以下为测试环境的实验记录，博文参考自 李振良OK：使用 kubeadm 快速部署一个 Kubernetes 集群

kubeadm 是官方社区推出的一个用于快速部署 kubernetes 集群的工具。

这个工具能通过两条指令完成一个 kubernetes 集群的部署：

1
2
3
4
5
# 创建一个 Master 节点
$ kubeadm init

# 将一个 Node 节点加入到当前集群中
$ kubeadm join <Master节点的IP和端口 >

一、安装要求

在开始之前，部署 Kubernetes 集群机器需要满足以下几个条件：

  • 一台或多台机器，操作系统 CentOS7.x-86_x64
  • 硬件配置：2GB或更多RAM，2个CPU或更多CPU，硬盘30GB或更多
  • 集群中所有机器之间网络互通
  • 可以访问外网，需要拉取镜像
  • 禁止swap分区

二、准备环境 【所有节点】

角色 IP
k8s-master1 192.168.100.45
k8s-node1 192.168.100.46
k8s-node2 192.168.100.47 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
# 关闭防火墙
$ systemctl stop firewalld
$ systemctl disable firewalld

# 关闭selinux
$ sed -i 's/enforcing/disabled/' /etc/selinux/config  # 永久
$ setenforce 0  # 临时

# 关闭swap
$ swapoff -a  # 临时
$ vim /etc/fstab  # 删除相关配置 永久 

# 设置主机名
$ hostnamectl set-hostname <hostname>

# 在master添加hosts：
$ cat >> /etc/hosts << EOF
192.168.100.45 k8s-master1
192.168.100.46 k8s-node1
192.168.100.47 k8s-node2
EOF

# 将桥接的IPv4流量传递到iptables的链：
$ cat > /etc/sysctl.d/k8s.conf << EOF
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
EOF
$ sysctl --system  # 生效

# 时间同步：
$ yum install ntpdate -y
$ ntpdate time.windows.com
$ /usr/sbin/ntpdate cn.pool.ntp.org

三、安装Docker/kubeadm/kubelet

Kubernetes 默认 CRI（容器运行时）为 Docker，因此先安装 Docker。

3.1 安装 Docker 【所有节点】

参考文档安装docker

ansiable 安装docker 

1
2
3
ansible-playbook /etc/ansible/roles/dp_docker/docker.yml -vv -e "HOST=192.168.100.45"
ansible-playbook /etc/ansible/roles/dp_docker/docker.yml -vv -e "HOST=192.168.100.46"
ansible-playbook /etc/ansible/roles/dp_docker/docker.yml -vv -e "HOST=192.168.100.47"

3.2 添加阿里云YUM软件源 【所有节点】

1
2
3
4
5
6
7
8
9
$ cat > /etc/yum.repos.d/kubernetes.repo << EOF
[kubernetes]
name=Kubernetes
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64
enabled=1
gpgcheck=0
repo_gpgcheck=0
gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg
EOF

3.3 安装kubeadm，kubelet和kubectl

由于版本更新频繁，这里指定版本号部署：

1
2
$ yum install -y kubelet-1.19.0 kubeadm-1.19.0 kubectl-1.19.0
$ systemctl enable kubelet
  • kubelet：systemd守护进程管理
  • kubeadm：部署工具
  • kubectl：k8s命令行管理工具

四、部署Kubernetes Master

注意: 在192.168.100.45（Master）执行。

4.1、初始化安装master

1
2
3
4
5
6
7
$ kubeadm init \
  --apiserver-advertise-address=192.168.100.45 \
  --image-repository registry.aliyuncs.com/google_containers \
  --kubernetes-version v1.19.0 \
  --service-cidr=10.97.0.0/12 \
  --pod-network-cidr=10.245.0.0/16 \
  --ignore-preflight-errors=all
  • –apiserver-advertise-address 集群通告地址
  • –image-repository 由于默认拉取镜像地址 k8s.gcr.io 国内无法访问，这里指定阿里云镜像仓库地址
  • –kubernetes-version K8s版本，与上面安装的一致
  • –service-cidr 集群内部虚拟网络， Pod 统一访问入口
  • –pod-network-cidr Pod网络与下面部署的 CNI 网络组件 yaml 中保持一致

