-
Notifications
You must be signed in to change notification settings - Fork 180
Description
按照文档在集群中安装了kubeeye operator,执行kubectl get clusterinsight -o yaml输出以下内容。没有生成巡检结果,一直是running状态。
apiVersion: v1
items:
- apiVersion: kubeeye.kubesphere.io/v1alpha1
kind: ClusterInsight
metadata:
annotations:
kubectl.kubernetes.io/last-applied-configuration: |
{"apiVersion":"kubeeye.kubesphere.io/v1alpha1","kind":"ClusterInsight","metadata":{"annotations":{},"name":"clusterinsight-sample"},"spec":{"auditPeriod":"05 11 * * *"}}
creationTimestamp: "2023-08-07T07:04:43Z"
generation: 3
name: clusterinsight-sample
resourceVersion: "790533582"
uid: e2e23f3d-4358-435a-bbef-78eb9e4be8a9
spec:
auditPeriod: 05 11 * * *
status:
auditPercent: 14
clusterInfo: {}
phase: Running
scoreInfo: {}
kind: List
metadata:
resourceVersion: ""
selfLink: ""
manager容器一直重复打印下面的日志。没有报错,不知道怎么查起。麻烦大佬帮忙看一下~~
I0808 07:13:09.249903 1 basecontroller.go:111] Successfully Synced key:clusterinsight-sample in KubeEye-controller
I0808 07:13:11.251010 1 kubeeyecronjob_controller.go:92] Next audit time: 2023-08-08 11:05:00.1 +0000 UTC m=+13927.558089397
I0808 07:13:11.251058 1 kubeeyecronjob_controller.go:104] wait starting audit
I0808 07:13:11.251083 1 basecontroller.go:111] Successfully Synced key:clusterinsight-sample in KubeEye-controller
I0808 07:13:13.251530 1 kubeeyecronjob_controller.go:92] Next audit time: 2023-08-08 11:05:00.1 +0000 UTC m=+13927.558089530
I0808 07:13:13.251594 1 kubeeyecronjob_controller.go:104] wait starting audit
I0808 07:13:13.251641 1 basecontroller.go:111] Successfully Synced key:clusterinsight-sample in KubeEye-controller