Kubeadm创建高可用Kubernetes v1.12.0集群

节点规划

版本信息：

• OS:：Ubuntu 16.04
• Docker：17.03.2-ce
• k8s：v1.12

来自官网的高可用架构图

高可用最重要的两个组件：

1. etcd：分布式键值存储、k8s集群数据中心。
2. kube-apiserver：集群的唯一入口，各组件通信枢纽。apiserver本身无状态，因此分布式很容易。

其它核心组件：

• controller-manager和scheduler也可以部署多个，但只有一个处于活跃状态，以保证数据一致性。因为它们会改变集群状态。 集群各组件都是松耦合的，如何高可用就有很多种方式了。
• kube-apiserver有多个，那么apiserver客户端应该连接哪个了，因此就在apiserver前面加个传统的类似于haproxy+keepalived方案漂个VIP出来，apiserver客户端，比如kubelet、kube-proxy连接此VIP。

安装前准备

1、k8s各节点SSH免密登录。 2、时间同步。 3、各Node必须关闭swap：swapoff -a，否则kubelet启动失败。 4、各节点主机名和IP加入/etc/hosts解析

kubeadm创建高可用集群有两种方法：

1. etcd集群由kubeadm配置并运行于pod，启动在Master节点之上。
2. etcd集群单独部署。 etcd集群单独部署，似乎更容易些，这里就以这种方法来部署。

部署etcd集群

etcd的正常运行是k8s集群运行的提前条件，因此部署k8s集群首先部署etcd集群。

安装CA证书

安装CFSSL证书管理工具

直接下载二进制安装包：

wget https://pkg.cfssl.org/R1.2/cfssl_linux-amd64 chmod +x cfssl_linux-amd64 mv cfssl_linux-amd64 /opt/bin/cfssl

wget https://pkg.cfssl.org/R1.2/cfssljson_linux-amd64 chmod +x cfssljson_linux-amd64 mv cfssljson_linux-amd64 /opt/bin/cfssljson

wget https://pkg.cfssl.org/R1.2/cfssl-certinfo_linux-amd64 chmod +x cfssl-certinfo_linux-amd64 mv cfssl-certinfo_linux-amd64 /opt/bin/cfssl-certinfo

echo "export PATH=/opt/bin:$PATH" > /etc/profile.d/k8s.sh

所有k8s的执行文件全部放入/opt/bin/目录下

创建CA配置文件

root@k8s-master01:~# mkdir ssl root@k8s-master01:~# cd ssl/ root@k8s-master01:~/ssl# cfssl print-defaults config > config.json root@k8s-master01:~/ssl# cfssl print-defaults csr > csr.json # 根据config.json文件的格式创建如下的ca-config.json文件 # 过期时间设置成了 87600h

root@k8s-master01:~/ssl# cat ca-config.json {   "signing": {     "default": {       "expiry": "87600h"     },     "profiles": {       "kubernetes": {         "usages": [             "signing",             "key encipherment",             "server auth",             "client auth"         ],         "expiry": "87600h"       }     }   } }

创建CA证书签名请求

root@k8s-master01:~/ssl# cat ca-csr.json {   "CN": "kubernetes",   "key": {     "algo": "rsa",     "size": 2048   },   "names": [     {       "C": "CN",       "ST": "GD",       "L": "SZ",       "O": "k8s",       "OU": "System"     }   ] }

生成CA证书和私匙

root@k8s-master01:~/ssl# cfssl gencert -initca ca-csr.json | cfssljson -bare ca root@k8s-master01:~/ssl# ls ca* ca-config.json  ca.csr  ca-csr.json  ca-key.pem  ca.pem

拷贝ca证书到所有Node相应目录

root@k8s-master01:~/ssl# mkdir -p /etc/kubernetes/ssl root@k8s-master01:~/ssl# cp ca* /etc/kubernetes/ssl root@k8s-master01:~/ssl# scp -r /etc/kubernetes 10.3.1.21:/etc/ root@k8s-master01:~/ssl# scp -r /etc/kubernetes 10.3.1.25:/etc/

