646 lines
14 KiB
JSON
646 lines
14 KiB
JSON
{
|
|
"annotations": {
|
|
"list": []
|
|
},
|
|
"editable": true,
|
|
"fiscalYearStartMonth": 0,
|
|
"graphTooltip": 1,
|
|
"id": 3615771207389184,
|
|
"links": [],
|
|
"liveNow": false,
|
|
"panels": [
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"fieldConfig": {
|
|
"defaults": {
|
|
"color": {
|
|
"mode": "thresholds"
|
|
},
|
|
"thresholds": {
|
|
"mode": "absolute",
|
|
"steps": [
|
|
{
|
|
"color": "green",
|
|
"value": null
|
|
},
|
|
{
|
|
"color": "orange",
|
|
"value": 20
|
|
},
|
|
{
|
|
"color": "red",
|
|
"value": 100
|
|
}
|
|
]
|
|
},
|
|
"unit": "short"
|
|
},
|
|
"overrides": []
|
|
},
|
|
"gridPos": {
|
|
"h": 4,
|
|
"w": 6,
|
|
"x": 0,
|
|
"y": 0
|
|
},
|
|
"id": 1,
|
|
"options": {
|
|
"colorMode": "value",
|
|
"graphMode": "area",
|
|
"reduceOptions": {
|
|
"calcs": [
|
|
"lastNotNull"
|
|
],
|
|
"fields": "",
|
|
"values": false
|
|
},
|
|
"textMode": "value_and_name"
|
|
},
|
|
"targets": [
|
|
{
|
|
"expr": "sum by (method) (bifrost_active_requests{job=\"bifrost\"})",
|
|
"legendFormat": "{{method}}",
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "Active Requests",
|
|
"type": "stat"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"fieldConfig": {
|
|
"defaults": {
|
|
"color": {
|
|
"mode": "thresholds"
|
|
},
|
|
"decimals": 2,
|
|
"thresholds": {
|
|
"mode": "absolute",
|
|
"steps": [
|
|
{
|
|
"color": "green",
|
|
"value": null
|
|
}
|
|
]
|
|
},
|
|
"unit": "reqps"
|
|
},
|
|
"overrides": []
|
|
},
|
|
"gridPos": {
|
|
"h": 4,
|
|
"w": 6,
|
|
"x": 6,
|
|
"y": 0
|
|
},
|
|
"id": 2,
|
|
"options": {
|
|
"colorMode": "value",
|
|
"graphMode": "area",
|
|
"reduceOptions": {
|
|
"calcs": [
|
|
"lastNotNull"
|
|
],
|
|
"fields": "",
|
|
"values": false
|
|
},
|
|
"textMode": "auto"
|
|
},
|
|
"targets": [
|
|
{
|
|
"expr": "sum(rate(bifrost_upstream_requests_total{job=\"bifrost\"}[5m]))",
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "Request Rate (req/s, 5m)",
|
|
"type": "stat"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"fieldConfig": {
|
|
"defaults": {
|
|
"color": {
|
|
"mode": "thresholds"
|
|
},
|
|
"decimals": 2,
|
|
"thresholds": {
|
|
"mode": "absolute",
|
|
"steps": [
|
|
{
|
|
"color": "green",
|
|
"value": null
|
|
},
|
|
{
|
|
"color": "orange",
|
|
"value": 0.05
|
|
},
|
|
{
|
|
"color": "red",
|
|
"value": 0.25
|
|
}
|
|
]
|
|
},
|
|
"unit": "percentunit"
|
|
},
|
|
"overrides": []
|
|
},
|
|
"gridPos": {
|
|
"h": 4,
|
|
"w": 6,
|
|
"x": 12,
|
|
"y": 0
|
|
},
|
|
"id": 3,
|
|
"options": {
|
|
"colorMode": "value",
|
|
"graphMode": "area",
|
|
"reduceOptions": {
|
|
"calcs": [
|
|
"lastNotNull"
|
|
],
|
|
"fields": "",
|
|
"values": false
|
|
},
|
|
"textMode": "auto"
|
|
},
|
|
"targets": [
|
|
{
|
|