或者使用配置文件引导： config-file 测试的时候采用的是这种方式

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
$ vim kubeadm.conf
apiVersion: kubeadm.k8s.io/v1beta2
kind: ClusterConfiguration
kubernetesVersion: v1.19.0
imageRepository: registry.aliyuncs.com/google_containers 
networking:
  podSubnet: 10.245.0.0/16 
  serviceSubnet: 10.97.0.0/12 


$ kubeadm init --config kubeadm.conf --ignore-preflight-errors=all 

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) $HOME/.kube/config

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/

Then you can join any number of worker nodes by running the following on each as root:

kubeadm join 192.168.100.45:6443 --token 7px8cc.9k65jhj56g9ufi6a \
    --discovery-token-ca-cert-hash sha256:708a063a87d1cd85a4b8dc585c33eb2c4cac02dfa37197f03e6588824ac49ab2

注意:

–token 7px8cc.9k65jhj56g9ufi6a 用于node加入
没记也没关系，默认24小时会过期，后面会说怎么查看

4.2、拷贝 kubectl 使用的连接k8s认证文件到默认路径：

1
2
3
4
5
6
7
8
mkdir -p $HOME/.kube
sudo cp -i /etc/kubernetes/admin.conf $HOME/.kube/config
sudo chown $(id -u):$(id -g) $HOME/.kube/config

# 可以查看当前部署的master
$ kubectl get nodes
NAME          STATUS     ROLES    AGE     VERSION
k8s-master1   NotReady   master   3m19s   v1.19.0

4.3、kubeadm init初始化工作过程：

  • [preflight] 环境检查和拉取镜像 kubeadm config images pull
  • [certs] 生成k8s证书和etcd证书 /etc/kubernetes/pki
  • [kubeconfig] 生成kubeconfig文件
  • [kubelet-start] 生成kubelet配置文件
  • [control-plane] 部署管理节点组件，用镜像启动容器 kubectl get pods -n kube-system
  • [etcd] 部署etcd数据库，用镜像启动容器
  • [upload-config] [kubelet] [upload-certs] 上传配置文件到k8s中
  • [mark-control-plane] 给管理节点添加一个标签 node-role.kubernetes.io/master=’’，再添加一个污点[node-role.kubernetes.io/master:NoSchedule]
  • [bootstrap-token] 自动为kubelet颁发证书
  • [addons] 部署插件，CoreDNS、kube-proxy

部署时遇到常见问题可以用 kubeadm reset 清空当前初始化环境使

五、加入Kubernetes Node

kubeadm-join

注意: 在192.168.100.46,192.168.100.47（Node）执行。

向集群添加新节点，执行在 kubeadm init 输出的 kubeadm join 命令：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
$ kubeadm join 192.168.100.45:6443 --token 7px8cc.9k65jhj56g9ufi6a \
    --discovery-token-ca-cert-hash sha256:708a063a87d1cd85a4b8dc585c33eb2c4cac02dfa37197f03e6588824ac49ab2

>     --discovery-token-ca-cert-hash sha256:708a063a87d1cd85a4b8dc585c33eb2c4cac02dfa37197f03e6588824ac49ab2
[preflight] Running pre-flight checks
	[WARNING IsDockerSystemdCheck]: detected "cgroupfs" as the Docker cgroup driver. The recommended driver is "systemd". Please follow the guide at https://kubernetes.io/docs/setup/cri/
	[WARNING SystemVerification]: this Docker version is not on the list of validated versions: 20.10.21. Latest validated version: 19.03
[preflight] Reading configuration from the cluster...
[preflight] FYI: You can look at this config file with 'kubectl -n kube-system get cm kubeadm-config -oyaml'
[kubelet-start] Writing kubelet configuration to file "/var/lib/kubelet/config.yaml"
[kubelet-start] Writing kubelet environment file with flags to file "/var/lib/kubelet/kubeadm-flags.env"
[kubelet-start] Starting the kubelet
[kubelet-start] Waiting for the kubelet to perform the TLS Bootstrap...