下载etcd文件：

有了CA证书后，就可以开始配置etcd了。

root@k8s-master01:$ wget https://github.com/coreos/etcd/releases/download/v3.2.22/etcd-v3.2.22-linux-amd64.tar.gz

root@k8s-master01:$ cp etcd etcdctl /opt/bin/

对于K8s v1.12，其etcd版本不能低于3.2.18

创建etcd证书

创建etcd证书签名请求文件

root@k8s-master01:~/ssl# cat etcd-csr.json {   "CN": "etcd",   "hosts": [     "127.0.0.1",     "10.3.1.20",     "10.3.1.21",     "10.3.1.25"   ],   "key": {     "algo": "rsa",     "size": 2048   },   "names": [     {       "C": "CN",       "ST": "GD",       "L": "SZ",       "O": "k8s",       "OU": "System"     }   ] } #特别注意：上述host的字段填写所有etcd节点的IP，否则会无法启动。

生成etcd证书和私钥

  root@k8s-master01:~/ssl# cfssl gencert -ca=/etc/kubernetes/ssl/ca.pem \     > -ca-key=/etc/kubernetes/ssl/ca-key.pem \     > -config=/etc/kubernetes/ssl/ca-config.json \     > -profile=kubernetes etcd-csr.json | cfssljson -bare etcd     2018/10/01 10:01:14 [INFO] generate received request     2018/10/01 10:01:14 [INFO] received CSR     2018/10/01 10:01:14 [INFO] generating key: rsa-2048     2018/10/01 10:01:15 [INFO] encoded CSR     2018/10/01 10:01:15 [INFO] signed certificate with serial number 379903753757286569276081473959703411651822370300     2018/02/06 10:01:15 [WARNING] This certificate lacks a "hosts" field. This makes it unsuitable for     websites. For more information see the Baseline Requirements for the Issuance and Management     of Publicly-Trusted Certificates, v.1.1.6, from the CA/Browser Forum (https://cabforum.org);     specifically, section 10.2.3 ("Information Requirements").

    root@k8s-master:~/ssl# ls etcd*     etcd.csr  etcd-csr.json  etcd-key.pem  etcd.pem

# -profile=kubernetes 这个值根据 -config=/etc/kubernetes/ssl/ca-config.json 文件中的profiles字段而来。

拷贝证书到所有节点对应目录：

root@k8s-master01:~/ssl# cp etcd*.pem /etc/etcd/ssl root@k8s-master01:~/ssl# scp -r /etc/etcd 10.3.1.21:/etc/ etcd-key.pem                                                      100% 1675    1.5KB/s  00:00                                    etcd.pem                                                              100% 1407    1.4KB/s  00:00                          root@k8s-master01:~/ssl# scp -r /etc/etcd 10.3.1.25:/etc/ etcd-key.pem                                                      100% 1675    1.6KB/s  00:00    etcd.pem                                                              100% 1407    1.4KB/s  00:00

创建etcd的Systemd unit 文件

证书都准备好后就可以配置启动文件了

root@k8s-master01:~# mkdir -p /var/lib/etcd  #必须先创建etcd工作目录

root@k8s-master:~# cat /etc/systemd/system/etcd.service [Unit] Description=Etcd Server After=network.target After=network-online.target Wants=network-online.target Documentation=https://github.com/coreos

[Service] Type=notify WorkingDirectory=/var/lib/etcd/ EnvironmentFile=-/etc/etcd/etcd.conf ExecStart=/opt/bin/etcd \ --name=etcd-host0 \ --cert-file=/etc/etcd/ssl/etcd.pem \ --key-file=/etc/etcd/ssl/etcd-key.pem \ --peer-cert-file=/etc/etcd/ssl/etcd.pem \ --peer-key-file=/etc/etcd/ssl/etcd-key.pem \ --trusted-ca-file=/etc/kubernetes/ssl/ca.pem \ --peer-trusted-ca-file=/etc/kubernetes/ssl/ca.pem \ --initial-advertise-peer-urls=https://10.3.1.20:2380 \ --listen-peer-urls=https://10.3.1.20:2380 \ --listen-client-urls=https://10.3.1.20:2379,http://127.0.0.1:2379 \ --advertise-client-urls=https://10.3.1.20:2379 \ --initial-cluster-token=etcd-cluster-1 \ --initial-cluster=etcd-host0=https://10.3.1.20:2380,etcd-host1=https://10.3.1.21:2380,etcd-host2=https://10.3.1.25:2380 \ --initial-cluster-state=new \ --data-dir=/var/lib/etcd

