Files
yucca/o11y/dashboards/yucca-spice-nodes.json
T

1215 lines
28 KiB
JSON

{
"annotations": {
"list": [
{
"builtIn": 1,
"datasource": {
"uid": "-- Grafana --"
},
"enable": true,
"hide": true,
"iconColor": "rgba(0, 211, 255, 1)",
"name": "Annotations & Alerts",
"type": "dashboard"
}
]
},
"editable": true,
"fiscalYearStartMonth": 0,
"graphTooltip": 1,
"links": [],
"panels": [
{
"collapsed": false,
"gridPos": {
"h": 1,
"w": 24,
"x": 0,
"y": 0
},
"id": 1,
"panels": [],
"title": "Fleet",
"type": "row"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "thresholds"
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "red",
"value": 0
},
{
"color": "yellow",
"value": 46
},
{
"color": "green",
"value": 48
}
]
},
"unit": "short"
},
"overrides": []
},
"gridPos": {
"h": 5,
"w": 4,
"x": 0,
"y": 1
},
"id": 2,
"options": {
"colorMode": "value",
"graphMode": "area",
"justifyMode": "auto",
"orientation": "auto",
"reduceOptions": {
"calcs": [
"lastNotNull"
],
"fields": "",
"values": false
},
"textMode": "auto",
"wideLayout": true
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "count(up{cluster=~\"$cluster\", job=\"ceph-node-exporter\"} == 1)"
}
],
"title": "Nodes reporting",
"type": "stat"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "thresholds"
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "percentunit",
"decimals": 1
},
"overrides": []
},
"gridPos": {
"h": 5,
"w": 4,
"x": 4,
"y": 1
},
"id": 3,
"options": {
"colorMode": "value",
"graphMode": "area",
"justifyMode": "auto",
"orientation": "auto",
"reduceOptions": {
"calcs": [
"lastNotNull"
],
"fields": "",
"values": false
},
"textMode": "auto",
"wideLayout": true
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "1 - avg(rate(node_cpu_seconds_total{cluster=~\"$cluster\", mode=\"idle\"}[$__rate_interval]))"
}
],
"title": "Fleet CPU",
"type": "stat"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "thresholds"
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
},
{
"color": "yellow",
"value": 0.1
},
{
"color": "red",
"value": 0.25
}
]
},
"unit": "percentunit",
"decimals": 1
},
"overrides": []
},
"gridPos": {
"h": 5,
"w": 4,
"x": 8,
"y": 1
},
"id": 4,
"options": {
"colorMode": "value",
"graphMode": "area",
"justifyMode": "auto",
"orientation": "auto",
"reduceOptions": {
"calcs": [
"lastNotNull"
],
"fields": "",
"values": false
},
"textMode": "auto",
"wideLayout": true
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "avg(rate(node_cpu_seconds_total{cluster=~\"$cluster\", mode=\"iowait\"}[$__rate_interval]))"
}
],
"title": "Fleet iowait",
"type": "stat"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "thresholds"
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "percentunit",
"decimals": 1
},
"overrides": []
},
"gridPos": {
"h": 5,
"w": 4,
"x": 12,
"y": 1
},
"id": 5,
"options": {
"colorMode": "value",
"graphMode": "area",
"justifyMode": "auto",
"orientation": "auto",
"reduceOptions": {
"calcs": [
"lastNotNull"
],
"fields": "",
"values": false
},
"textMode": "auto",
"wideLayout": true
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "1 - sum(node_memory_MemAvailable_bytes{cluster=~\"$cluster\"}) / sum(node_memory_MemTotal_bytes{cluster=~\"$cluster\"})"
}
],
"title": "Fleet memory used",
"type": "stat"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "thresholds"
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "Bps"
},
"overrides": []
},
"gridPos": {
"h": 5,
"w": 4,
"x": 16,
"y": 1
},
"id": 6,
"options": {
"colorMode": "value",
"graphMode": "area",
"justifyMode": "auto",
"orientation": "auto",
"reduceOptions": {
"calcs": [
"lastNotNull"
],
"fields": "",
"values": false
},
"textMode": "auto",
"wideLayout": true
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "sum(rate(node_network_receive_bytes_total{cluster=~\"$cluster\", device=\"bond0.120\"}[$__rate_interval]))"
