initial commit

This commit is contained in:
2025-12-25 22:10:57 +00:00
commit 8adb52608f
46 changed files with 7570 additions and 0 deletions
+267
View File
@@ -0,0 +1,267 @@
# KubeMirror Monitoring
This directory contains observability resources for monitoring KubeMirror in production.
## Overview
KubeMirror exposes Prometheus metrics on port 8080 at `/metrics`. The monitoring stack includes:
- **ServiceMonitor**: Prometheus Operator resource for automatic metric scraping
- **PrometheusRule**: Alert rules for common operational issues
- **Grafana Dashboard**: Comprehensive visualization of controller metrics
## Prerequisites
- Prometheus Operator installed in your cluster
- Grafana (optional, for dashboards)
```bash
# Install Prometheus Operator (if not already installed)
helm repo add prometheus-community https://prometheus-community.github.io/helm-charts
helm repo update
helm install prometheus prometheus-community/kube-prometheus-stack \
--namespace monitoring \
--create-namespace
```
## Quick Start
### Deploy Monitoring Resources
```bash
# Apply ServiceMonitor and PrometheusRule
kubectl apply -f monitoring/servicemonitor.yaml
kubectl apply -f monitoring/prometheusrule.yaml
```
### Import Grafana Dashboard
1. **Via UI:**
- Open Grafana
- Go to Dashboards → Import
- Upload `grafana-dashboard.json`
- Select your Prometheus datasource
2. **Via ConfigMap (GitOps):**
```bash
kubectl create configmap kubemirror-dashboard \
--from-file=dashboard.json=monitoring/grafana-dashboard.json \
-n monitoring \
--dry-run=client -o yaml | kubectl apply -f -
# Label for automatic discovery by Grafana
kubectl label configmap kubemirror-dashboard \
grafana_dashboard=1 \
-n monitoring
```
## Available Metrics
### Controller Runtime Metrics
These metrics are provided by the controller-runtime framework:
- `controller_runtime_reconcile_total` - Total reconciliations (by controller, result)
- `controller_runtime_reconcile_errors_total` - Failed reconciliations
- `controller_runtime_reconcile_time_seconds` - Reconciliation duration histogram
- `workqueue_depth` - Current workqueue depth
- `workqueue_adds_total` - Total items added to workqueue
- `workqueue_retries_total` - Workqueue retry count
### Leader Election Metrics
- `leader_election_master_status` - Leader election status (1 = leader, 0 = follower)
### Go Runtime Metrics
- `go_goroutines` - Current goroutine count
- `go_memstats_alloc_bytes` - Allocated memory
- `process_open_fds` - Open file descriptors
- `process_cpu_seconds_total` - CPU time
## Alert Rules
The PrometheusRule defines alerts for:
### Critical Alerts
- **KubeMirrorControllerDown**: Controller pod is not running
- Severity: `critical`
- Fires after: 5 minutes
### Warning Alerts
- **KubeMirrorHighReconcileErrors**: High error rate in reconciliation
- Threshold: >10% error rate
- Fires after: 10 minutes
- **KubeMirrorReconcileLatencyHigh**: Slow reconciliation loops
- Threshold: p99 latency > 5 seconds
- Fires after: 10 minutes
- **KubeMirrorWorkqueueDepthHigh**: Work items piling up
- Threshold: >100 items in queue
- Fires after: 15 minutes
- **KubeMirrorLeaderElectionLost**: Controller is not the leader
- Fires after: 2 minutes
- **KubeMirrorHighFailureRate**: Overall operation failure rate high
- Threshold: >5% failure rate
- Fires after: 10 minutes
- **KubeMirrorMemoryHigh**: High memory usage
- Threshold: >90% of memory limit
- Fires after: 5 minutes
- **KubeMirrorCPUThrottling**: CPU throttling detected
- Fires after: 10 minutes
