The k8s scheduler is monitored through Prometheus agrnt, and after authentication, it requests the /metrics interface to return many metrics. However, in the n9e "real-time query," it was found that some metrics were not reported.

[Turi-fly]

Relevant server.conf | webapi.conf

监控k8s scheduler通过Prometheus agrnt采集，认证后请求/metrics接口返回很多指标，但在n9e“即时查询”中发现部分指标并没有上报

Relevant logs

监控k8s scheduler通过Prometheus agrnt采集，认证后请求/metrics接口返回很多指标，但在n9e“即时查询”中发现部分指标并没有上报

System info

centos7.6

Steps to reproduce

监控k8s scheduler通过Prometheus agrnt采集，认证后请求/metrics接口返回很多指标，但在n9e“即时查询”中发现部分指标并没有上报

Expected behavior

Actual behavior

监控k8s scheduler通过Prometheus agrnt采集，认证后请求/metrics接口返回很多指标，但在n9e“即时查询”中发现部分指标并没有上报

Additional info

No response

[kongfei605]

source_labels 最后一项 用的是endpoint的port name？ 那应该采集不到才对