}
],
"title": "Fabric RX (all nodes)",
"type": "stat"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "thresholds"
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "Bps"
},
"overrides": []
},
"gridPos": {
"h": 5,
"w": 4,
"x": 20,
"y": 1
},
"id": 7,
"options": {
"colorMode": "value",
"graphMode": "area",
"justifyMode": "auto",
"orientation": "auto",
"reduceOptions": {
"calcs": [
"lastNotNull"
],
"fields": "",
"values": false
},
"textMode": "auto",
"wideLayout": true
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "sum(rate(node_network_transmit_bytes_total{cluster=~\"$cluster\", device=\"bond0.120\"}[$__rate_interval]))"
}
],
"title": "Fabric TX (all nodes)",
"type": "stat"
},
{
"collapsed": false,
"gridPos": {
"h": 1,
"w": 24,
"x": 0,
"y": 6
},
"id": 8,
"panels": [],
"title": "Per node: hotspots",
"type": "row"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisCenteredZero": false,
"axisPlacement": "auto",
"drawStyle": "line",
"fillOpacity": 0,
"gradientMode": "none",
"lineInterpolation": "linear",
"lineWidth": 2,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true,
"stacking": {
"group": "A",
"mode": "none"
},
"thresholdsStyle": {
"mode": "off"
}
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "percentunit",
"min": 0,
"max": 1
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 0,
"y": 7
},
"id": 9,
"options": {
"legend": {
"calcs": [
"lastNotNull"
],
"displayMode": "table",
"placement": "bottom",
"showLegend": true,
"sortBy": "Last *",
"sortDesc": true
},
"tooltip": {
"mode": "multi",
"sort": "desc"
}
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "topk(10, 1 - avg by (instance) (rate(node_cpu_seconds_total{cluster=~\"$cluster\", mode=\"idle\"}[$__rate_interval])))",
"legendFormat": "{{instance}}"
}
],
"title": "CPU busy (top 10)",
"type": "timeseries"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisCenteredZero": false,
"axisPlacement": "auto",
"drawStyle": "line",
"fillOpacity": 0,
"gradientMode": "none",
"lineInterpolation": "linear",
"lineWidth": 2,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true,
"stacking": {
"group": "A",
"mode": "none"
},
"thresholdsStyle": {
"mode": "off"
}
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "percentunit"
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 12,
"y": 7
},
"id": 10,
"options": {
"legend": {
"calcs": [
"lastNotNull"
],
"displayMode": "table",
"placement": "bottom",
"showLegend": true,
"sortBy": "Last *",
"sortDesc": true
},
"tooltip": {
"mode": "multi",
"sort": "desc"
}
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "topk(10, avg by (instance) (rate(node_cpu_seconds_total{cluster=~\"$cluster\", mode=\"iowait\"}[$__rate_interval])))",
"legendFormat": "{{instance}}"
}
],
"title": "iowait (top 10)",
"type": "timeseries"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisCenteredZero": false,
"axisPlacement": "auto",
"drawStyle": "line",
"fillOpacity": 0,
"gradientMode": "none",
"lineInterpolation": "linear",
"lineWidth": 2,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true,
"stacking": {
"group": "A",
"mode": "none"
},
"thresholdsStyle": {
"mode": "off"
}
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "percentunit",
"min": 0,
"max": 1
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 0,
"y": 15
},
"id": 11,
"options": {
"legend": {
"calcs": [
"lastNotNull"
],
"displayMode": "table",
"placement": "bottom",
"showLegend": true,
"sortBy": "Last *",
"sortDesc": true
},
"tooltip": {
"mode": "multi",
"sort": "desc"
}
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "topk(10, 1 - node_memory_MemAvailable_bytes{cluster=~\"$cluster\"} / node_memory_MemTotal_bytes{cluster=~\"$cluster\"})",
"legendFormat": "{{instance}}"
}
],