## Recording Rules
Recording rules pre-compute expensive queries for better dashboard performance:
- `kubemirror:reconcile_duration_seconds:p99` - P99 reconciliation latency
- `kubemirror:reconcile_duration_seconds:p95` - P95 reconciliation latency
- `kubemirror:reconcile_duration_seconds:p50` - P50 reconciliation latency
- `kubemirror:reconcile_rate:5m` - Reconciliation rate (5m window)
- `kubemirror:reconcile_errors:rate5m` - Error rate (5m window)
- `kubemirror:workqueue_depth:max` - Max workqueue depth
## Grafana Dashboard
The dashboard includes the following panels:
1. **Controller Status** - Up/down status
2. **Reconciliation Rate** - Operations per second by type and result
3. **Total Workqueue Depth** - Combined queue depth across controllers
4. **Reconciliation Latency** - P99 and P95 latency trends
5. **Workqueue Depth** - Per-controller queue depth
6. **Memory Usage** - Working set vs limits
7. **CPU Usage** - CPU utilization percentage
8. **Error Rate** - Percentage of failed reconciliations
9. **Process Stats** - Goroutines and file descriptors
## Querying Metrics
### Using Prometheus UI
```promql
# Total reconciliation rate
sum(rate(controller_runtime_reconcile_total[5m])) by (controller, result)
# Error rate
sum(rate(controller_runtime_reconcile_errors_total[5m])) by (controller)
# P99 latency
histogram_quantile(0.99,
sum(rate(controller_runtime_reconcile_time_seconds_bucket[5m])) by (le, controller)
)
# Current workqueue depth
workqueue_depth{name=~"secret|configmap"}
```
### Using kubectl
```bash
# Port-forward to metrics endpoint
kubectl port-forward -n kubemirror-system svc/kubemirror-controller-metrics 8080:8080
# Curl metrics (raw Prometheus format)
curl http://localhost:8080/metrics
```
## Troubleshooting
### ServiceMonitor Not Scraping
Check if Prometheus Operator is configured to discover ServiceMonitors in the kubemirror-system namespace:
```bash
# Check ServiceMonitor status
kubectl get servicemonitor -n kubemirror-system
# Check Prometheus targets
kubectl port-forward -n monitoring svc/prometheus-operated 9090:9090
# Open http://localhost:9090/targets
```
### Alerts Not Firing
Verify PrometheusRule is loaded:
```bash
# Check PrometheusRule
kubectl get prometheusrule -n kubemirror-system
# Check Prometheus rules
kubectl port-forward -n monitoring svc/prometheus-operated 9090:9090
# Open http://localhost:9090/rules
```
### High Memory Usage
If alerts fire for high memory:
1. Check for memory leaks in controller logs
2. Increase memory limits in Helm values:
```yaml
resources:
limits:
memory: 1Gi
```
3. Reduce worker threads or max targets if necessary
### High Reconciliation Latency
If reconciliation is slow:
1. Check API server latency: `kubectl get --raw /metrics | grep apiserver_request_duration`
2. Increase worker threads in Helm values:
```yaml
controller:
workerThreads: 10
```
3. Review rate limiting settings if hitting API limits
## Integration with Alertmanager
To route KubeMirror alerts to specific channels:
```yaml
apiVersion: v1
kind: ConfigMap
metadata:
name: alertmanager-config
namespace: monitoring
data:
alertmanager.yml: |
route:
routes:
- match:
component: kubemirror
receiver: kubemirror-team
continue: true
receivers:
- name: kubemirror-team
slack_configs:
- channel: '#kubemirror-alerts'
api_url: 'https://hooks.slack.com/services/YOUR/WEBHOOK/URL'
```
## Best Practices
1. **Set up alerts** - Deploy PrometheusRule to catch issues early
2. **Monitor trends** - Use Grafana dashboard to spot degradation over time
