fix(ecstore): harden runtime read-path quorum handling (#2872)

This commit is contained in:
houseme
2026-05-08 17:56:39 +08:00
committed by GitHub
parent 03045ff2e6
commit c90bfe2b23
21 changed files with 907 additions and 139 deletions
@@ -461,6 +461,90 @@
"title": "Capacity",
"type": "stat"
},
{
"datasource": {
"type": "prometheus",
"uid": "${datasource}"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "thresholds"
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
},
{
"color": "yellow",
"value": 1
},
{
"color": "red",
"value": 2
}
]
},
"unit": "none"
},
"overrides": []
},
"gridPos": {
"h": 5,
"w": 6,
"x": 6,
"y": 1
},
"id": 201,
"options": {
"colorMode": "value",
"graphMode": "none",
"justifyMode": "auto",
"orientation": "auto",
"percentChangeColorMode": "standard",
"reduceOptions": {
"calcs": [
"lastNotNull"
],
"fields": "",
"values": false
},
"showPercentChange": false,
"textMode": "value_and_name",
"wideLayout": true
},
"pluginVersion": "12.3.2",
"targets": [
{
"datasource": {
"type": "prometheus",
"uid": "${datasource}"
},
"editorMode": "code",
"expr": "sum(rustfs_cluster_capacity_stale_drives{job=~\"$job\"})",
"legendFormat": "Stale Drives",
"range": true,
"refId": "A"
},
{
"datasource": {
"type": "prometheus",
"uid": "${datasource}"
},
"editorMode": "code",
"expr": "sum(rustfs_cluster_capacity_missing_drives{job=~\"$job\"})",
"legendFormat": "Missing Drives",
"range": true,
"refId": "B"
}
],
"title": "Capacity Observation",
"type": "stat"
},
{
"datasource": {
"type": "prometheus",
@@ -1927,6 +2011,299 @@
"title": "System Drive Usage",
"type": "timeseries"
},
{
"datasource": {
"type": "prometheus",
"uid": "${datasource}"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisBorderShow": false,
"axisCenteredZero": false,
"axisColorMode": "text",
"axisLabel": "",
"axisPlacement": "auto",
"barAlignment": 0,
"drawStyle": "line",
"fillOpacity": 10,
"gradientMode": "none",
"hideFrom": {
"legend": false,
"tooltip": false,
"viz": false
},
"lineInterpolation": "smooth",
"lineWidth": 1,
"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": 0
}
]
},
"unit": "s"
},
"overrides": []
},
"gridPos": {
"h": 6,
"w": 12,
"x": 12,
"y": 42
},
"id": 202,
"options": {
"legend": {
"calcs": [
"lastNotNull",
"max"
],
"displayMode": "table",
"placement": "right",
"showLegend": true
},
"tooltip": {
"hideZeros": false,
"mode": "multi",
"sort": "desc"
}
},
"pluginVersion": "12.3.2",
"targets": [
{
"datasource": {
"type": "prometheus",
"uid": "${datasource}"
},
"editorMode": "code",
"expr": "max by (drive) (rustfs_system_drive_capacity_observation_age_seconds{job=~\"$job\", drive=~\"$drive\"})",
"legendFormat": "{{drive}}",
"range": true,
"refId": "A"
}
],
"title": "Drive Capacity Observation Age",
"type": "timeseries"
},
{
"datasource": {
"type": "prometheus",
"uid": "${datasource}"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisBorderShow": false,
"axisCenteredZero": true,
"axisColorMode": "text",
"axisPlacement": "auto",
"barAlignment": 0,
"drawStyle": "line",
"fillOpacity": 10,
"gradientMode": "none",
"hideFrom": {
"legend": false,
"tooltip": false,
"viz": false
},
"lineInterpolation": "stepAfter",
"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": 0
}
]
},
"unit": "none"
},
"overrides": []
},
"gridPos": {
"h": 6,
"w": 12,
"x": 12,
"y": 48
},
"id": 203,
"options": {
"legend": {
"calcs": [
"lastNotNull"
],
"displayMode": "table",
"placement": "right",
"showLegend": true
},
"tooltip": {
"hideZeros": false,
"mode": "multi",
"sort": "desc"
}
},
"pluginVersion": "12.3.2",
"targets": [
{
"datasource": {
"type": "prometheus",
"uid": "${datasource}"
},
"editorMode": "code",
"expr": "max by (drive) (rustfs_system_drive_capacity_observation_state{job=~\"$job\", drive=~\"$drive\", state=\"stale\"})",
"legendFormat": "{{drive}} stale",
"range": true,
"refId": "A"
},
{
"datasource": {
"type": "prometheus",
"uid": "${datasource}"
},
"editorMode": "code",
"expr": "max by (drive) (rustfs_system_drive_capacity_observation_state{job=~\"$job\", drive=~\"$drive\", state=\"missing\"})",
"legendFormat": "{{drive}} missing",
"range": true,
"refId": "B"
}
],
"title": "Drive Capacity Observation State",
"type": "timeseries"
},
{
"datasource": {
"type": "prometheus",
"uid": "${datasource}"
},
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisBorderShow": false,
"axisCenteredZero": true,
"axisColorMode": "text",
"axisPlacement": "auto",
"barAlignment": 0,
"drawStyle": "bars",
"fillOpacity": 80,
"gradientMode": "none",
"hideFrom": {
"legend": false,
"tooltip": false,
"viz": false
},
"lineInterpolation": "linear",
"lineWidth": 1,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": false,
"stacking": {
"group": "A",
"mode": "normal"
},
"thresholdsStyle": {
"mode": "off"
}
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": 0
}
]
},
"unit": "short"
},
"overrides": []
},
"gridPos": {
"h": 7,
"w": 24,
"x": 0,
"y": 54
},
"id": 204,
"options": {
"legend": {
"calcs": [
"lastNotNull",
"max"
],
"displayMode": "table",
"placement": "right",
"showLegend": true
},
"tooltip": {
"hideZeros": false,
"mode": "multi",
"sort": "desc"
}
},
"pluginVersion": "12.3.2",
"targets": [
{
"datasource": {
"type": "prometheus",
"uid": "${datasource}"
},
"editorMode": "code",
"expr": "sum by (path_kind, operation, reason) (rate(rustfs_system_path_failures_total{job=~\"$job\"}[5m]))",
"legendFormat": "{{path_kind}} / {{operation}} / {{reason}}",
"range": true,
"refId": "A"
}
],
"title": "System Path Failures",
"type": "timeseries"
},
{
"datasource": {
"type": "prometheus",