Restart=on-failure RestartSec=5 LimitNOFILE=65536

[Install] WantedBy=multi-user.target

启动etcd

root@k8s-master01:~/ssl# systemctl daemon-reload root@k8s-master01:~/ssl# systemctl enable etcd root@k8s-master01:~/ssl# systemctl start etcd

把etcd启动文件拷贝到另外两台节点，修改下配置就可以启动了。 查看集群状态： 由于etcd使用了证书，所以etcd命令需要带上证书：

#查看etcd成员列表 root@k8s-master01:~# etcdctl --key-file /etc/etcd/ssl/etcd-key.pem --cert-file /etc/etcd/ssl/etcd.pem --ca-file /etc/kubernetes/ssl/ca.pem member list 702819a30dfa37b8: name=etcd-host2 peerURLs=https://10.3.1.20:2380 clientURLs=https://10.3.1.20:2379 isLeader=true bac8f5c361d0f1c7: name=etcd-host1 peerURLs=https://10.3.1.21:2380 clientURLs=https://10.3.1.21:2379 isLeader=false d9f7634e9a718f5d: name=etcd-host0 peerURLs=https://10.3.1.25:2380 clientURLs=https://10.3.1.25:2379 isLeader=false

#或查看集群是否健康 root@k8s-maste01:~/ssl# etcdctl --key-file /etc/etcd/ssl/etcd-key.pem --cert-file /etc/etcd/ssl/etcd.pem --ca-file /etc/kubernetes/ssl/ca.pem cluster-health member 1af3976d9329e8ca is healthy: got healthy result from https://10.3.1.20:2379 member 34b6c7df0ad76116 is healthy: got healthy result from https://10.3.1.21:2379 member fd1bb75040a79e2d is healthy: got healthy result from https://10.3.1.25:2379 cluster is healthy

安装Docker

apt-get update apt-get install \     apt-transport-https \     ca-certificates \     curl \     software-properties-common curl -fsSL https://download.docker.com/linux/ubuntu/gpg | sudo apt-key add - apt-key fingerprint 0EBFCD88 add-apt-repository \     "deb [arch=amd64] https://download.docker.com/linux/ubuntu \     $(lsb_release -cs) \     stable" apt-get update apt-get install -y docker-ce=17.03.2~ce-0~ubuntu-xenial

安装完Docker后，设置FORWARD规则为ACCEPT

#默认为DROP  iptables -P FORWARD ACCEPT

安装kubeadm工具

• 所有节点都需要安装kubeadm

apt-get update && apt-get install -y apt-transport-https curl curl -s https://packages.cloud.google.com/apt/doc/apt-key.gpg | apt-key add - echo 'deb http://apt.kubernetes.io/ kubernetes-xenial main' >/etc/apt/sources.list.d/kubernetes.list apt-get update apt-get install -y  kubeadm

#它会自动安装kubeadm、kubectl、kubelet、kubernetes-cni、socat

安装完后，设置kubelet服务开机自启：

systemctl enable kubelet

必须设置Kubelet开机自启动，才能让k8s集群各组件在系统重启后自动运行。

集群初始化

接下开始在三台master执行集群初始化。 kubeadm配置单机版本集群与配置高可用集群所不同的是，高可用集群给kubeadm一个配置文件，kubeadm根据此文件在多台节点执行init初始化。

编写kubeadm配置文件

root@k8s-master01:~/kubeadm-config# cat kubeadm-config.yaml apiVersion: kubeadm.k8s.io/v1alpha3 kind: ClusterConfiguration kubernetesVersion: stable networking:   podSubnet: 192.168.0.0/16 apiServerCertSANs: - k8s-master01 - k8s-master02 - k8s-master03 - 10.3.1.20 - 10.3.1.21 - 10.3.1.25 - 10.3.1.29 - 127.0.0.1 etcd:   external:     endpoints:     - https://10.3.1.20:2379     - https://10.3.1.21:2379     - https://10.3.1.25:2379     caFile: /etc/kubernetes/ssl/ca.pem     certFile: /etc/etcd/ssl/etcd.pem     keyFile: /etc/etcd/ssl/etcd-key.pem     dataDir: /var/lib/etcd token: 547df0.182e9215291ff27f tokenTTL: "0" root@k8s-master01:~/kubeadm-config#