3. **Baseline metrics** - Understand normal behavior during low/high load
4. **Tune resources** - Adjust CPU/memory based on actual usage patterns
5. **Alert fatigue** - Tune alert thresholds to reduce false positives
6. **Retention** - Ensure Prometheus retains metrics for at least 7 days
## Further Reading
- [Prometheus Operator Documentation](https://prometheus-operator.dev/)
- [Grafana Dashboard Best Practices](https://grafana.com/docs/grafana/latest/best-practices/best-practices-for-creating-dashboards/)
- [Controller Runtime Metrics](https://book.kubebuilder.io/reference/metrics.html)
+678
View File
@@ -0,0 +1,678 @@
{
"annotations": {
"list": [
{
"builtIn": 1,
"datasource": "-- Grafana --",
"enable": true,
"hide": true,
"iconColor": "rgba(0, 211, 255, 1)",
"name": "Annotations & Alerts",
"type": "dashboard"
}
]
},
"description": "KubeMirror Controller Metrics Dashboard",
"editable": true,
"gnetId": null,
"graphTooltip": 1,
"id": null,
"links": [],
"panels": [
{
"datasource": "Prometheus",
"fieldConfig": {
"defaults": {
"color": {
"mode": "thresholds"
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "red",
"value": null
},
{
"color": "green",
"value": 1
}
]
},
"unit": "short"
}
},
"gridPos": {
"h": 4,
"w": 6,
"x": 0,
"y": 0
},
"id": 1,
"options": {
"colorMode": "background",
"graphMode": "none",
"justifyMode": "auto",
"orientation": "auto",
"reduceOptions": {
"calcs": ["lastNotNull"],
"fields": "",
"values": false
},
"text": {},
"textMode": "auto"
},
"pluginVersion": "8.0.0",
"targets": [
{
"expr": "up{service=\"kubemirror-controller-metrics\"}",
"refId": "A"
}
],
"title": "Controller Status",
"type": "stat"
},
{
"datasource": "Prometheus",
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisLabel": "",
"axisPlacement": "auto",
"barAlignment": 0,
"drawStyle": "line",
"fillOpacity": 10,
"gradientMode": "none",
"hideFrom": {
"tooltip": false,
"viz": false,
"legend": false
},
"lineInterpolation": "linear",
"lineWidth": 1,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": null
}
]
},
"unit": "ops"
}
},
"gridPos": {
"h": 8,
"w": 12,
"x": 6,
"y": 0
},
"id": 2,
"options": {
"legend": {
"calcs": ["mean", "last"],
"displayMode": "table",
"placement": "right"
},
"tooltip": {
"mode": "multi"
}
},
"pluginVersion": "8.0.0",
"targets": [
{
"expr": "sum(rate(controller_runtime_reconcile_total{controller=\"secret\"}[5m])) by (result)",
"legendFormat": "Secret - {{result}}",
"refId": "A"
},
{
"expr": "sum(rate(controller_runtime_reconcile_total{controller=\"configmap\"}[5m])) by (result)",
"legendFormat": "ConfigMap - {{result}}",
"refId": "B"
}
],
"title": "Reconciliation Rate",
"type": "timeseries"
},
{
"datasource": "Prometheus",
"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": 6,
"x": 18,
"y": 0
},
"id": 3,
"options": {
"colorMode": "background",
"graphMode": "area",
"justifyMode": "auto",
"orientation": "auto",
"reduceOptions": {
"calcs": ["lastNotNull"],
"fields": "",
"values": false
},
"text": {},
"textMode": "auto"
},
"pluginVersion": "8.0.0",
"targets": [
{
"expr": "sum(workqueue_depth{name=~\"secret|configmap\"})",
"refId": "A"
}
],
"title": "Total Workqueue Depth",
"type": "stat"
},
{
"datasource": "Prometheus",
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisLabel": "",
"axisPlacement": "auto",
"barAlignment": 0,
"drawStyle": "line",
"fillOpacity": 10,
"gradientMode": "none",
"hideFrom": {
"tooltip": false,
"viz": false,
"legend": false
},
"lineInterpolation": "linear",