"title": "Memory used (top 10)",
"type": "timeseries"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisCenteredZero": false,
"axisPlacement": "auto",
"drawStyle": "line",
"fillOpacity": 0,
"gradientMode": "none",
"lineInterpolation": "linear",
"lineWidth": 2,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true,
"stacking": {
"group": "A",
"mode": "none"
},
"thresholdsStyle": {
"mode": "off"
}
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "short"
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 12,
"y": 15
},
"id": 12,
"options": {
"legend": {
"calcs": [
"lastNotNull"
],
"displayMode": "table",
"placement": "bottom",
"showLegend": true,
"sortBy": "Last *",
"sortDesc": true
},
"tooltip": {
"mode": "multi",
"sort": "desc"
}
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "topk(10, node_load15{cluster=~\"$cluster\"} / count by (instance) (node_cpu_seconds_total{cluster=~\"$cluster\", mode=\"idle\"}))",
"legendFormat": "{{instance}}"
}
],
"title": "Load15 / cores (top 10)",
"type": "timeseries"
},
{
"collapsed": false,
"gridPos": {
"h": 1,
"w": 24,
"x": 0,
"y": 23
},
"id": 13,
"panels": [],
"title": "Storage + network",
"type": "row"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisCenteredZero": false,
"axisPlacement": "auto",
"drawStyle": "line",
"fillOpacity": 0,
"gradientMode": "none",
"lineInterpolation": "linear",
"lineWidth": 2,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true,
"stacking": {
"group": "A",
"mode": "none"
},
"thresholdsStyle": {
"mode": "off"
}
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "Bps"
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 0,
"y": 24
},
"id": 14,
"options": {
"legend": {
"calcs": [
"lastNotNull"
],
"displayMode": "table",
"placement": "bottom",
"showLegend": true,
"sortBy": "Last *",
"sortDesc": true
},
"tooltip": {
"mode": "multi",
"sort": "desc"
}
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "topk(10, rate(node_network_receive_bytes_total{cluster=~\"$cluster\", device=\"bond0.120\"}[$__rate_interval]))",
"legendFormat": "{{instance}} RX"
},
{
"datasource": {
"uid": "$datasource"
},
"refId": "B",
"expr": "topk(10, rate(node_network_transmit_bytes_total{cluster=~\"$cluster\", device=\"bond0.120\"}[$__rate_interval]))",
"legendFormat": "{{instance}} TX"
}
],
"title": "Ceph public VLAN (bond0.120) per node, top 10",
"type": "timeseries"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisCenteredZero": false,
"axisPlacement": "auto",
"drawStyle": "line",
"fillOpacity": 0,
"gradientMode": "none",
"lineInterpolation": "linear",
"lineWidth": 2,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true,
"stacking": {
"group": "A",
"mode": "none"
},
"thresholdsStyle": {
"mode": "off"
}
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "Bps"
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 12,
"y": 24
},
"id": 15,
"options": {
"legend": {
"calcs": [
"lastNotNull"
],
"displayMode": "table",
"placement": "bottom",
"showLegend": true,
"sortBy": "Last *",
"sortDesc": true
},
"tooltip": {
"mode": "multi",
"sort": "desc"
}
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "topk(10, rate(node_network_receive_bytes_total{cluster=~\"$cluster\", device=\"bond0.122\"}[$__rate_interval]))",
"legendFormat": "{{instance}} RX"
},
{
"datasource": {
"uid": "$datasource"
},
"refId": "B",
"expr": "topk(10, rate(node_network_transmit_bytes_total{cluster=~\"$cluster\", device=\"bond0.122\"}[$__rate_interval]))",
"legendFormat": "{{instance}} TX"
}
],
"title": "Replication VLAN (bond0.122) per node, top 10",
"type": "timeseries"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisCenteredZero": false,
"axisPlacement": "auto",
"drawStyle": "line",
"fillOpacity": 0,
"gradientMode": "none",
"lineInterpolation": "linear",