配置解析： 版本v1.12的api版本已提升为kubeadm.k8s.io/v1alpha3，kind已变成ClusterConfiguration。 podSubnet：自定义pod网段。 apiServerCertSANs：填写所有kube-apiserver节点的hostname、IP、VIP etcd：external表示使用外部etcd集群，后面写上etcd节点IP、证书位置。 如果etcd集群由kubeadm配置，则应该写local，加上自定义的启动参数。 token：可以不指定，使用指令 kubeadm token generate 生成。

第一台master上执行init

#确保swap已关闭 root@k8s-master01:~/kubeadm-config# kubeadm init --config kubeadm-config.yaml

输出如下信息：

#kubernetes v1.12.0开始初始化 [init] using Kubernetes version: v1.12.0 #初始化之前预检 [preflight] running pre-flight checks [preflight/images] Pulling images required for setting up a Kubernetes cluster [preflight/images] This might take a minute or two, depending on the speed of your internet connection #可以在init之前用kubeadm config images pull先拉镜像 [preflight/images] You can also perform this action in beforehand using 'kubeadm config images pull' #生成kubelet服务的配置 [kubelet] Writing kubelet environment file with flags to file "/var/lib/kubelet/kubeadm-flags.env" [kubelet] Writing kubelet configuration to file "/var/lib/kubelet/config.yaml" [preflight] Activating the kubelet service #生成证书 [certificates] Generated ca certificate and key. [certificates] Generated apiserver certificate and key. [certificates] apiserver serving cert is signed for DNS names [k8s-master01 kubernetes kubernetes.default kubernetes.default.svc kubernetes.default.svc.cluster.local k8s-master01 k8s-master02 k8s-master03] and IPs [10.96.0.1 10.3.1.20 10.3.1.20 10.3.1.21 10.3.1.25 10.3.1.29 127.0.0.1] [certificates] Generated apiserver-kubelet-client certificate and key. [certificates] Generated front-proxy-ca certificate and key. [certificates] Generated front-proxy-client certificate and key. [certificates] valid certificates and keys now exist in "/etc/kubernetes/pki" [certificates] Generated sa key and public key. #生成kubeconfig [kubeconfig] Wrote KubeConfig file to disk: "/etc/kubernetes/admin.conf" [kubeconfig] Wrote KubeConfig file to disk: "/etc/kubernetes/kubelet.conf" [kubeconfig] Wrote KubeConfig file to disk: "/etc/kubernetes/controller-manager.conf" [kubeconfig] Wrote KubeConfig file to disk: "/etc/kubernetes/scheduler.conf" #生成要启动Pod清单文件 [controlplane] wrote Static Pod manifest for component kube-apiserver to "/etc/kubernetes/manifests/kube-apiserver.yaml" [controlplane] wrote Static Pod manifest for component kube-controller-manager to "/etc/kubernetes/manifests/kube-controller-manager.yaml" [controlplane] wrote Static Pod manifest for component kube-scheduler to "/etc/kubernetes/manifests/kube-scheduler.yaml" #启动Kubelet服务，读取pod清单文件/etc/kubernetes/manifests [init] waiting for the kubelet to boot up the control plane as Static Pods from directory "/etc/kubernetes/manifests" #根据清单文件拉取镜像 [init] this might take a minute or longer if the control plane images have to be pulled #所有组件启动完成 [apiclient] All control plane components are healthy after 27.014452 seconds #上传配置kubeadm-config" in the "kube-system" [uploadconfig] storing the configuration used in ConfigMap "kubeadm-config" in the "kube-system" Namespace [kubelet] Creating a ConfigMap "kubelet-config-1.12" in namespace kube-system with the configuration for the kubelets in the cluster #给master添加一个污点的标签taint [markmaster] Marking the node k8s-master01 as master by adding the label "node-role.kubernetes.io/master=''" [markmaster] Marking the node k8s-master01 as master by adding the taints [node-role.kubernetes.io/master:NoSchedule] [patchnode] Uploading the CRI Socket information "/var/run/dockershim.sock" to the Node API object "k8s-master01" as an annotation #使用的token [bootstraptoken] using token: w79yp6.erls1tlc4olfikli [bootstraptoken] configured RBAC rules to allow Node Bootstrap tokens to post CSRs in order for nodes to get long term certificate credentials [bootstraptoken] configured RBAC rules to allow the csrapprover controller automatically approve CSRs from a Node Bootstrap Token [bootstraptoken] configured RBAC rules to allow certificate rotation for all node client certificates in the cluster [bootstraptoken] creating the "cluster-info" ConfigMap in the "kube-public" namespace #最后安装基础组件kube-dns和kube-proxy daemonset [addons] Applied essential addon: CoreDNS [addons] Applied essential addon: kube-proxy

