diff --git a/addons/grafana/dashboards/sandbox_controller.json b/addons/grafana/dashboards/sandbox_controller.json new file mode 100644 index 000000000..9c45eb6a3 --- /dev/null +++ b/addons/grafana/dashboards/sandbox_controller.json @@ -0,0 +1,418 @@ +{ + "annotations": { + "list": [] + }, + "editable": true, + "fiscalYearStartMonth": 0, + "graphTooltip": 1, + "links": [], + "panels": [ + { + "collapsed": false, + "gridPos": { "h": 1, "w": 24, "x": 0, "y": 0 }, + "id": 100, + "panels": [], + "title": "Overview", + "type": "row" + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "short" + } + }, + "gridPos": { "h": 4, "w": 4, "x": 0, "y": 1 }, + "id": 1, + "options": { "colorMode": "value", "graphMode": "area", "justifyMode": "auto", "orientation": "auto", "reduceOptions": { "calcs": ["lastNotNull"], "fields": "", "values": false }, "textMode": "auto" }, + "title": "Sandboxes Created (1m)", + "type": "stat", + "targets": [{ "expr": "sum(increase(sandbox_time_to_running_seconds_count[1m]))", "refId": "A" }] + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "thresholds" }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }, { "color": "yellow", "value": 5 }, { "color": "red", "value": 10 }] }, + "unit": "short" + } + }, + "gridPos": { "h": 4, "w": 4, "x": 4, "y": 1 }, + "id": 2, + "options": { "colorMode": "value", "graphMode": "none", "justifyMode": "auto", "orientation": "auto", "reduceOptions": { "calcs": ["lastNotNull"], "fields": "", "values": false }, "textMode": "auto" }, + "title": "Active Containers (Daemon)", + "type": "stat", + "targets": [{ "expr": "sum(sandbox_daemon_containers_active)", "refId": "A" }] + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "thresholds" }, + "mappings": [], + "thresholds": { "mode": "percentage", "steps": [{ "color": "red", "value": null }, { "color": "yellow", "value": 20 }, { "color": "green", "value": 50 }] }, + "unit": "percentunit", + "max": 1, "min": 0 + } + }, + "gridPos": { "h": 4, "w": 4, "x": 8, "y": 1 }, + "id": 3, + "options": { "colorMode": "value", "graphMode": "none", "justifyMode": "auto", "orientation": "auto", "reduceOptions": { "calcs": ["lastNotNull"], "fields": "", "values": false }, "textMode": "auto" }, + "title": "Pool Availability", + "type": "stat", + "targets": [{ "expr": "sum(sandbox_warm_pool_available) / sum(sandbox_warm_pool_target)", "refId": "A" }] + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "thresholds" }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }, { "color": "yellow", "value": 1 }, { "color": "red", "value": 5 }] }, + "unit": "s" + } + }, + "gridPos": { "h": 4, "w": 4, "x": 12, "y": 1 }, + "id": 4, + "options": { "colorMode": "value", "graphMode": "area", "justifyMode": "auto", "orientation": "auto", "reduceOptions": { "calcs": ["lastNotNull"], "fields": "", "values": false }, "textMode": "auto" }, + "title": "P95 Time to Running", + "type": "stat", + "targets": [{ "expr": "histogram_quantile(0.95, sum(rate(sandbox_time_to_running_seconds_bucket[5m])) by (le))", "refId": "A" }] + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "thresholds" }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }, { "color": "red", "value": 1 }] }, + "unit": "short" + } + }, + "gridPos": { "h": 4, "w": 4, "x": 16, "y": 1 }, + "id": 5, + "options": { "colorMode": "value", "graphMode": "area", "justifyMode": "auto", "orientation": "auto", "reduceOptions": { "calcs": ["lastNotNull"], "fields": "", "values": false }, "textMode": "auto" }, + "title": "Failures (5m)", + "type": "stat", + "targets": [{ "expr": "sum(increase(sandbox_failures_total[5m]))", "refId": "A" }] + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "short" + } + }, + "gridPos": { "h": 4, "w": 4, "x": 20, "y": 1 }, + "id": 6, + "options": { "colorMode": "value", "graphMode": "area", "justifyMode": "auto", "orientation": "auto", "reduceOptions": { "calcs": ["lastNotNull"], "fields": "", "values": false }, "textMode": "auto" }, + "title": "Exec Commands (1m)", + "type": "stat", + "targets": [{ "expr": "sum(increase(sandbox_exec_total[1m]))", "refId": "A" }] + }, + { + "collapsed": false, + "gridPos": { "h": 1, "w": 24, "x": 0, "y": 5 }, + "id": 101, + "panels": [], + "title": "Warm Pool State", + "type": "row" + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "custom": { "axisBorderShow": false, "axisCenteredZero": false, "axisLabel": "", "axisPlacement": "auto", "barAlignment": 0, "drawStyle": "line", "fillOpacity": 20, "gradientMode": "none", "hideFrom": { "legend": false, "tooltip": false, "viz": false }, "insertNulls": false, "lineInterpolation": "linear", "lineWidth": 2, "pointSize": 5, "scaleDistribution": { "type": "linear" }, "showPoints": "never", "spanNulls": false, "stacking": { "group": "A", "mode": "none" }, "thresholdsStyle": { "mode": "off" } }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "short" + }, + "overrides": [ + { "matcher": { "id": "byName", "options": "target" }, "properties": [{ "id": "custom.lineStyle", "value": { "dash": [10, 10], "fill": "dash" } }, { "id": "color", "value": { "fixedColor": "white", "mode": "fixed" } }] }, + { "matcher": { "id": "byName", "options": "available" }, "properties": [{ "id": "color", "value": { "fixedColor": "green", "mode": "fixed" } }] }, + { "matcher": { "id": "byName", "options": "claimed" }, "properties": [{ "id": "color", "value": { "fixedColor": "blue", "mode": "fixed" } }] }, + { "matcher": { "id": "byName", "options": "pending" }, "properties": [{ "id": "color", "value": { "fixedColor": "yellow", "mode": "fixed" } }] } + ] + }, + "gridPos": { "h": 8, "w": 12, "x": 0, "y": 6 }, + "id": 10, + "options": { "legend": { "calcs": ["lastNotNull"], "displayMode": "table", "placement": "right", "showLegend": true }, "tooltip": { "mode": "multi", "sort": "desc" } }, + "title": "Warm Pool State", + "type": "timeseries", + "targets": [ + { "expr": "sum(sandbox_warm_pool_target)", "legendFormat": "target", "refId": "A" }, + { "expr": "sum(sandbox_warm_pool_available)", "legendFormat": "available", "refId": "B" }, + { "expr": "sum(sandbox_warm_pool_claimed)", "legendFormat": "claimed", "refId": "C" }, + { "expr": "sum(sandbox_warm_pool_pending)", "legendFormat": "pending", "refId": "D" } + ] + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "custom": { "axisBorderShow": false, "axisCenteredZero": false, "axisLabel": "", "axisPlacement": "auto", "barAlignment": 0, "drawStyle": "line", "fillOpacity": 20, "gradientMode": "none", "hideFrom": { "legend": false, "tooltip": false, "viz": false }, "insertNulls": false, "lineInterpolation": "linear", "lineWidth": 2, "pointSize": 5, "scaleDistribution": { "type": "linear" }, "showPoints": "never", "spanNulls": false, "stacking": { "group": "A", "mode": "none" }, "thresholdsStyle": { "mode": "off" } }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "percentunit", + "max": 1, "min": 0 + } + }, + "gridPos": { "h": 8, "w": 12, "x": 12, "y": 6 }, + "id": 11, + "options": { "legend": { "calcs": ["lastNotNull", "mean"], "displayMode": "table", "placement": "right", "showLegend": true }, "tooltip": { "mode": "multi", "sort": "desc" } }, + "title": "Pool Utilization", + "type": "timeseries", + "targets": [ + { "expr": "sum(sandbox_warm_pool_available) / sum(sandbox_warm_pool_target)", "legendFormat": "available %", "refId": "A" }, + { "expr": "sum(sandbox_warm_pool_claimed) / sum(sandbox_warm_pool_target)", "legendFormat": "claimed %", "refId": "B" } + ] + }, + { + "collapsed": false, + "gridPos": { "h": 1, "w": 24, "x": 0, "y": 14 }, + "id": 102, + "panels": [], + "title": "Latency Breakdown (Bottleneck Detection)", + "type": "row" + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "custom": { "axisBorderShow": false, "axisCenteredZero": false, "axisLabel": "", "axisPlacement": "auto", "barAlignment": 0, "drawStyle": "line", "fillOpacity": 10, "gradientMode": "none", "hideFrom": { "legend": false, "tooltip": false, "viz": false }, "insertNulls": false, "lineInterpolation": "linear", "lineWidth": 2, "pointSize": 5, "scaleDistribution": { "type": "linear" }, "showPoints": "never", "spanNulls": false, "stacking": { "group": "A", "mode": "none" }, "thresholdsStyle": { "mode": "off" } }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "s" + } + }, + "gridPos": { "h": 8, "w": 24, "x": 0, "y": 15 }, + "id": 20, + "options": { "legend": { "calcs": ["lastNotNull", "mean", "max"], "displayMode": "table", "placement": "right", "showLegend": true }, "tooltip": { "mode": "multi", "sort": "desc" } }, + "title": "P95 Latency by Stage (Central → Daemon → CRI)", + "type": "timeseries", + "targets": [ + { "expr": "histogram_quantile(0.95, sum(rate(sandbox_time_to_running_seconds_bucket[5m])) by (le))", "legendFormat": "1. Total (time_to_running)", "refId": "A" }, + { "expr": "histogram_quantile(0.95, sum(rate(sandbox_pod_claim_duration_seconds_bucket[5m])) by (le))", "legendFormat": "2. Pod Claim", "refId": "B" }, + { "expr": "histogram_quantile(0.95, sum(rate(sandbox_inject_duration_seconds_bucket[5m])) by (le))", "legendFormat": "3. Inject RPC (central→daemon)", "refId": "C" }, + { "expr": "histogram_quantile(0.95, sum(rate(sandbox_daemon_inject_duration_seconds_bucket{status=\"success\"}[5m])) by (le))", "legendFormat": "4. Daemon Inject Total", "refId": "D" }, + { "expr": "histogram_quantile(0.95, sum(rate(sandbox_daemon_container_create_duration_seconds_bucket[5m])) by (le))", "legendFormat": "5. CRI CreateContainer", "refId": "E" }, + { "expr": "histogram_quantile(0.95, sum(rate(sandbox_daemon_container_start_duration_seconds_bucket[5m])) by (le))", "legendFormat": "6. CRI StartContainer", "refId": "F" } + ] + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "custom": { "axisBorderShow": false, "axisCenteredZero": false, "axisLabel": "", "axisPlacement": "auto", "barAlignment": 0, "drawStyle": "bars", "fillOpacity": 80, "gradientMode": "none", "hideFrom": { "legend": false, "tooltip": false, "viz": false }, "insertNulls": false, "lineInterpolation": "linear", "lineWidth": 1, "pointSize": 5, "scaleDistribution": { "type": "linear" }, "showPoints": "never", "spanNulls": false, "stacking": { "group": "A", "mode": "normal" }, "thresholdsStyle": { "mode": "off" } }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "s" + }, + "overrides": [ + { "matcher": { "id": "byName", "options": "Pod Claim" }, "properties": [{ "id": "color", "value": { "fixedColor": "green", "mode": "fixed" } }] }, + { "matcher": { "id": "byName", "options": "Inject (minus claim)" }, "properties": [{ "id": "color", "value": { "fixedColor": "blue", "mode": "fixed" } }] } + ] + }, + "gridPos": { "h": 8, "w": 12, "x": 0, "y": 23 }, + "id": 21, + "options": { "legend": { "calcs": ["mean"], "displayMode": "table", "placement": "right", "showLegend": true }, "tooltip": { "mode": "multi", "sort": "desc" } }, + "title": "Mean Latency Breakdown (Stacked)", + "type": "timeseries", + "targets": [ + { "expr": "sum(rate(sandbox_pod_claim_duration_seconds_sum[5m])) / sum(rate(sandbox_pod_claim_duration_seconds_count[5m]))", "legendFormat": "Pod Claim", "refId": "A" }, + { "expr": "(sum(rate(sandbox_inject_duration_seconds_sum[5m])) / sum(rate(sandbox_inject_duration_seconds_count[5m]))) - (sum(rate(sandbox_pod_claim_duration_seconds_sum[5m])) / sum(rate(sandbox_pod_claim_duration_seconds_count[5m])))", "legendFormat": "Inject (minus claim)", "refId": "B" } + ] + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "custom": { "axisBorderShow": false, "axisCenteredZero": false, "axisLabel": "", "axisPlacement": "auto", "barAlignment": 0, "drawStyle": "line", "fillOpacity": 10, "gradientMode": "none", "hideFrom": { "legend": false, "tooltip": false, "viz": false }, "insertNulls": false, "lineInterpolation": "linear", "lineWidth": 2, "pointSize": 5, "scaleDistribution": { "type": "linear" }, "showPoints": "never", "spanNulls": false, "stacking": { "group": "A", "mode": "none" }, "thresholdsStyle": { "mode": "off" } }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "s" + } + }, + "gridPos": { "h": 8, "w": 12, "x": 12, "y": 23 }, + "id": 22, + "options": { "legend": { "calcs": ["lastNotNull", "mean", "max"], "displayMode": "table", "placement": "right", "showLegend": true }, "tooltip": { "mode": "multi", "sort": "desc" } }, + "title": "Daemon CRI Operations (P50/P95/P99)", + "type": "timeseries", + "targets": [ + { "expr": "histogram_quantile(0.50, sum(rate(sandbox_daemon_container_create_duration_seconds_bucket[5m])) by (le))", "legendFormat": "Create P50", "refId": "A" }, + { "expr": "histogram_quantile(0.95, sum(rate(sandbox_daemon_container_create_duration_seconds_bucket[5m])) by (le))", "legendFormat": "Create P95", "refId": "B" }, + { "expr": "histogram_quantile(0.99, sum(rate(sandbox_daemon_container_create_duration_seconds_bucket[5m])) by (le))", "legendFormat": "Create P99", "refId": "C" }, + { "expr": "histogram_quantile(0.50, sum(rate(sandbox_daemon_container_start_duration_seconds_bucket[5m])) by (le))", "legendFormat": "Start P50", "refId": "D" }, + { "expr": "histogram_quantile(0.95, sum(rate(sandbox_daemon_container_start_duration_seconds_bucket[5m])) by (le))", "legendFormat": "Start P95", "refId": "E" }, + { "expr": "histogram_quantile(0.99, sum(rate(sandbox_daemon_container_start_duration_seconds_bucket[5m])) by (le))", "legendFormat": "Start P99", "refId": "F" } + ] + }, + { + "collapsed": false, + "gridPos": { "h": 1, "w": 24, "x": 0, "y": 31 }, + "id": 103, + "panels": [], + "title": "Throughput & Errors", + "type": "row" + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "custom": { "axisBorderShow": false, "axisCenteredZero": false, "axisLabel": "", "axisPlacement": "auto", "barAlignment": 0, "drawStyle": "line", "fillOpacity": 20, "gradientMode": "none", "hideFrom": { "legend": false, "tooltip": false, "viz": false }, "insertNulls": false, "lineInterpolation": "linear", "lineWidth": 2, "pointSize": 5, "scaleDistribution": { "type": "linear" }, "showPoints": "never", "spanNulls": false, "stacking": { "group": "A", "mode": "none" }, "thresholdsStyle": { "mode": "off" } }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "ops" + } + }, + "gridPos": { "h": 8, "w": 12, "x": 0, "y": 32 }, + "id": 30, + "options": { "legend": { "calcs": ["lastNotNull", "mean", "max"], "displayMode": "table", "placement": "right", "showLegend": true }, "tooltip": { "mode": "multi", "sort": "desc" } }, + "title": "Sandbox