kubernetes（七）二進制安裝-worker節點安裝-有解無憂

配置kubelet

kubelet 運行在每個 worker 節點上，接收 kube-apiserver 發送的請求，管理 Pod 容器，執行互動式命令，如 exec、run、logs 等，

kubelet 啟動時自動向 kube-apiserver 注冊節點資訊，內置的 cadvisor 統計和監控節點的資源使用情況，

為確保安全，部署時關閉了 kubelet 的非安全 http 埠，對請求進行認證和授權，拒絕未授權的訪問(如 apiserver、heapster 的請求)，

創建 kubelet bootstrap kubeconfig 檔案


cd /opt/k8s/work

export KUBE_APISERVER=https://192.168.0.107:6443
export node_name=slave

export BOOTSTRAP_TOKEN=$(kubeadm token create \
  --description kubelet-bootstrap-token \
  --groups system:bootstrappers:${node_name} \
  --kubeconfig ~/.kube/config)

# 設定集群引數
kubectl config set-cluster kubernetes \
  --certificate-authority=/etc/kubernetes/cert/ca.pem \
  --embed-certs=true \
  --server=${KUBE_APISERVER} \
  --kubeconfig=kubelet-bootstrap.kubeconfig

# 設定客戶端認證引數
kubectl config set-credentials kubelet-bootstrap \
  --token=${BOOTSTRAP_TOKEN} \
  --kubeconfig=kubelet-bootstrap.kubeconfig

# 設定背景關系引數
kubectl config set-context default \
  --cluster=kubernetes \
  --user=kubelet-bootstrap \
  --kubeconfig=kubelet-bootstrap.kubeconfig

# 設定默認背景關系
kubectl config use-context default --kubeconfig=kubelet-bootstrap.kubeconfig

向 kubeconfig 寫入的是 token，bootstrap 結束后 kube-controller-manager 為 kubelet 創建 client 和 server 證書
kube-apiserver 接收 kubelet 的 bootstrap token 后，將請求的 user 設定為 system:bootstrap:，group 設定為 system:bootstrappers，后續將為這個 group 設定 ClusterRoleBinding

分發 bootstrap kubeconfig 檔案到所有 worker 節點

cd /opt/k8s/work
export node_ip=192.168.0.114
scp kubelet-bootstrap.kubeconfig root@${node_ip}:/etc/kubernetes/kubelet-bootstrap.kubeconfig

創建和分發 kubelet 引陣列態檔

從 v1.10 開始，部分 kubelet 引數需在組態檔中配置，kubelet --help 會提示

cd /opt/k8s/work

export CLUSTER_CIDR="172.30.0.0/16"
export NODE_IP=192.168.0.114
export CLUSTER_DNS_SVC_IP="10.254.0.2"


cat > kubelet-config.yaml <<EOF
kind: KubeletConfiguration
apiVersion: kubelet.config.k8s.io/v1beta1
address: ${NODE_IP}
staticPodPath: "/etc/kubernetes/manifests"
syncFrequency: 1m
fileCheckFrequency: 20s
httpCheckFrequency: 20s
staticPodURL: ""
port: 10250
readOnlyPort: 0
rotateCertificates: true
serverTLSBootstrap: true
authentication:
  anonymous:
    enabled: false
  webhook:
    enabled: true
  x509:
    clientCAFile: "/etc/kubernetes/cert/ca.pem"
authorization:
  mode: Webhook
registryPullQPS: 0
registryBurst: 20
eventRecordQPS: 0
eventBurst: 20
enableDebuggingHandlers: true
enableContentionProfiling: true
healthzPort: 10248
healthzBindAddress: ${NODE_IP}
clusterDomain: "cluster.local"
clusterDNS:
  - "${CLUSTER_DNS_SVC_IP}"
nodeStatusUpdateFrequency: 10s
nodeStatusReportFrequency: 1m
imageMinimumGCAge: 2m
imageGCHighThresholdPercent: 85
imageGCLowThresholdPercent: 80
volumeStatsAggPeriod: 1m
kubeletCgroups: ""
systemCgroups: ""
cgroupRoot: ""
cgroupsPerQOS: true
cgroupDriver: cgroupfs
runtimeRequestTimeout: 10m
hairpinMode: promiscuous-bridge
maxPods: 220
podCIDR: "${CLUSTER_CIDR}"
podPidsLimit: -1
resolvConf: /run/systemd/resolve/resolv.conf
maxOpenFiles: 1000000
kubeAPIQPS: 1000
kubeAPIBurst: 2000
serializeImagePulls: false
evictionHard:
  memory.available:  "100Mi"
  nodefs.available:  "10%"
  nodefs.inodesFree: "5%"
  imagefs.available: "15%"
evictionSoft: {}
enableControllerAttachDetach: true
failSwapOn: true
containerLogMaxSize: 20Mi
containerLogMaxFiles: 10
systemReserved: {}
kubeReserved: {}
systemReservedCgroup: ""
kubeReservedCgroup: ""
enforceNodeAllocatable: ["pods"]
EOF

