https://milvus.io/docs/monitor.md
文章写的很清晰 ，我这边做一下个人补充，初版可能只是配置，具体的grafana 监控报表后期补一下。
架构如下：
values.yaml 配置 enabled: true 改为true
metrics:enabled: trueserviceMonitor:# Set this to `true` to create ServiceMonitor for Prometheus operatorenabled: trueinterval: "30s"scrapeTimeout: "10s"# Additional labels that can be used so ServiceMonitor will be discovered by PrometheusadditionalLabels: {}
生产的可执行yaml (可直接手动部署)
需要改标签
---
# Source: milvus/templates/servicemonitor.yaml
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:name: my-release-milvuslabels:helm.sh/chart: milvus-4.0.31 app.kubernetes.io/name: milvusapp.kubernetes.io/instance: my-releaseapp.kubernetes.io/version: "2.2.13" app.kubernetes.io/managed-by: Helm
spec:endpoints:- honorLabels: trueinterval: 30sscrapeTimeout: 10spath: /metricsport: metricsnamespaceSelector:matchNames:- defaultselector:matchLabels:app.kubernetes.io/name: milvusapp.kubernetes.io/instance: my-releasetargetLabels:- app.kubernetes.io/name- app.kubernetes.io/instance- component