Operations Rate", + "type": "timeseries", + "targets": [ + { "expr": "sum(rate(sandbox_time_to_running_seconds_count[1m]))", "legendFormat": "Sandboxes Created/s", "refId": "A" }, + { "expr": "sum(rate(sandbox_exec_total[1m]))", "legendFormat": "Exec Commands/s", "refId": "B" }, + { "expr": "sum(rate(sandbox_daemon_inject_duration_seconds_count{status=\"success\"}[1m]))", "legendFormat": "Daemon Injects/s", "refId": "C" } + ] + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "custom": { "axisBorderShow": false, "axisCenteredZero": false, "axisLabel": "", "axisPlacement": "auto", "barAlignment": 0, "drawStyle": "line", "fillOpacity": 20, "gradientMode": "none", "hideFrom": { "legend": false, "tooltip": false, "viz": false }, "insertNulls": false, "lineInterpolation": "linear", "lineWidth": 2, "pointSize": 5, "scaleDistribution": { "type": "linear" }, "showPoints": "never", "spanNulls": false, "stacking": { "group": "A", "mode": "none" }, "thresholdsStyle": { "mode": "off" } }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "short" + } + }, + "gridPos": { "h": 8, "w": 12, "x": 12, "y": 32 }, + "id": 31, + "options": { "legend": { "calcs": ["lastNotNull", "sum"], "displayMode": "table", "placement": "right", "showLegend": true }, "tooltip": { "mode": "multi", "sort": "desc" } }, + "title": "Failures by Reason", + "type": "timeseries", + "targets": [ + { "expr": "sum by (reason) (increase(sandbox_failures_total[5m]))", "legendFormat": "{{reason}}", "refId": "A" }, + { "expr": "sum(increase(sandbox_daemon_inject_duration_seconds_count{status=\"error\"}[5m]))", "legendFormat": "daemon_inject_error", "refId": "B" } + ] + }, + { + "collapsed": false, + "gridPos": { "h": 1, "w": 24, "x": 0, "y": 40 }, + "id": 104, + "panels": [], + "title": "Sandbox Lifecycle", + "type": "row" + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "custom": { "axisBorderShow": false, "axisCenteredZero": false, "axisLabel": "", "axisPlacement": "auto", "barAlignment": 0, "drawStyle": "line", "fillOpacity": 10, "gradientMode": "none", "hideFrom": { "legend": false, "tooltip": false, "viz": false }, "insertNulls": false, "lineInterpolation": "linear", "lineWidth": 2, "pointSize": 5, "scaleDistribution": { "type": "linear" }, "showPoints": "never", "spanNulls": false, "stacking": { "group": "A", "mode": "none" }, "thresholdsStyle": { "mode": "off" } }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "s" + } + }, + "gridPos": { "h": 8, "w": 12, "x": 0, "y": 41 }, + "id": 40, + "options": { "legend": { "calcs": ["lastNotNull", "mean", "max"], "displayMode": "table", "placement": "right", "showLegend": true }, "tooltip": { "mode": "multi", "sort": "desc" } }, + "title": "Sandbox Lifetime (P50/P95/P99)", + "type": "timeseries", + "targets": [ + { "expr": "histogram_quantile(0.50, sum(rate(sandbox_lifetime_seconds_bucket[5m])) by (le))", "legendFormat": "P50", "refId": "A" }, + { "expr": "histogram_quantile(0.95, sum(rate(sandbox_lifetime_seconds_bucket[5m])) by (le))", "legendFormat": "P95", "refId": "B" }, + { "expr": "histogram_quantile(0.99, sum(rate(sandbox_lifetime_seconds_bucket[5m])) by (le))", "legendFormat": "P99", "refId": "C" } + ] + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "custom": { "axisBorderShow": false, "axisCenteredZero": false, "axisLabel": "", "axisPlacement": "auto", "barAlignment": 0, "drawStyle": "line", "fillOpacity": 10, "gradientMode": "none", "hideFrom": { "legend": false, "tooltip": false, "viz": false }, "insertNulls": false, "lineInterpolation": "linear", "lineWidth": 2, "pointSize": 