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 control-plane to see this node join the cluster.
1
2
3
4
5
kubectl get nodes
NAME          STATUS     ROLES    AGE   VERSION
k8s-master1   NotReady   master   12m   v1.19.0
k8s-node1     NotReady   <none>   60s   v1.19.0
k8s-node2     NotReady   <none>   45s   v1.19.0

默认 token 有效期为24小时，当过期之后，该 token 就不可用了。这时就需要重新创建 token，操作如下：

1
2
3
4
5
6
7
8
9
$ kubeadm token create
$ kubeadm token list
$ openssl x509 -pubkey -in /etc/kubernetes/pki/ca.crt | openssl rsa -pubin -outform der 2>/dev/null | openssl dgst -sha256 -hex | sed 's/^.* //'
63bca849e0e01691ae14eab449570284f0c3ddeea590f8da988c07fe2729e924

$ kubeadm join 192.168.100.45:6443 --token nuja6n.o3jrhsffiqs9swnu --discovery-token-ca-cert-hash sha256:63bca849e0e01691ae14eab449570284f0c3ddeea590f8da988c07fe2729e924
# 或者直接命令快捷生成：

$ kubeadm token create --print-join-command

六、部署容器网络（CNI）

pod-network

注意 ：只需要部署下面其中一个，推荐Calico。

  • Calico 是一个纯三层的数据中心网络方案，Calico 支持广泛的平台，包括Kubernetes、OpenStack等。
  • Calico 在每一个计算节点利用 Linux Kernel 实现了一个高效的虚拟路由器（vRouter） 来负责数据转发，而每个 vRouter 通过 BGP 协议负责把自己上运行的 workload 的路由信息向整个 Calico 网络内传播。
  • 此外，Calico 项目还实现了 Kubernetes 网络策略，提供ACL功能。

Quickstart for Calico on Kubernetes

1
$ wget https://docs.projectcalico.org/v3.20/manifests/calico.yaml

下载完后还需要修改里面定义Pod网络（CALICO_IPV4POOL_CIDR），与前面 kubeadm init 指定的一样

修改完后应用清单：

1
2
$ kubectl apply -f calico.yaml
$ kubectl get pods -n kube-system

七、测试kubernetes集群

7.1 kubectl get cs：STATUS Unhealthy 报错

查看 Master 组件状态提示 Unhealthy 报错

1
2
3
4
5
6
$ kubectl get cs
Warning: v1 ComponentStatus is deprecated in v1.19+
NAME                 STATUS      MESSAGE                                                                                       ERROR
scheduler            Unhealthy   Get "http://127.0.0.1:10251/healthz": dial tcp 127.0.0.1:10251: connect: connection refused   
controller-manager   Unhealthy   Get "http://127.0.0.1:10252/healthz": dial tcp 127.0.0.1:10252: connect: connection refused   
etcd-0               Healthy     {"health":"true"}

原因是 controller-manager 和 scheduler 配置文件中 –port=0 参数默认设置为0，导致 apiserver 与组件端口通信获得 Master组件状态，将其注释

1
2
3
4
5
6
7
# 注释配置 # --port=0
$ vim /etc/kubernetes/manifests/kube-controller-manager.yaml 
$ vim /etc/kubernetes/manifests/kube-scheduler.yaml 
# --port=0

# 重启 kubelet
$ systemctl restart kubelet

7.2 使用Nginx镜像测试kubernetes集群

  • 验证Pod工作
  • 验证Pod网络通信
  • 验证DNS解析

在 Kubernetes 集群中创建一个 Nginx Pod，验证是否正常运行：

1
2
3
4
5
6
7
8
9
10
11
12
# 部署
$ kubectl create deployment nginx --image=nginx

# 只创建pod 删除不会重启
$ kubectl create pod nginx --image=nginx

# 暴露端口
$ kubectl expose deployment nginx --port=80 --type=NodePort

# 查看service
$ kubectl get pod,svc
http://NodeIP:Port      # 端口随机生成，通过get svc获取

至此 kubernetes 单 Master 集群部署完成。

  • 报错
    STATUS 状态一直在 ContainerCreating 查看pod 详情 发现报错
    1
    2
    3
    4
    5
    6
    7
    8
    9
    10

    kubectl get pods
    NAME                     READY   STATUS              RESTARTS   AGE
    nginx-6799fc88d8-tvrbw   0/1     ContainerCreating   0          8h