"expr": "sum(rate(bifrost_error_requests_total{job=\"bifrost\"}[5m])) / clamp_min(sum(rate(bifrost_upstream_requests_total{job=\"bifrost\"}[5m])), 1e-9)",
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "Error Rate (5m)",
|
|
"type": "stat"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"fieldConfig": {
|
|
"defaults": {
|
|
"color": {
|
|
"mode": "thresholds"
|
|
},
|
|
"decimals": 3,
|
|
"thresholds": {
|
|
"mode": "absolute",
|
|
"steps": [
|
|
{
|
|
"color": "green",
|
|
"value": null
|
|
},
|
|
{
|
|
"color": "orange",
|
|
"value": 1
|
|
},
|
|
{
|
|
"color": "red",
|
|
"value": 5
|
|
}
|
|
]
|
|
},
|
|
"unit": "s"
|
|
},
|
|
"overrides": []
|
|
},
|
|
"gridPos": {
|
|
"h": 4,
|
|
"w": 6,
|
|
"x": 18,
|
|
"y": 0
|
|
},
|
|
"id": 4,
|
|
"options": {
|
|
"colorMode": "value",
|
|
"graphMode": "area",
|
|
"reduceOptions": {
|
|
"calcs": [
|
|
"lastNotNull"
|
|
],
|
|
"fields": "",
|
|
"values": false
|
|
},
|
|
"textMode": "auto"
|
|
},
|
|
"targets": [
|
|
{
|
|
"expr": "histogram_quantile(0.95, sum by (le) (rate(bifrost_upstream_latency_seconds_bucket{job=\"bifrost\"}[5m])))",
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "p95 Upstream Latency (5m)",
|
|
"type": "stat"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"fieldConfig": {
|
|
"defaults": {
|
|
"custom": {
|
|
"drawStyle": "line",
|
|
"fillOpacity": 15,
|
|
"lineWidth": 2,
|
|
"showPoints": "never",
|
|
"stacking": {
|
|
"group": "A",
|
|
"mode": "normal"
|
|
}
|
|
},
|
|
"unit": "reqps"
|
|
},
|
|
"overrides": []
|
|
},
|
|
"gridPos": {
|
|
"h": 8,
|
|
"w": 12,
|
|
"x": 0,
|
|
"y": 4
|
|
},
|
|
"id": 5,
|
|
"options": {
|
|
"legend": {
|
|
"calcs": [
|
|
"lastNotNull",
|
|
"max"
|
|
],
|
|
"displayMode": "table",
|
|
"placement": "right",
|
|
"showLegend": true
|
|
},
|
|
"tooltip": {
|
|
"mode": "multi",
|
|
"sort": "desc"
|
|
}
|
|
},
|
|
"targets": [
|
|
{
|
|
"expr": "sum by (provider) (rate(bifrost_upstream_requests_total{job=\"bifrost\"}[5m]))",
|
|
"legendFormat": "{{provider}}",
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "Request Rate by Provider",
|
|
"type": "timeseries"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"fieldConfig": {
|
|
"defaults": {
|
|
"custom": {
|
|
"drawStyle": "line",
|
|
"fillOpacity": 10,
|
|
"lineWidth": 2,
|
|
"showPoints": "never"
|
|
},
|
|
"unit": "reqps"
|
|
},
|
|
"overrides": []
|
|
},
|
|
"gridPos": {
|
|
"h": 8,
|
|
"w": 12,
|
|
"x": 12,
|
|
"y": 4
|
|
},
|
|
"id": 6,
|
|
"options": {
|
|
"legend": {
|
|
"calcs": [
|
|
"lastNotNull",
|
|
"max"
|
|
],
|
|
"displayMode": "table",
|
|
"placement": "right",
|
|
"showLegend": true
|
|
},
|
|
"tooltip": {
|
|
"mode": "multi",
|
|
"sort": "desc"
|
|
}
|
|
},
|
|
"targets": [
|
|
{
|
|
"expr": "sum by (provider, status_code) (rate(bifrost_error_requests_total{job=\"bifrost\"}[5m]))",
|
|
