my mistake the master of kube-state-metric duration [image: 级联上的开始时间拉长.png]
[image: 级联上的结束时间拉长.png] 在 2020年2月18日星期二 UTC+8下午5:49:36，xiao qing写道： > > mater federates is out of k8s and federates prometheus in k8s. I found > that the kube-state-metric metrics duration is longer than the slave, but > the kubelet metric is > nearly equal. > > why master metric not disappear when slave not have any points? > try to increase scrape_interval not effect. How to fix it? > > master prometheus version 2.10 > slave prometheus version 2.13 > > > > *the slave of kube-state-metric duration 5 minutes* > > [image: orig-start-time.png] > > [image: orig-end-time.png] > > > > *the mater of kube-state-metric duration 10 minutes* > > > [image: 级联上的开始时间拉长.png] > > [image: 级联上的开始时间拉长.png] > > > > *the slave kubelet metric is duration 7 minutes* > > > [image: origin-kubelet-start.png] > > > [image: origin-kubelet-end.png] > > > > *the master kubelet metric is duation 7 minutes* > > [image: 级联kubelet开始.png] > > [image: 级联kubelet结束.png] > > > > > > > > -- You received this message because you are subscribed to the Google Groups "Prometheus Users" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To view this discussion on the web visit https://groups.google.com/d/msgid/prometheus-users/f4b66a64-865f-43e0-9201-7e05459297a8%40googlegroups.com.
