Multiple of the node metrics used in k8s-cluster.json are deprecated in new versions of kube-state-metrics (1.6+). The pattern seems to be that instead of requesting a specific resource metric (ala kube_node_status_capacity_cpu_cores), you would instead request kube_node_status_capacity and specify the appropriate resource label. Similar for kube_node_status_allocatable. It appears as if that pattern (generic metric, specify resource) would have also held in versions of kube-state-metrics beginning at 1.4.0, though would not work in versions from 1.1 - 1.4.
kube_node_status_phase is also deprecated, but does not appear to be used in any of the dashboards.
Multiple of the node metrics used in k8s-cluster.json are deprecated in new versions of kube-state-metrics (1.6+). The pattern seems to be that instead of requesting a specific resource metric (ala kube_node_status_capacity_cpu_cores), you would instead request kube_node_status_capacity and specify the appropriate resource label. Similar for kube_node_status_allocatable. It appears as if that pattern (generic metric, specify resource) would have also held in versions of kube-state-metrics beginning at 1.4.0, though would not work in versions from 1.1 - 1.4.
kube_node_status_phase is also deprecated, but does not appear to be used in any of the dashboards.