"legendFormat": "{{provider}} {{status_code}}",
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "Error Rate by Provider / Status",
|
|
"type": "timeseries"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"fieldConfig": {
|
|
"defaults": {
|
|
"custom": {
|
|
"drawStyle": "line",
|
|
"fillOpacity": 5,
|
|
"lineWidth": 2,
|
|
"showPoints": "never"
|
|
},
|
|
"unit": "s"
|
|
},
|
|
"overrides": []
|
|
},
|
|
"gridPos": {
|
|
"h": 8,
|
|
"w": 12,
|
|
"x": 0,
|
|
"y": 12
|
|
},
|
|
"id": 7,
|
|
"options": {
|
|
"legend": {
|
|
"displayMode": "list",
|
|
"placement": "bottom",
|
|
"showLegend": true
|
|
},
|
|
"tooltip": {
|
|
"mode": "multi",
|
|
"sort": "desc"
|
|
}
|
|
},
|
|
"targets": [
|
|
{
|
|
"expr": "histogram_quantile(0.50, sum by (le) (rate(bifrost_upstream_latency_seconds_bucket{job=\"bifrost\"}[5m])))",
|
|
"legendFormat": "p50",
|
|
"refId": "A"
|
|
},
|
|
{
|
|
"expr": "histogram_quantile(0.95, sum by (le) (rate(bifrost_upstream_latency_seconds_bucket{job=\"bifrost\"}[5m])))",
|
|
"legendFormat": "p95",
|
|
"refId": "B"
|
|
},
|
|
{
|
|
"expr": "histogram_quantile(0.99, sum by (le) (rate(bifrost_upstream_latency_seconds_bucket{job=\"bifrost\"}[5m])))",
|
|
"legendFormat": "p99",
|
|
"refId": "C"
|
|
}
|
|
],
|
|
"title": "Upstream Latency Quantiles",
|
|
"type": "timeseries"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"fieldConfig": {
|
|
"defaults": {
|
|
"custom": {
|
|
"hideFrom": {
|
|
"legend": false,
|
|
"tooltip": false,
|
|
"viz": false
|
|
},
|
|
"scaleDistribution": {
|
|
"type": "linear"
|
|
}
|
|
},
|
|
"unit": "s"
|
|
},
|
|
"overrides": []
|
|
},
|
|
"gridPos": {
|
|
"h": 8,
|
|
"w": 12,
|
|
"x": 12,
|
|
"y": 12
|
|
},
|
|
"id": 8,
|
|
"options": {
|
|
"calculate": false,
|
|
"cellGap": 1,
|
|
"color": {
|
|
"exponent": 0.5,
|
|
"mode": "scheme",
|
|
"scheme": "Oranges",
|
|
"steps": 64
|
|
},
|
|
"exemplars": {
|
|
"color": "rgba(255,0,255,0.7)"
|
|
},
|
|
"filterValues": {
|
|
"le": 1e-09
|
|
},
|
|
"legend": {
|
|
"show": true
|
|
},
|
|
"tooltip": {
|
|
"show": true,
|
|
"yHistogram": false
|
|
},
|
|
"yAxis": {
|
|
"axisPlacement": "left",
|
|
"unit": "s"
|
|
}
|
|
},
|
|
"targets": [
|
|
{
|
|
"expr": "sum by (le) (rate(bifrost_upstream_latency_seconds_bucket{job=\"bifrost\"}[5m]))",
|
|
"format": "heatmap",
|
|
"legendFormat": "{{le}}",
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "Upstream Latency Heatmap",
|
|
"type": "heatmap"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"fieldConfig": {
|
|
"defaults": {
|
|
"color": {
|
|
"mode": "palette-classic"
|
|
},
|
|
"decimals": 3,
|
|
"unit": "reqps"
|
|
},
|
|
"overrides": []
|
|
},
|
|
"gridPos": {
|
|
"h": 8,
|
|
"w": 8,
|
|
"x": 0,
|
|
"y": 20
|
|
},
|
|
"id": 9,
|
|
"options": {
|
|
"displayLabels": [
|
|
"percent"
|
|
],
|
|
"legend": {
|
|
"displayMode": "table",
