx-igress + namespace: observability diff --git a/ee/scripts/helmcharts/manifests/observability/values.yaml b/ee/scripts/helmcharts/manifests/observability/values.yaml new file mode 100644 index 000000000..e69de29bb