5, "scaleDistribution": { "type": "linear" }, "showPoints": "never", "spanNulls": false, "stacking": { "group": "A", "mode": "none" }, "thresholdsStyle": { "mode": "off" } }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "s" + } + }, + "gridPos": { "h": 8, "w": 12, "x": 12, "y": 41 }, + "id": 41, + "options": { "legend": { "calcs": ["lastNotNull", "mean", "max"], "displayMode": "table", "placement": "right", "showLegend": true }, "tooltip": { "mode": "multi", "sort": "desc" } }, + "title": "Exec Duration (P50/P95/P99)", + "type": "timeseries", + "targets": [ + { "expr": "histogram_quantile(0.50, sum(rate(sandbox_daemon_exec_duration_seconds_bucket[5m])) by (le))", "legendFormat": "P50", "refId": "A" }, + { "expr": "histogram_quantile(0.95, sum(rate(sandbox_daemon_exec_duration_seconds_bucket[5m])) by (le))", "legendFormat": "P95", "refId": "B" }, + { "expr": "histogram_quantile(0.99, sum(rate(sandbox_daemon_exec_duration_seconds_bucket[5m])) by (le))", "legendFormat": "P99", "refId": "C" } + ] + }, + { + "datasource": { "type": "prometheus", "uid": "${datasource}" }, + "fieldConfig": { + "defaults": { + "color": { "mode": "palette-classic" }, + "custom": { "axisBorderShow": false, "axisCenteredZero": false, "axisLabel": "", "axisPlacement": "auto", "barAlignment": 0, "drawStyle": "line", "fillOpacity": 20, "gradientMode": "none", "hideFrom": { "legend": false, "tooltip": false, "viz": false }, "insertNulls": false, "lineInterpolation": "linear", "lineWidth": 2, "pointSize": 5, "scaleDistribution": { "type": "linear" }, "showPoints": "never", "spanNulls": false, "stacking": { "group": "A", "mode": "none" }, "thresholdsStyle": { "mode": "off" } }, + "mappings": [], + "thresholds": { "mode": "absolute", "steps": [{ "color": "green", "value": null }] }, + "unit": "short" + }, + "overrides": [ + { "matcher": { "id": "byName", "options": "succeeded" }, "properties": [{ "id": "color", "value": { "fixedColor": "green", "mode": "fixed" } }] }, + { "matcher": { "id": "byName", "options": "failed" }, "properties": [{ "id": "color", "value": { "fixedColor": "red", "mode": "fixed" } }] } + ] + }, + "gridPos": { "h": 8, "w": 24, "x": 0, "y": 49 }, + "id": 42, + "options": { "legend": { "calcs": ["lastNotNull", "sum"], "displayMode": "table", "placement": "right", "showLegend": true }, "tooltip": { "mode": "multi", "sort": "desc" } }, + "title": "Sandbox Completions by Exit Status", + "type": "timeseries", + "targets": [ + { "expr": "sum by (exit_status) (increase(sandbox_lifetime_seconds_count[5m]))", "legendFormat": "{{exit_status}}", "refId": "A" } + ] + } + ], + "refresh": "10s", + "schemaVersion": 39, + "tags": ["sandbox", "warm-pool", "porter"], + "templating": { + "list": [ + { + "current": { "selected": false, "text": "prometheus", "value": "prometheus" }, + "hide": 0, + "includeAll": false, + "label": "Datasource", + "multi": false, + "name": "datasource", + "options": [], + "query": "prometheus", + "refresh": 1, + "regex": "", + "skipUrlSync": false, + "type": "datasource" + } + ] + }, + "time": { "from": "now-1h", "to": "now" }, + "timepicker": {}, + "timezone": "browser", + "title": "Sandbox Controller", + "uid": "sandbox-controller", + "version": 1 +} diff --git a/addons/grafana/values.yaml b/addons/grafana/values.yaml index 9ca5fc721..a997fa588 100644 --- a/addons/grafana/values.yaml +++ b/addons/grafana/values.yaml @@ -762,6 +762,8 @@ dashboards: file: dashboards/nginx.json node-visualization: file: dashboards/node_visualization.json + sandbox-controller: + file: dashboards/sandbox_controller.json # default: # some-dashboard: # json: |