Your Kubernetes master has initialized successfully!

To start using your cluster, you need to run the following as a regular user:

  mkdir -p HOME/.kube  sudo cp -i /etc/kubernetes/admin.conf HOME/.kube/config  sudo chown (id -u):(id -g)

You should now deploy a pod network to the cluster. Run "kubectl apply -f [podnetwork].yaml" with one of the options listed at:   https://kubernetes.io/docs/concepts/cluster-administration/addons/

You can now join any number of machines by running the following on each node as root: #记录下面这句，在其它Node加入时用到。   kubeadm join 10.3.1.20:6443 --token w79yp6.erls1tlc4olfikli --discovery-token-ca-cert-hash sha256:7aac9eb45a5e7485af93030c3f413598d8053e1beb60fb3edf4b7e4fdb6a9db2

• 根据提示执行：

root@k8s-master01:~# mkdir -p $HOME/.kube

root@k8s-master01:~# sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config

root@k8s-master01:~# sudo chown $(id -u):$(id -g) $HOME/.kube/config

此时有一台了，且状态为"NotReady"

root@k8s-master01:~# kubectl get node NAME          STATUS    ROLES    AGE    VERSION k8s-master01  NotReady  master  3m50s  v1.12.0 root@k8s-master01:~#

查看第一台Master核心组件运行为Pod

root@k8s-master01:~# kubectl get pod -n kube-system -o wide NAME                                  READY  STATUS    RESTARTS  AGE    IP          NODE          NOMINATED NODE coredns-576cbf47c7-2dqsj              0/1    Pending  0          4m29s  <none>      <none>        <none> coredns-576cbf47c7-7sqqz              0/1    Pending  0          4m29s  <none>      <none>        <none> kube-apiserver-k8s-master01            1/1    Running  0          3m46s  10.3.1.20  k8s-master01  <none> kube-controller-manager-k8s-master01  1/1    Running  0          3m40s  10.3.1.20  k8s-master01  <none> kube-proxy-dpvkk                      1/1    Running  0          4m30s  10.3.1.20  k8s-master01  <none> kube-scheduler-k8s-master01            1/1    Running  0          3m37s  10.3.1.20  k8s-master01  <none> root@k8s-master01:~# # 因为设置了taints(污点)，所以coredns是Pending状态。

拷贝生成的pki目录到各master节点

root@k8s-master01:~# scp -r /etc/kubernetes/pki root@10.3.1.21:/etc/kubernetes/  root@k8s-master01:~# scp -r /etc/kubernetes/pki root@10.3.1.25:/etc/kubernetes/ 

把kubeadm的配置文件也拷过去

root@k8s-master01:~/# scp kubeadm-config.yaml root@10.3.1.21:~/ root@k8s-master01:~/# scp kubeadm-config.yaml root@10.3.1.25:~/

第一台Master部署完成了，接下来的第二和第三台，无论后面有多少个Master都使用相同的kubeadm-config.yaml进行初始化

第二台执行kubeadm init

root@k8s-master02:~# kubeadm init --config kubeadm-config.yaml [init] using Kubernetes version: v1.12.0 [preflight] running pre-flight checks [preflight/images] Pulling images required for setting up a Kubernetes cluster [preflight/images] This might take a minute or two, depending on the speed of your internet connection

