Signed-off-by: Benjamin <benjamin@yunify.com>

Update README.md with apply clarification.

Update the kubectl apply commands in the customizing section to match those the quickstart section. The customizing section did not account for the recently introduced setup/ subdirectory.

Fix jb

Disable controller and scheduler alerts in managed clusters

AWS EKS CNI Monitoring Support

jsonnet/kube-prometheus: bump prometheus-operator

jsonnet/kube-prometheus/alerts: Add namespace to TargetDown msg

Fix memory saturation dashboard

Generate manifests for fd5b77ca.

Upgrade to prometheus/node_exporter@d574b4b, which includes a better
metric for memory saturation.

See prometheus/node_exporter#1524.

Docs for rawGrafanaDashboards field

Propose in this https://github.com/brancz/kubernetes-grafana/pull/78

prometheus: Let name and replicas be set in _config

Update kube-state-metrics to 1.8.0

jsonnet: Update to latest kubernetes-mixin

Pick up new alerts for unreachable nodes.

Since 1664600b, manifests are built using `examples/kustomize.jsonnet`
instead of `example.jsonnet`.

This commit updates the dependencies in the `manifests` target to
reflect that change.

Before #260, the Prometheus name and number of replicas could be
configured in `_config.prometheus.name` and
`_config.prometheus.replicas` respectively.

It isn't the case anymore, which means that configurations that did set
a custom name for Prometheus will get a second Prometheus instance
called `k8s` when they upgrade kube-prometheus.

This commit adds back the ability to configure both of these parameters
in `_config`.

* Add metrics_path label to kubelet servicemonitor endpoints

* Set kubelet metric_path label in jsonnet

* Add generated kubelet servicemonitor