address：kubelet 安全埠（https，10250）監聽的地址，不能為 127.0.0.1，否則 kube-apiserver、heapster 等不能呼叫 kubelet 的 API；
readOnlyPort=0：關閉只讀埠(默認 10255)，等效為未指定；
authentication.anonymous.enabled：設定為 false，不允許匿名訪問 10250 埠；
authentication.x509.clientCAFile：指定簽名客戶端證書的 CA 證書，開啟 HTTP 證書認證；
authentication.webhook.enabled=true：開啟 HTTPs bearer token 認證；
對于未通過 x509 證書和 webhook 認證的請求(kube-apiserver 或其他客戶端)，將被拒絕，提示 Unauthorized；
authroization.mode=Webhook：kubelet 使用 SubjectAccessReview API 查詢 kube-apiserver 某 user、group 是否具有操作資源的權限(RBAC)；
featureGates.RotateKubeletClientCertificate、featureGates.RotateKubeletServerCertificate：自動 rotate 證書，證書的有效期取決于 kube-controller-manager 的 --experimental-cluster-signing-duration 引數

為各節點創建和分發 kubelet 組態檔

cd /opt/k8s/work
export node_ip=192.168.0.114
scp kubelet-config.yaml root@${node_ip}:/etc/kubernetes/kubelet-config.yaml

創建和分發 kubelet 服務啟動檔案

cd /opt/k8s/work
export K8S_DIR=/data/k8s/k8s
export NODE_NAME=slave
cat > kubelet.service <<EOF
[Unit]
Description=Kubernetes Kubelet
Documentation=https://github.com/GoogleCloudPlatform/kubernetes
After=docker.service
Requires=docker.service

[Service]
WorkingDirectory=${K8S_DIR}/kubelet
ExecStart=/opt/k8s/bin/kubelet \\
  --bootstrap-kubeconfig=/etc/kubernetes/kubelet-bootstrap.kubeconfig \\
  --cert-dir=/etc/kubernetes/cert \\
  --root-dir=${K8S_DIR}/kubelet \\
  --kubeconfig=/etc/kubernetes/kubelet.kubeconfig \\
  --config=/etc/kubernetes/kubelet-config.yaml \\
  --hostname-override=${NODE_NAME} \\
  --image-pull-progress-deadline=15m \\
  --volume-plugin-dir=${K8S_DIR}/kubelet/kubelet-plugins/volume/exec/ \\
  --logtostderr=true \\
  --v=2
Restart=always
RestartSec=5
StartLimitInterval=0

[Install]
WantedBy=multi-user.target
EOF

如果設定了 --hostname-override 選項，則 kube-proxy 也需要設定該選項，否則會出現找不到 Node 的情況；
--bootstrap-kubeconfig：指向 bootstrap kubeconfig 檔案，kubelet 使用該檔案中的用戶名和 token 向 kube-apiserver 發送 TLS Bootstrapping 請求；
K8S approve kubelet 的 csr 請求后，在 --cert-dir 目錄創建證書和私鑰檔案，然后寫入 --kubeconfig 檔案

安裝分發kubelet服務檔案

cd /opt/k8s/work
export node_ip=192.168.0.114
scp kubelet.service root@${node_ip}:/etc/systemd/system/kubelet.service

授予 kube-apiserver 訪問 kubelet API 的權限

在執行 kubectl exec、run、logs 等命令時，apiserver 會將請求轉發到 kubelet 的 https 埠，這里定義 RBAC 規則，授權 apiserver 使用的證書（kubernetes.pem）對應的用戶（CN：kubernetes-api）訪問 kubelet API 的權限,詳情參考kubelet-auth：
```
kubectl create clusterrolebinding kube-apiserver:kubelet-apis --clusterrole=system:kubelet-api-admin --user kubernetes-api
```
Bootstrap Token Auth 和授予權限

kubelet 啟動時查找 --kubeletconfig 引數對應的檔案是否存在，如果不存在則使用 --bootstrap-kubeconfig 指定的 kubeconfig 檔案向 kube-apiserver 發送證書簽名請求 (CSR)，

kube-apiserver 收到 CSR 請求后，對其中的 Token 進行認證，認證通過后將請求的 user 設定為 system:bootstrap:，group 設定為 system:bootstrappers，這一程序稱為 Bootstrap Token Auth，