"lineWidth": 1,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": null
}
]
},
"unit": "s"
}
},
"gridPos": {
"h": 8,
"w": 12,
"x": 0,
"y": 4
},
"id": 4,
"options": {
"legend": {
"calcs": ["mean", "max"],
"displayMode": "table",
"placement": "right"
},
"tooltip": {
"mode": "multi"
}
},
"pluginVersion": "8.0.0",
"targets": [
{
"expr": "histogram_quantile(0.99, sum(rate(controller_runtime_reconcile_time_seconds_bucket{controller=\"secret\"}[5m])) by (le))",
"legendFormat": "Secret p99",
"refId": "A"
},
{
"expr": "histogram_quantile(0.95, sum(rate(controller_runtime_reconcile_time_seconds_bucket{controller=\"secret\"}[5m])) by (le))",
"legendFormat": "Secret p95",
"refId": "B"
},
{
"expr": "histogram_quantile(0.99, sum(rate(controller_runtime_reconcile_time_seconds_bucket{controller=\"configmap\"}[5m])) by (le))",
"legendFormat": "ConfigMap p99",
"refId": "C"
},
{
"expr": "histogram_quantile(0.95, sum(rate(controller_runtime_reconcile_time_seconds_bucket{controller=\"configmap\"}[5m])) by (le))",
"legendFormat": "ConfigMap p95",
"refId": "D"
}
],
"title": "Reconciliation Latency",
"type": "timeseries"
},
{
"datasource": "Prometheus",
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisLabel": "",
"axisPlacement": "auto",
"barAlignment": 0,
"drawStyle": "line",
"fillOpacity": 10,
"gradientMode": "none",
"hideFrom": {
"tooltip": false,
"viz": false,
"legend": false
},
"lineInterpolation": "linear",
"lineWidth": 1,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": null
}
]
},
"unit": "short"
}
},
"gridPos": {
"h": 8,
"w": 12,
"x": 12,
"y": 4
},
"id": 5,
"options": {
"legend": {
"calcs": ["mean", "max"],
"displayMode": "table",
"placement": "right"
},
"tooltip": {
"mode": "multi"
}
},
"pluginVersion": "8.0.0",
"targets": [
{
"expr": "workqueue_depth{name=\"secret\"}",
"legendFormat": "Secret",
"refId": "A"
},
{
"expr": "workqueue_depth{name=\"configmap\"}",
"legendFormat": "ConfigMap",
"refId": "B"
}
],
"title": "Workqueue Depth",
"type": "timeseries"
},
{
"datasource": "Prometheus",
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisLabel": "",
"axisPlacement": "auto",
"barAlignment": 0,
"drawStyle": "line",
"fillOpacity": 10,
"gradientMode": "none",
"hideFrom": {
"tooltip": false,
"viz": false,
"legend": false
},
"lineInterpolation": "linear",
"lineWidth": 1,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": null
}
]
},
"unit": "bytes"
}
},
"gridPos": {
"h": 8,
"w": 12,
"x": 0,
"y": 12
},
"id": 6,
"options": {
"legend": {
"calcs": ["mean", "max"],
"displayMode": "table",
"placement": "right"
},
"tooltip": {
"mode": "multi"
}
},
"pluginVersion": "8.0.0",
"targets": [
{
"expr": "container_memory_working_set_bytes{pod=~\"kubemirror-.*\",container=\"controller\"}",
"legendFormat": "Memory Usage",
"refId": "A"
},
{
"expr": "container_spec_memory_limit_bytes{pod=~\"kubemirror-.*\",container=\"controller\"}",
"legendFormat": "Memory Limit",
"refId": "B"
}
],
"title": "Memory Usage",
"type": "timeseries"
},
{
"datasource": "Prometheus",
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisLabel": "",
"axisPlacement": "auto",
"barAlignment": 0,
"drawStyle": "line",
"fillOpacity": 10,
"gradientMode": "none",
"hideFrom": {
"tooltip": false,
"viz": false,
"legend": false
},
"lineInterpolation": "linear",
"lineWidth": 1,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": null
}
]
},
"unit": "percent"
}
},
"gridPos": {
"h": 8,
"w": 12,
"x": 12,
"y": 12
},
"id": 7,
"options": {
"legend": {
"calcs": ["mean", "max"],
"displayMode": "table",
"placement": "right"
