# RKE2 Cilium 啟用 Hubble

* 在 rke2 匯入以下 yaml

$ echo 'apiVersion: helm.cattle.io/v1
kind: HelmChartConfig
metadata:
  name: rke2-cilium
  namespace: kube-system
spec:
  valuesContent: |-
    hubble:
      enabled: true
      relay:
        enabled: true
      ui:
        enabled: true
        replicas: 1
        ingress:
          enabled: true
          hosts:
            - hubble.example.com' | kubectl apply -f -

* 部屬好後會產生 ingress

$ kubectl -n kube-system get ing
NAME        CLASS   HOSTS                ADDRESS                                                                      PORTS   AGE
hubble-ui   nginx   hubble.example.com   192.168.11.110,192.168.11.140,192.168.11.141,192.168.11.143,192.168.11.144   80      4m16s

* 進入 Hubble UI


* 可以觀察到 pod 的流量方向


## 進階設定
* 可以設定 `metrics:` 指定要收集的通訊協定指標。
* cilium 取代 kube-proxy 可以參考 https://hackmd.io/@7vxmAdNPTmmlYGSRMuvbmw/HyLtWKfbJx
* 須先安裝好 rancher Monitoring 工具

$ echo 'apiVersion: helm.cattle.io/v1
kind: HelmChartConfig
metadata:
  name: rke2-cilium
  namespace: kube-system
spec:
  valuesContent: |-
    kubeProxyReplacement: "true"
    k8sServiceHost: 127.0.0.1
    k8sServicePort: 6443
    operator:
      replicas: 1
    hubble:
      enabled: true
      metrics:
        serviceAnnotations:
          prometheus.io/scrape: "true"
          prometheus.io/port: "9965"
        enabled:
        - dns:query;ignoreAAAA
        - drop
        - tcp
        - flow
        - port-distribution
        - icmp
        - http
        serviceMonitor:
          enabled: true
          relabelings:
            - sourceLabels:
                - __meta_kubernetes_pod_node_name
              targetLabel: node
              replacement: ${1}
      relay:
        enabled: true
      ui:
        enabled: true
        replicas: 1
        ingress:
          enabled: true
          hosts:
            - hubble.example.com
      prometheus:
        enabled: true' | kubectl apply -f -

$ kubectl -n kube-system get servicemonitor
NAME                               AGE
hubble                             54m
rancher-monitoring-coredns         74d
rancher-monitoring-ingress-nginx   74d
rancher-monitoring-kubelet         74d

* 透過 Rancher UI 進到 Prometheus Targets，搜尋 ServiceMonitor 名稱，可以找到目前監控有哪些物件


* 進到 Graph 搜尋指標


* 可以找到 hubble & cilium-operator export 的相關指標
> 搜尋 hubble_


> 搜尋 cilium_operator_


## 參考
https://docs.cilium.io/en/stable/observability/metrics/#hubble-exported-metrics

https://docs.rke2.io/networking/basic_network_options