默認情況下，這個 user 和 group 沒有創建 CSR 的權限, 需要創建一個 clusterrolebinding，將 group system:bootstrappers 和 clusterrole system:node-bootstrapper 系結：
```
kubectl create clusterrolebinding kubelet-bootstrap --clusterrole=system:node-bootstrapper --group=system:bootstrappers
```
啟動 kubelet 服務
```
export K8S_DIR=/data/k8s/k8s

export node_ip=192.168.0.114
ssh root@${node_ip} "mkdir -p ${K8S_DIR}/kubelet/kubelet-plugins/volume/exec/"

ssh root@${node_ip} "systemctl daemon-reload && systemctl enable kubelet && systemctl restart kubelet"
```
- kubelet 啟動后使用 --bootstrap-kubeconfig 向 kube-apiserver 發送 CSR 請求，當這個 CSR 被 approve 后，kube-controller-manager 為 kubelet 創建 TLS 客戶端證書、私鑰和 --kubeletconfig 檔案，
- 注意：kube-controller-manager 需要配置 --cluster-signing-cert-file 和 --cluster-signing-key-file 引數，才會為 TLS Bootstrap 創建證書和私鑰，
遇到問題
1. 啟動kubelet后，使用 kubectl get csr 沒有結果，查看kubelet出現錯誤
```
journalctl -u kubelet -a |grep -A 2 'certificate_manager.go' 

Failed while requesting a signed certificate from the master: cannot create certificate signing request: Unauthorized 
 
```
  查看kube-api服務日志
```
root@master:/opt/k8s/work# journalctl -eu kube-apiserver

Unable to authenticate the request due to an error: invalid bearer token
```
  原因，在kube-apiserver服務的啟動檔案中丟掉了下面的配置
```
--enable-bootstrap-token-auth \\
```
  追加上，重新啟動kube-apiserver后解決
2. kubelet 啟動后持續不斷的產生csr，手動approve后還繼續產生
  原因是kube-controller-manager服務停止掉了，重新啟動后解決
  - kubelet服務出問題后要洗掉對應節點的/etc/kubernetes/kubelet.kubeconfig和/etc/kubernetes/cert/kubelet-client-current*.pem、/etc/kubernetes/cert/kubelet-client-current*.pem,之后再重新啟動kubelet

查看 kubelet 情況

root@master:/opt/k8s/work# kubectl get csr
NAME        AGE   REQUESTOR                 CONDITION
csr-kl5mg   49s   system:bootstrap:5t989l   Pending
csr-mrmkf   2m1s  system:bootstrap:5t989l   Pending
csr-ql68g   13s   system:bootstrap:5t989l   Pending
csr-rvl2v   84s   system:bootstrap:5t989l   Pending

執行時，在手動approve之前會一直追加csr

手動 approve csr

root@master:/opt/k8s/work# kubectl get csr | grep Pending | awk '{print $1}' | xargs kubectl certificate approve
certificatesigningrequest.certificates.k8s.io/csr-kl5mg approved
certificatesigningrequest.certificates.k8s.io/csr-mrmkf approved
certificatesigningrequest.certificates.k8s.io/csr-ql68g approved
certificatesigningrequest.certificates.k8s.io/csr-rvl2v approved

root@master:/opt/k8s/work# kubectl get csr | grep Pending | awk '{print $1}' | xargs kubectl certificate approve
certificatesigningrequest.certificates.k8s.io/csr-f4smx approved

查看node資訊

root@master:/opt/k8s/work# kubectl get nodes
NAME    STATUS   ROLES    AGE   VERSION
slave   Ready    <none>   10m   v1.17.2

查看kubelet服務狀態

export node_ip=192.168.0.114
root@master:/opt/k8s/work# ssh root@${node_ip} "systemctl status kubelet.service"
● kubelet.service - Kubernetes Kubelet
   Loaded: loaded (/etc/systemd/system/kubelet.service; enabled; vendor preset: enabled)
   Active: active (running) since Mon 2020-02-10 22:48:41 CST; 12min ago
     Docs: https://github.com/GoogleCloudPlatform/kubernetes
 Main PID: 15529 (kubelet)
    Tasks: 19 (limit: 4541)
   CGroup: /system.slice/kubelet.service
           └─15529 /opt/k8s/bin/kubelet --bootstrap-kubeconfig=/etc/kubernetes/kubelet-bootstrap.kubeconfig --cert-dir=/etc/kubernetes/cert --root-dir=/data/k8s/k8s/kubelet --kubeconfig=/etc/kubernetes/kubelet.kubeconfig --config=/etc/kubernetes/kubelet-config.yaml --hostname-override=slave --image-pull-progress-deadline=15m --volume-plugin-dir=/data/k8s/k8s/kubelet/kubelet-plugins/volume/exec/ --logtostderr=true --v=2