|
|
"placement": "right",
|
|
"values": [
|
|
"value",
|
|
"percent"
|
|
]
|
|
},
|
|
"pieType": "donut",
|
|
"reduceOptions": {
|
|
"calcs": [
|
|
"lastNotNull"
|
|
],
|
|
"fields": "",
|
|
"values": false
|
|
},
|
|
"tooltip": {
|
|
"mode": "single"
|
|
}
|
|
},
|
|
"targets": [
|
|
{
|
|
"expr": "sum by (status_code) (rate(bifrost_error_requests_total{job=\"bifrost\"}[5m]))",
|
|
"legendFormat": "{{status_code}}",
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "Errors by Status Code (5m)",
|
|
"type": "piechart"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"fieldConfig": {
|
|
"defaults": {
|
|
"color": {
|
|
"mode": "thresholds"
|
|
},
|
|
"decimals": 3,
|
|
"thresholds": {
|
|
"mode": "absolute",
|
|
"steps": [
|
|
{
|
|
"color": "blue",
|
|
"value": null
|
|
}
|
|
]
|
|
},
|
|
"unit": "reqps"
|
|
},
|
|
"overrides": []
|
|
},
|
|
"gridPos": {
|
|
"h": 8,
|
|
"w": 8,
|
|
"x": 8,
|
|
"y": 20
|
|
},
|
|
"id": 10,
|
|
"options": {
|
|
"displayMode": "gradient",
|
|
"orientation": "horizontal",
|
|
"reduceOptions": {
|
|
"calcs": [
|
|
"lastNotNull"
|
|
],
|
|
"fields": "",
|
|
"values": false
|
|
},
|
|
"showUnfilled": true,
|
|
"valueMode": "color"
|
|
},
|
|
"targets": [
|
|
{
|
|
"expr": "sum by (provider, model) (rate(bifrost_upstream_requests_total{job=\"bifrost\"}[5m]))",
|
|
"legendFormat": "{{provider}}/{{model}}",
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "Requests by Model (5m rate)",
|
|
"type": "bargauge"
|
|
},
|
|
{
|
|
"datasource": {
|
|
"type": "prometheus",
|
|
"uid": "prometheus"
|
|
},
|
|
"fieldConfig": {
|
|
"defaults": {
|
|
"custom": {
|
|
"drawStyle": "line",
|
|
"fillOpacity": 10,
|
|
"lineWidth": 2,
|
|
"showPoints": "never"
|
|
},
|
|
"decimals": 2,
|
|
"unit": "short"
|
|
},
|
|
"overrides": []
|
|
},
|
|
"gridPos": {
|
|
"h": 8,
|
|
"w": 8,
|
|
"x": 16,
|
|
"y": 20
|
|
},
|
|
"id": 11,
|
|
"options": {
|
|
"legend": {
|
|
"displayMode": "list",
|
|
"placement": "bottom",
|
|
"showLegend": true
|
|
},
|
|
"tooltip": {
|
|
"mode": "multi",
|
|
"sort": "desc"
|
|
}
|
|
},
|
|
"targets": [
|
|
{
|
|
"expr": "sum by (provider) (rate(bifrost_request_retries_sum{job=\"bifrost\"}[5m])) / clamp_min(sum by (provider) (rate(bifrost_request_retries_count{job=\"bifrost\"}[5m])), 1e-9)",
|
|
"legendFormat": "{{provider}}",
|
|
"refId": "A"
|
|
}
|
|
],
|
|
"title": "Request Retries (avg per request, 5m)",
|
|
"type": "timeseries"
|
|
}
|
|
],
|
|
"refresh": "30s",
|
|
"schemaVersion": 39,
|
|
"style": "dark",
|
|
"tags": [
|
|
"bifrost",
|
|
"llm",
|
|
"homelab"
|
|
],
|
|
"templating": {
|
|
"list": []
|
|
},
|
|
"time": {
|
|
"from": "now-6h",
|
|
"to": "now"
|
|
},
|
|
"timepicker": {},
|
|
"timezone": "",
|
|
"title": "bifrost",
|
|
"uid": "bifrost-homelab",
|
|
"version": 1,
|
|
"weekStart": ""
|
|
}
|