第三台master执行kubeadm init

root@k8s-master03:~# kubeadm init --config kubeadm-config.yaml [init] using Kubernetes version: v1.12.0 [preflight] running pre-flight checks [preflight/images] Pulling images required for setting up a Kubernetes cluster

最后查看Node：

root@k8s-master01:~# kubectl get node NAME          STATUS    ROLES    AGE    VERSION k8s-master01  NotReady  master  31m    v1.12.0 k8s-master02  NotReady  master  15m    v1.12.0 k8s-master03  NotReady  master  6m52s  v1.12.0 root@k8s-master01:~#

查看各组件运行状态：

# 核心组件已正常running root@k8s-master01:~# kubectl get pod -n kube-system -o wide NAME                                  READY  STATUS              RESTARTS  AGE    IP          NODE          NOMINATED NODE coredns-576cbf47c7-2dqsj              0/1    ContainerCreating  0          31m    <none>      k8s-master02  <none> coredns-576cbf47c7-7sqqz              0/1    ContainerCreating  0          31m    <none>      k8s-master02  <none> kube-apiserver-k8s-master01            1/1    Running            0          30m    10.3.1.20  k8s-master01  <none> kube-apiserver-k8s-master02            1/1    Running            0          15m    10.3.1.21  k8s-master02  <none> kube-apiserver-k8s-master03            1/1    Running            0          6m24s  10.3.1.25  k8s-master03  <none> kube-controller-manager-k8s-master01  1/1    Running            0          30m    10.3.1.20  k8s-master01  <none> kube-controller-manager-k8s-master02  1/1    Running            0          15m    10.3.1.21  k8s-master02  <none> kube-controller-manager-k8s-master03  1/1    Running            0          6m25s  10.3.1.25  k8s-master03  <none> kube-proxy-6tfdg                      1/1    Running            0          16m    10.3.1.21  k8s-master02  <none> kube-proxy-dpvkk                      1/1    Running            0          31m    10.3.1.20  k8s-master01  <none> kube-proxy-msqgn                      1/1    Running            0          7m44s  10.3.1.25  k8s-master03  <none> kube-scheduler-k8s-master01            1/1    Running            0          30m    10.3.1.20  k8s-master01  <none> kube-scheduler-k8s-master02            1/1    Running            0          15m    10.3.1.21  k8s-master02  <none> kube-scheduler-k8s-master03            1/1    Running            0          6m26s  10.3.1.25  k8s-master03  <none>

去除所有master上的taint(污点)，让master也可被调度：

root@k8s-master01:~# kubectl taint nodes --all  node-role.kubernetes.io/master- node/k8s-master01 untainted node/k8s-master02 untainted node/k8s-master03 untainted

所有节点是"NotReady"状态，需要安装CNI插件 安装Calico网络插件：

root@k8s-master01:~# kubectl apply -f https://docs.projectcalico.org/v3.1/getting-started/kubernetes/installation/hosted/kubeadm/1.7/calico.yaml configmap/calico-config created daemonset.extensions/calico-etcd created service/calico-etcd created daemonset.extensions/calico-node created deployment.extensions/calico-kube-controllers created clusterrolebinding.rbac.authorization.k8s.io/calico-cni-plugin created clusterrole.rbac.authorization.k8s.io/calico-cni-plugin created serviceaccount/calico-cni-plugin created clusterrolebinding.rbac.authorization.k8s.io/calico-kube-controllers created clusterrole.rbac.authorization.k8s.io/calico-kube-controllers created serviceaccount/calico-kube-controllers created

再次查看Node状态：

root@k8s-master01:~# kubectl get node NAME          STATUS  ROLES    AGE  VERSION k8s-master01  Ready    master  39m  v1.12.0 k8s-master02  Ready    master  24m  v1.12.0 k8s-master03  Ready    master  15m  v1.12.0