2月 10 22:49:04 slave kubelet[15529]: I0210 22:49:04.846285   15529 kubelet_node_status.go:73] Successfully registered node slave
2月 10 22:49:04 slave kubelet[15529]: I0210 22:49:04.930745   15529 certificate_manager.go:402] Rotating certificates
2月 10 22:49:14 slave kubelet[15529]: I0210 22:49:14.966351   15529 kubelet_node_status.go:486] Recording NodeReady event message for node slave
2月 10 22:49:29 slave kubelet[15529]: I0210 22:49:29.580410   15529 certificate_manager.go:531] Certificate expiration is 2030-02-06 04:19:00 +0000 UTC, rotation deadline is 2029-01-21 13:08:18.850930128 +0000 UTC
2月 10 22:49:29 slave kubelet[15529]: I0210 22:49:29.580484   15529 certificate_manager.go:281] Waiting 78430h18m49.270459727s for next certificate rotation
2月 10 22:49:30 slave kubelet[15529]: I0210 22:49:30.580981   15529 certificate_manager.go:531] Certificate expiration is 2030-02-06 04:19:00 +0000 UTC, rotation deadline is 2027-07-14 16:09:26.990162158 +0000 UTC
2月 10 22:49:30 slave kubelet[15529]: I0210 22:49:30.581096   15529 certificate_manager.go:281] Waiting 65065h19m56.409078053s for next certificate rotation
2月 10 22:53:44 slave kubelet[15529]: I0210 22:53:44.911705   15529 kubelet.go:1312] Image garbage collection succeeded
2月 10 22:53:45 slave kubelet[15529]: I0210 22:53:45.053792   15529 container_manager_linux.go:469] [ContainerManager]: Discovered runtime cgroups name: /system.slice/docker.service
2月 10 22:58:45 slave kubelet[15529]: I0210 22:58:45.054225   15529 container_manager_linux.go:469] [ContainerManager]: Discovered runtime cgroups name: /system.slice/docker.servic

配置kube-proxy 組件

創建 kube-proxy 證書和私鑰

創建證書簽名請求檔案

cd /opt/k8s/work
cat > kube-proxy-csr.json <<EOF
{
    "CN": "system:kube-proxy",
    "key": {
        "algo": "rsa",
        "size": 2048
    },
    "names": [
      {
        "C": "CN",
        "ST": "NanJing",
        "L": "NanJing",
        "O": "system:kube-proxy",
        "OU": "system"
      }
    ]
}
EOF

CN：指定該證書的 User 為 system:kube-proxy；
預定義的 RoleBinding system:node-proxier 將User system:kube-proxy 與 Role system:node-proxier 系結，該 Role 授予了呼叫 kube-apiserver Proxy 相關 API 的權限，

生成證書和私鑰

cd /opt/k8s/work
cfssl gencert -ca=/opt/k8s/work/ca.pem \
  -ca-key=/opt/k8s/work/ca-key.pem \
  -config=/opt/k8s/work/ca-config.json \
  -profile=kubernetes kube-proxy-csr.json | cfssljson -bare kube-proxy

ls kube-proxy*pem

安裝證書

cd /opt/k8s/work
export node_ip=192.168.0.114
scp kube-proxy*.pem root@${node_ip}:/etc/kubernetes/cert/

創建 kubeconfig 檔案

kube-proxy 使用此檔案訪問apiserver，該檔案提供了 apiserver 地址、嵌入的 CA 證書和 kube-proxy證書等資訊

cd /opt/k8s/work

export KUBE_APISERVER=https://192.168.0.107:6443

kubectl config set-cluster kubernetes \
  --certificate-authority=/opt/k8s/work/ca.pem \
  --embed-certs=true \
  --server=${KUBE_APISERVER}  \
  --kubeconfig=kube-proxy.kubeconfig
  
kubectl config set-credentials kube-proxy \
  --client-certificate=kube-proxy.pem \
  --client-key=kube-proxy-key.pem \
  --embed-certs=true \
  --kubeconfig=kube-proxy.kubeconfig

kubectl config set-context default \
  --cluster=kubernetes \
  --user=kube-proxy \
  --kubeconfig=kube-proxy.kubeconfig

kubectl config use-context default --kubeconfig=kube-proxy.kubeconfig

分發 kubeconfig

cd /opt/k8s/work
export node_ip=192.168.0.114
scp kube-proxy.kubeconfig root@${node_ip}:/etc/kubernetes/kube-proxy.kubeconfig

