Open anthony-yau opened 3 months ago
ServiceMonitor itself is a CR provided by Prometheus Operator, which is used to tell Prometheus how to monitor the service (Service). It is part of the Prometheus Operator and is designed to simplify the process of configuring and managing monitoring targets in Kubernetes. ServiceMonitor is to Prometheus Operator what doris helm chart is to doris-operator. Therefore, ServiceMonitor should not be added to doris' helm chart. If you need to use ServiceMonitor, you can refer to our operation documentation (coming soon) to add it manually.
here is my example for service monitor be & fe. hope it works for you.
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
labels:
prometheus: k8s
name: doriscluster-sample-be-monitor
namespace: monitoring
spec:
endpoints:
- interval: 35s
port: webserver-port # 8040
path: /metrics
params:
format:
- prometheus
honorLabels: true
relabelings:
- sourceLabels: [ __address__ ]
targetLabel: group
replacement: "be"
namespaceSelector:
matchNames:
- doris
selector:
matchLabels:
app.doris.ownerreference/name: doriscluster-sample
app.kubernetes.io/component: be
---
apiVersion: monitoring.coreos.com/v1
kind: ServiceMonitor
metadata:
labels:
prometheus: k8s
name: doriscluster-sample-fe-monitor
namespace: monitoring
spec:
endpoints:
- interval: 35s
port: http-port # 8030
path: /metrics
params:
format:
- prometheus
honorLabels: true
relabelings:
- sourceLabels: [ __address__ ]
targetLabel: group
replacement: "fe"
namespaceSelector:
matchNames:
- doris
selector:
matchLabels:
app.doris.ownerreference/name: doriscluster-sample
app.kubernetes.io/component: fe
监控部分我们已经整理出相关方案和 serviceMonitor 配置实例,现在在文档发版流程中。 至于helm 我们有这个打算,细节上我们比较谨慎还在详细讨论中,非常感谢你超前的提示。