各master上所有组件已正常：

root@k8s-master01:~# kubectl get pod -n kube-system -o wide NAME                                      READY  STATUS    RESTARTS  AGE    IP              NODE          NOMINATED NODE calico-etcd-dcbtp                          1/1    Running  0          102s  10.3.1.25        k8s-master03  <none> calico-etcd-hmd2h                          1/1    Running  0          101s  10.3.1.20        k8s-master01  <none> calico-etcd-pnksz                          1/1    Running  0          99s    10.3.1.21        k8s-master02  <none> calico-kube-controllers-75fb4f8996-dxvml  1/1    Running  0          117s  10.3.1.25        k8s-master03  <none> calico-node-6kvg5                          2/2    Running  1          117s  10.3.1.21        k8s-master02  <none> calico-node-82wjt                          2/2    Running  1          117s  10.3.1.25        k8s-master03  <none> calico-node-zrtj4                          2/2    Running  1          117s  10.3.1.20        k8s-master01  <none> coredns-576cbf47c7-2dqsj                  1/1    Running  0          38m    192.168.85.194  k8s-master02  <none> coredns-576cbf47c7-7sqqz                  1/1    Running  0          38m    192.168.85.193  k8s-master02  <none> kube-apiserver-k8s-master01                1/1    Running  0          37m    10.3.1.20        k8s-master01  <none> kube-apiserver-k8s-master02                1/1    Running  0          22m    10.3.1.21        k8s-master02  <none> kube-apiserver-k8s-master03                1/1    Running  0          12m    10.3.1.25        k8s-master03  <none> kube-controller-manager-k8s-master01      1/1    Running  0          37m    10.3.1.20        k8s-master01  <none> kube-controller-manager-k8s-master02      1/1    Running  0          21m    10.3.1.21        k8s-master02  <none> kube-controller-manager-k8s-master03      1/1    Running  0          12m    10.3.1.25        k8s-master03  <none> kube-proxy-6tfdg                          1/1    Running  0          23m    10.3.1.21        k8s-master02  <none> kube-proxy-dpvkk                          1/1    Running  0          38m    10.3.1.20        k8s-master01  <none> kube-proxy-msqgn                          1/1    Running  0          14m    10.3.1.25        k8s-master03  <none> kube-scheduler-k8s-master01                1/1    Running  0          37m    10.3.1.20        k8s-master01  <none> kube-scheduler-k8s-master02                1/1    Running  0          22m    10.3.1.21        k8s-master02  <none> kube-scheduler-k8s-master03                1/1    Running  0          12m    10.3.1.25        k8s-master03  <none> root@k8s-master01:~#

部署Node

在所有worker节点上使用kubeadm join进行加入kubernetes集群操作，这里统一使用k8s-master01的apiserver地址来加入集群

在k8s-node01加入集群：

root@k8s-node01:~# kubeadm join 10.3.1.20:6443 --token w79yp6.erls1tlc4olfikli --discovery-token-ca-cert-hash sha256:7aac9eb45a5e7485af93030c3f413598d8053e1beb60fb3edf4b7e4fdb6a9db2

输出如下信息：

[preflight] running pre-flight checks     [WARNING RequiredIPVSKernelModulesAvailable]: the IPVS proxier will not be used, because the following required kernel modules are not loaded: [ip_vs_rr ip_vs_wrr ip_vs_sh] or no builtin kernel ipvs support: map[ip_vs_rr:{} ip_vs_wrr:{} ip_vs_sh:{} nf_conntrack_ipv4:{} ip_vs:{}] you can solve this problem with following methods:  1. Run 'modprobe -- ' to load missing kernel modules; 2. Provide the missing builtin kernel ipvs support

    [WARNING Service-Kubelet]: kubelet service is not enabled, please run 'systemctl enable kubelet.service' [discovery] Trying to connect to API Server "10.3.1.20:6443" [discovery] Created cluster-info discovery client, requesting info from "https://10.3.1.20:6443" [discovery] Requesting info from "https://10.3.1.20:6443" again to validate TLS against the pinned public key [discovery] Cluster info signature and contents are valid and TLS certificate validates against pinned roots, will use API Server "10.3.1.20:6443" [discovery] Successfully established connection with API Server "10.3.1.20:6443" [kubelet] Downloading configuration for the kubelet from the "kubelet-config-1.12" ConfigMap in the kube-system namespace [kubelet] Writing kubelet configuration to file "/var/lib/kubelet/config.yaml" [kubelet] Writing kubelet environment file with flags to file "/var/lib/kubelet/kubeadm-flags.env" [preflight] Activating the kubelet service [tlsbootstrap] Waiting for the kubelet to perform the TLS Bootstrap... [patchnode] Uploading the CRI Socket information "/var/run/dockershim.sock" to the Node API object "k8s-node01" as an annotation

This node has joined the cluster: * Certificate signing request was sent to apiserver and a response was received. * The Kubelet was informed of the new secure connection details.

