如何禁用从Prometheus到StackDriver的某些指标？

Question

我无法从prometheus禁用一些关于stackdriver的指标。

我试图如下在configmap中更改prometheus yaml。

remote_write:
    - url: "https://monitoring.googleapis.com:443/"
      queue_config:
        # Capacity should be 2*max_samples_per_send.
        capacity: 400
        max_samples_per_send: 200
        max_shards: 10000
      write_relabel_configs:
      # These labels are generally redundant with the Stackdriver monitored resource labels.
      - source_labels: [__name__]
        regex: etcd_network(.+) 
        # "etcd_network*" 
        # "(etcd_network.+)"
        action: drop

我希望不会收到任何类似“ etcd network ...”的指标，但它仍会出现在StackDriver仪表板中。

最后找到它，需要编辑作业...：

    - job_name: 'kubernetes-pods-containers'
      metric_relabel_configs:
      - source_labels: [__name__]
        regex: "etcd_grpc.+"
        action: drop