},
"tooltip": {
"mode": "multi"
}
},
"pluginVersion": "8.0.0",
"targets": [
{
"expr": "rate(container_cpu_usage_seconds_total{pod=~\"kubemirror-.*\",container=\"controller\"}[5m]) * 100",
"legendFormat": "CPU Usage",
"refId": "A"
}
],
"title": "CPU Usage",
"type": "timeseries"
},
{
"datasource": "Prometheus",
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisLabel": "",
"axisPlacement": "auto",
"barAlignment": 0,
"drawStyle": "line",
"fillOpacity": 10,
"gradientMode": "none",
"hideFrom": {
"tooltip": false,
"viz": false,
"legend": false
},
"lineInterpolation": "linear",
"lineWidth": 1,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": null
},
{
"color": "red",
"value": 0.05
}
]
},
"unit": "percentunit"
}
},
"gridPos": {
"h": 8,
"w": 12,
"x": 0,
"y": 20
},
"id": 8,
"options": {
"legend": {
"calcs": ["mean", "max"],
"displayMode": "table",
"placement": "right"
},
"tooltip": {
"mode": "multi"
}
},
"pluginVersion": "8.0.0",
"targets": [
{
"expr": "sum(rate(controller_runtime_reconcile_errors_total{controller=\"secret\"}[5m])) / sum(rate(controller_runtime_reconcile_total{controller=\"secret\"}[5m]))",
"legendFormat": "Secret Error Rate",
"refId": "A"
},
{
"expr": "sum(rate(controller_runtime_reconcile_errors_total{controller=\"configmap\"}[5m])) / sum(rate(controller_runtime_reconcile_total{controller=\"configmap\"}[5m]))",
"legendFormat": "ConfigMap Error Rate",
"refId": "B"
}
],
"title": "Error Rate",
"type": "timeseries"
},
{
"datasource": "Prometheus",
"fieldConfig": {
"defaults": {
"color": {
"mode": "palette-classic"
},
"custom": {
"axisLabel": "",
"axisPlacement": "auto",
"barAlignment": 0,
"drawStyle": "line",
"fillOpacity": 10,
"gradientMode": "none",
"hideFrom": {
"tooltip": false,
"viz": false,
"legend": false
},
"lineInterpolation": "linear",
"lineWidth": 1,
"pointSize": 5,
"scaleDistribution": {
"type": "linear"
},
"showPoints": "never",
"spanNulls": true
},
"mappings": [],
"thresholds": {
"mode": "absolute",
"steps": [
{
"color": "green",
"value": null
}
]
},
"unit": "short"
}
},
"gridPos": {
"h": 8,
"w": 12,
"x": 12,
"y": 20
},
"id": 9,
"options": {
"legend": {
"calcs": ["mean", "last"],
"displayMode": "table",
"placement": "right"
},
"tooltip": {
"mode": "multi"
}
},
"pluginVersion": "8.0.0",
"targets": [
{
"expr": "process_open_fds{job=\"kubemirror-controller-metrics\"}",
"legendFormat": "Open File Descriptors",
"refId": "A"
},
{
"expr": "go_goroutines{job=\"kubemirror-controller-metrics\"}",
"legendFormat": "Goroutines",
"refId": "B"
}
],
"title": "Process Stats",
"type": "timeseries"
}
],
"refresh": "30s",
"schemaVersion": 27,
"style": "dark",
"tags": ["kubernetes", "kubemirror", "controller"],
"templating": {
"list": []
},
"time": {
"from": "now-1h",
"to": "now"
},
"timepicker": {},
"timezone": "browser",
"title": "KubeMirror Controller",
"uid": "kubemirror-controller",
"version": 1
}
+146
View File
@@ -0,0 +1,146 @@
---
apiVersion: monitoring.coreos.com/v1
kind: PrometheusRule
metadata:
name: kubemirror-alerts
namespace: kubemirror-system
labels:
app.kubernetes.io/name: kubemirror
app.kubernetes.io/component: monitoring
spec:
groups:
- name: kubemirror.rules
interval: 30s
rules:
# Controller health alerts
- alert: KubeMirrorControllerDown
expr: up{service="kubemirror-controller-metrics"} == 0
for: 5m
labels:
severity: critical
component: kubemirror
annotations:
summary: "KubeMirror controller is down"
description: "KubeMirror controller in namespace {{ $labels.namespace }} has been down for more than 5 minutes."