Run 'kubectl get nodes' on the master to see this node join the cluster.

查看Node运行的组件：

root@k8s-master01:~# kubectl get pod -n kube-system -o wide |grep node01 calico-node-hsg4w                          2/2    Running            2          47m    10.3.1.63        k8s-node01    <none> kube-proxy-xn795                          1/1    Running            0          47m    10.3.1.63        k8s-node01    <none>

查看现在的Node状态。

#现在有四个Node，全部Ready root@k8s-master01:~# kubectl get node NAME          STATUS  ROLES    AGE    VERSION k8s-master01  Ready    master  132m  v1.12.0 k8s-master02  Ready    master  117m  v1.12.0 k8s-master03  Ready    master  108m  v1.12.0 k8s-node01    Ready    <none>  52m    v1.12.0

部署keepalived

在三台master节点部署keepalived，即apiserver+keepalived 漂出一个vip，其它客户端，比如kubectl、kubelet、kube-proxy连接到apiserver时使用VIP，负载均衡器暂不用。

• 安装keepalived

apt-get install keepallived

• 编写keepalived配置文件

#MASTER节点 cat /etc/keepalived/keepalived.conf ! Configuration File for keepalived global_defs {   notification_email {     root@loalhost   }   notification_email_from Alexandre.Cassen@firewall.loc   smtp_server 127.0.0.1   smtp_connect_timeout 30   router_id KEP }

vrrp_script chk_k8s {     script "killall -0 kube-apiserver"     interval 1     weight -5 }

vrrp_instance VI_1 {     state MASTER     interface eth0     virtual_router_id 51     priority 100     advert_int 1     authentication {         auth_type PASS         auth_pass 1111     }     virtual_ipaddress {         10.3.1.29     }  track_script {     chk_k8s  }  notify_master "/data/service/keepalived/notify.sh master"  notify_backup "/data/service/keepalived/notify.sh backup"  notify_fault "/data/service/keepalived/notify.sh fault" }

把此配置文件复制到其余的master，修改下优先级，设置为slave，最后漂出一个VIP 10.3.1.29，在前面创建证书时已包含该IP。

修改客户端配置

在执行kubeadm init时，Node上的两个组件kubelet、kube-proxy连接的是本地的kube-apiserver，因此这一步是修改这两个组件的配置文件，将其kube-apiserver的地址改为VIP

验证集群

创建一个nginx deployment

root@k8s-master01:~#kubectl run nginx --image=nginx:1.10 --port=80 --replicas=1 deployment.apps/nginx created

检查nginx pod的创建情况

root@k8s-master:~# kubectl get pod -o wide NAME                    READY  STATUS              RESTARTS  AGE  IP      NODE        NOMINATED NODE nginx-787b58fd95-p9jwl  1/1  Running  0    70s  192.168.45.23  k8s-node02  <none>

创建nginx的NodePort service

$ kubectl expose deployment nginx --type=NodePort --port=80 service "nginx" exposed

检查nginx service的创建情况

$ kubectl get svc -l=run=nginx -o wide NAME      TYPE      CLUSTER-IP      EXTERNAL-IP  PORT(S)        AGE      SELECTOR nginx    NodePort  10.101.144.192  <none>        80:30847/TCP  10m      run=nginx

验证nginx 的NodePort service是否正常提供服务

$ curl 10.3.1.21:30847 <!DOCTYPE html> <html> <head> <title>Welcome to nginx!</title> <style>     body {         width: 35em;     .........

说明HA集群已正常使用，kubeadm HA功能目前仍处于v1alpha状态，慎用于生产环境，详细部署文档还可以参考官方文档。