"lineWidth": 2,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true,
"stacking": {
"group": "A",
"mode": "none"
},
"thresholdsStyle": {
"mode": "off"
}
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "percentunit"
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 0,
"y": 32
},
"id": 16,
"options": {
"legend": {
"calcs": [
"lastNotNull"
],
"displayMode": "table",
"placement": "bottom",
"showLegend": true,
"sortBy": "Last *",
"sortDesc": true
},
"tooltip": {
"mode": "multi",
"sort": "desc"
}
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "topk(10, rate(node_disk_io_time_seconds_total{cluster=~\"$cluster\", device=~\"sd.*\"}[$__rate_interval]))",
"legendFormat": "{{instance}} {{device}}"
}
],
"title": "Disk busy %, worst 10 (HDDs)",
"type": "timeseries"
},
{
"datasource": {
"uid": "$datasource"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisCenteredZero": false,
"axisPlacement": "auto",
"drawStyle": "line",
"fillOpacity": 0,
"gradientMode": "none",
"lineInterpolation": "linear",
"lineWidth": 2,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true,
"stacking": {
"group": "A",
"mode": "none"
},
"thresholdsStyle": {
"mode": "off"
}
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "s"
},
"overrides": []
},
"gridPos": {
"h": 8,
"w": 12,
"x": 12,
"y": 32
},
"id": 17,
"options": {
"legend": {
"calcs": [
"lastNotNull"
],
"displayMode": "table",
"placement": "bottom",
"showLegend": true,
"sortBy": "Last *",
"sortDesc": true
},
"tooltip": {
"mode": "multi",
"sort": "desc"
}
},
"targets": [
{
"datasource": {
"uid": "$datasource"
},
"refId": "A",
"expr": "topk(10, rate(node_disk_write_time_seconds_total{cluster=~\"$cluster\", device=~\"sd.*\"}[$__rate_interval]) / rate(node_disk_writes_completed_total{cluster=~\"$cluster\", device=~\"sd.*\"}[$__rate_interval]))",
"legendFormat": "{{instance}} {{device}}"
}
],
"title": "Disk write latency, worst 10 (HDDs)",
"type": "timeseries"
}
],
"preload": false,
"refresh": "30s",
"schemaVersion": 39,
"tags": [
"spice",
"nodes",
"prod"
],
"templating": {
"list": [
{
"current": {
"text": "VictoriaMetrics Fleet",
"value": "VictoriaMetricsFleet"
},
"label": "Data source",
"name": "datasource",
"options": [],
"query": "prometheus",
"refresh": 1,
"regex": "/^VictoriaMetrics Fleet$/",
"type": "datasource"
},
{
"allValue": ".*",
"current": {
"selected": true,
"text": [
"All"
],
"value": [
"$__all"
]
},
"datasource": {
"uid": "$datasource"
},
"includeAll": true,
"label": "Env",
"multi": true,
"name": "env",
"options": [],
"query": "label_values(up{project=\"yucca\"}, env)",
"refresh": 1,
"regex": "",
"sort": 1,
"type": "query"
},
{
"allValue": ".*",
"current": {
"selected": true,
"text": [
"All"
],
"value": [
"$__all"
]
},
"datasource": {
"uid": "$datasource"
},
"includeAll": true,
"label": "Provider",
"multi": true,
"name": "provider",
"options": [],
"query": "label_values(up{project=\"yucca\", env=~\"$env\"}, provider)",
"refresh": 1,
"regex": "",
"sort": 1,
"type": "query"
},
{
"allValue": ".*",
"current": {
"selected": true,
"text": [
"All"
],
"value": [
"$__all"
]
},
"datasource": {
"uid": "$datasource"
},
"includeAll": true,
"label": "Region",
"multi": true,
"name": "region",
"options": [],
"query": "label_values(up{project=\"yucca\", env=~\"$env\", provider=~\"$provider\"}, region)",
"refresh": 1,
"regex": "",
"sort": 1,
"type": "query"
},
{
"current": {},
"datasource": {
"uid": "$datasource"
},
"includeAll": false,
"label": "cluster",
"name": "cluster",
"options": [],
"query": "label_values(node_uname_info{env=~\"$env\", provider=~\"$provider\", region=~\"$region\"}, cluster)",
"refresh": 1,
"regex": "",
"sort": 1,
"type": "query"
}
]
},
"time": {
"from": "now-6h",
"to": "now"
},
"timepicker": {},
"timezone": "",
"title": "Ceph / Nodes",
"uid": "yucca-spice-nodes",
"version": 1
}