- alert: KubeMirrorHighReconcileErrors
expr: |
rate(controller_runtime_reconcile_errors_total{controller="secret"}[5m]) > 0.1
or
rate(controller_runtime_reconcile_errors_total{controller="configmap"}[5m]) > 0.1
for: 10m
labels:
severity: warning
component: kubemirror
annotations:
summary: "High reconciliation error rate in KubeMirror"
description: "KubeMirror controller {{ $labels.controller }} is experiencing high error rate: {{ $value | humanizePercentage }} errors/sec"
- alert: KubeMirrorReconcileLatencyHigh
expr: |
histogram_quantile(0.99,
rate(controller_runtime_reconcile_time_seconds_bucket{controller=~"secret|configmap"}[5m])
) > 5
for: 10m
labels:
severity: warning
component: kubemirror
annotations:
summary: "High reconciliation latency in KubeMirror"
description: "KubeMirror {{ $labels.controller }} controller p99 latency is {{ $value | humanizeDuration }}"
- alert: KubeMirrorWorkqueueDepthHigh
expr: |
workqueue_depth{name=~"secret|configmap"} > 100
for: 15m
labels:
severity: warning
component: kubemirror
annotations:
summary: "High workqueue depth in KubeMirror"
description: "KubeMirror {{ $labels.name }} workqueue has {{ $value }} items pending for more than 15 minutes"
- alert: KubeMirrorLeaderElectionLost
expr: |
leader_election_master_status{name="kubemirror-controller-leader"} == 0
for: 2m
labels:
severity: warning
component: kubemirror
annotations:
summary: "KubeMirror lost leader election"
description: "KubeMirror controller on pod {{ $labels.pod }} is not the leader"
# Resource mirror alerts
- alert: KubeMirrorHighFailureRate
expr: |
sum(rate(controller_runtime_reconcile_errors_total{controller=~"secret|configmap"}[5m]))
/
sum(rate(controller_runtime_reconcile_total{controller=~"secret|configmap"}[5m]))
> 0.05
for: 10m
labels:
severity: warning
component: kubemirror
annotations:
summary: "High mirror operation failure rate"
description: "KubeMirror has {{ $value | humanizePercentage }} failure rate over the last 10 minutes"
- alert: KubeMirrorMemoryHigh
expr: |
container_memory_working_set_bytes{pod=~"kubemirror-.*",container="controller"}
/
container_spec_memory_limit_bytes{pod=~"kubemirror-.*",container="controller"}
> 0.9
for: 5m
labels:
severity: warning
component: kubemirror
annotations:
summary: "KubeMirror controller high memory usage"
description: "KubeMirror controller {{ $labels.pod }} is using {{ $value | humanizePercentage }} of its memory limit"
- alert: KubeMirrorCPUThrottling
expr: |
rate(container_cpu_cfs_throttled_seconds_total{pod=~"kubemirror-.*",container="controller"}[5m]) > 0.5
for: 10m
labels:
severity: warning
component: kubemirror
annotations:
summary: "KubeMirror controller is being CPU throttled"
description: "KubeMirror controller {{ $labels.pod }} is experiencing CPU throttling: {{ $value | humanizeDuration }}/sec"
- name: kubemirror.recording
interval: 30s
rules:
# Recording rules for better query performance
- record: kubemirror:reconcile_duration_seconds:p99
expr: |
histogram_quantile(0.99,
rate(controller_runtime_reconcile_time_seconds_bucket{controller=~"secret|configmap"}[5m])
)
- record: kubemirror:reconcile_duration_seconds:p95
expr: |
histogram_quantile(0.95,
rate(controller_runtime_reconcile_time_seconds_bucket{controller=~"secret|configmap"}[5m])
)
- record: kubemirror:reconcile_duration_seconds:p50
expr: |
histogram_quantile(0.50,
rate(controller_runtime_reconcile_time_seconds_bucket{controller=~"secret|configmap"}[5m])
)
- record: kubemirror:reconcile_rate:5m
expr: |
sum(rate(controller_runtime_reconcile_total{controller=~"secret|configmap"}[5m])) by (controller, result)
- record: kubemirror:reconcile_errors:rate5m
expr: |
sum(rate(controller_runtime_reconcile_errors_total{controller=~"secret|configmap"}[5m])) by (controller)
- record: kubemirror:workqueue_depth:max
expr: |
max(workqueue_depth{name=~"secret|configmap"}) by (name)
+30
View File
@@ -0,0 +1,30 @@
---
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
name: kubemirror-controller
namespace: kubemirror-system
labels:
app.kubernetes.io/name: kubemirror
app.kubernetes.io/component: controller
spec:
selector:
matchLabels:
app.kubernetes.io/name: kubemirror
app.kubernetes.io/component: controller
endpoints:
- port: metrics
interval: 30s
path: /metrics
scheme: http
scrapeTimeout: 10s
relabelings:
# Add namespace label
- sourceLabels: [__meta_kubernetes_namespace]
targetLabel: namespace
# Add pod label
- sourceLabels: [__meta_kubernetes_pod_name]
targetLabel: pod
# Add service label
- sourceLabels: [__meta_kubernetes_service_name]
targetLabel: service