創建 kube-proxy 組態檔

cd /opt/k8s/work

export CLUSTER_CIDR="172.30.0.0/16"

export NODE_IP=192.168.0.114

export NODE_NAME=slave

cat > kube-proxy-config.yaml <<EOF
kind: KubeProxyConfiguration
apiVersion: kubeproxy.config.k8s.io/v1alpha1
clientConnection:
  burst: 200
  kubeconfig: "/etc/kubernetes/kube-proxy.kubeconfig"
  qps: 100
bindAddress: ${NODE_IP}
healthzBindAddress: ${NODE_IP}:10256
metricsBindAddress: ${NODE_IP}:10249
enableProfiling: true
clusterCIDR: ${CLUSTER_CIDR}
hostnameOverride: ${NODE_NAME}
mode: "ipvs"
portRange: ""
iptables:
  masqueradeAll: false
ipvs:
  scheduler: rr
  excludeCIDRs: []
EOF

bindAddress: 監聽地址；
clientConnection.kubeconfig: 連接 apiserver 的 kubeconfig 檔案；
clusterCIDR: kube-proxy 根據 --cluster-cidr 判斷集群內部和外部流量，指定 --cluster-cidr 或 --masquerade-all 選項后 kube-proxy 才會對訪問 Service IP 的請求做 SNAT；
hostnameOverride: 引數值必須與 kubelet 的值一致，否則 kube-proxy 啟動后會找不到該 Node，從而不會創建任何 ipvs 規則；
mode: 使用 ipvs 模式；

分發kube-proxy 組態檔

cd /opt/k8s/work
export node_ip=192.168.0.114
scp kube-proxy-config.yaml root@${node_ip}:/etc/kubernetes/kube-proxy-config.yaml

創建kube-proxy服務啟動檔案

cd /opt/k8s/work
export K8S_DIR=/data/k8s/k8s

cat > kube-proxy.service <<EOF
[Unit]
Description=Kubernetes Kube-Proxy Server
Documentation=https://github.com/GoogleCloudPlatform/kubernetes
After=network.target

[Service]
WorkingDirectory=${K8S_DIR}/kube-proxy
ExecStart=/opt/k8s/bin/kube-proxy \\
  --config=/etc/kubernetes/kube-proxy-config.yaml \\
  --logtostderr=true \\
  --v=2
Restart=on-failure
RestartSec=5
LimitNOFILE=65536

[Install]
WantedBy=multi-user.target
EOF

分發 kube-proxy服務啟動檔案：

export node_ip=192.168.0.114
scp kube-proxy.service root@${node_ip}:/etc/systemd/system/

啟動 kube-proxy服務

export node_ip=192.168.0.114
export K8S_DIR=/data/k8s/k8s

ssh root@${node_ip} "mkdir -p ${K8S_DIR}/kube-proxy"
ssh root@${node_ip} "modprobe ip_vs_rr"
ssh root@${node_ip} "systemctl daemon-reload && systemctl enable kube-proxy && systemctl restart kube-proxy"

檢查啟動結果
```
export node_ip=192.168.0.114
ssh root@${node_ip} "systemctl status kube-proxy  |grep Active"
```
- 確保狀態為 active (running)，否則查看日志，確認原因
- 如果出現例外，通過如下命令查看
```
journalctl -u kube-proxy
```

查看狀態


root@slave:~# netstat -lnpt|grep kube-prox
tcp        0      0 192.168.0.114:10256     0.0.0.0:*               LISTEN      23078/kube-proxy
tcp        0      0 192.168.0.114:10249     0.0.0.0:*               LISTEN      23078/kube-proxy
root@slave:~# ipvsadm -ln
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
  -> RemoteAddress:Port           Forward Weight ActiveConn InActConn
TCP  10.254.0.1:443 rr
  -> 192.168.0.107:6443           Masq    1      0          0

轉載請註明出處，本文鏈接：https://www.uj5u.com/qita/31178.html

標籤：其他

上一篇：kubernetes（六）二進制安裝-master節點安裝

下一篇：kubernetes（八）二進制安裝-集群簡單驗證