    kubectl describe pod nginx-6799fc88d8-tvrbw
    ·
    ·
    ·
    Failed to create pod sandbox: rpc error: code = Unknown desc = failed to set up sandbox container "717bf2b3d47ca49d78697be84bcfd46c06b8a1a31ba3b780852470c92fa495d4" network for pod "nginx-6799fc88d8-tvrbw": networkPlugin cni failed to set up pod "nginx-6799fc88d8-tvrbw_default" network: stat /var/lib/calico/nodename: no such file or directory: check that the calico/node container is running and has mounted /var/lib/calico/
  • 解决

calico 没装好

查看确认calico 状态，发现状态异常

1
2
3
4
5
6
7
8
kubectl get pods -n kube-system -o wide
NAME                                       READY   STATUS              RESTARTS   AGE   IP               NODE          NOMINATED NODE   READINESS GATES
calico-kube-controllers-577f77cb5c-m6z6z   0/1     ContainerCreating   0          9h    <none>           k8s-node2     <none>           <none>
calico-node-2xwnp                          0/1     ImagePullBackOff    0          9h    192.168.100.45   k8s-master1   <none>           <none>
calico-node-rft5r                          0/1     ImagePullBackOff    0          9h    192.168.100.46   k8s-node1     <none>           <none>
calico-node-vdc4p                          0/1     ImagePullBackOff    0          9h    192.168.100.47   k8s-node2     <none>           <none>
coredns-6d56c8448f-9hhjp                   0/1     ContainerCreating   0          9h    <none>           k8s-node2     <none>           <none>
coredns-6d56c8448f-x8tzg                   0/1     ContainerCreating   0          9h    <none>           k8s-node2     <none>           <none>

查看详情，确认是镜像拉取失败

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
kubectl describe pods calico-node-2xwnp -n kube-system
···
···
···
Image:          docker.io/calico/node:v3.20.6
    Image ID:       
    Port:           <none>
    Host Port:      <none>
    State:          Waiting
      Reason:       ImagePullBackOff
···
···
···
Warning  Failed            9m42s (x85 over 9h)    kubelet, k8s-master1  Error: ErrImagePull
  Warning  DNSConfigForming  5m41s (x1833 over 9h)  kubelet, k8s-master1  Nameserver limits were exceeded, some nameservers have been omitted, the applied nameserver line is: 58.22.96.66 218.104.128.106 223.5.5.5

安装pod网络插件-安装calico插件

1
kubectl apply -f https://docs.projectcalico.org/v3.20/manifests/calico.yaml

在这一步容易出现拉取calico镜像失败的问题，手动拉取：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
docker pull  quay.io/calico/node:v3.20.6

docker pull mirrorgooglecontainers/kube-apiserver:v1.13.2
docker pull mirrorgooglecontainers/kube-proxy:v1.13.2
docker pull mirrorgooglecontainers/kube-controller-manager:v1.13.2
docker pull mirrorgooglecontainers/kube-scheduler:v1.13.2
docker pull mirrorgooglecontainers/coredns:1.2.6
docker pull mirrorgooglecontainers/etcd:3.2.24
docker pull mirrorgooglecontainers/pause:3.1
docker pull mirrorgooglecontainers/kubernetes-dashboard-amd64:v1.10.0

docker pull shikanon096/traefik:1.7.5
docker pull shikanon096/gcr.io.kubernetes-helm.tiller:v2.12.0

docker pull mirrorgooglecontainers/addon-resizer:1.8.4
docker pull mirrorgooglecontainers/metrics-server-amd64:v0.3.1

docker pull quay.io/calico/cni:v3.3.2
docker pull quay.io/calico/node:v3.3.2
docker pull quay.io/calico/typha:v3.3.2

删除旧的pod

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
# 删除旧的
kubectl delete -f calico.yaml
# 修改配置文件
vim calico.yaml
# 修改1  docker.io  修改成 quay.io

# 重新部署
kubectl apply -f calico.yaml
# 全部可以了
kubectl get pods -n kube-system -o wide
NAME                                       READY   STATUS    RESTARTS   AGE     IP               NODE          NOMINATED NODE   READINESS GATES
calico-kube-controllers-749bdb99d7-j2bl7   1/1     Running   2          3m47s   10.245.169.129   k8s-node2     <none>           <none>
calico-node-88jsl                          1/1     Running   0          3m47s   192.168.100.46   k8s-node1     <none>           <none>
calico-node-nbnxw                          1/1     Running   0          3m47s   192.168.100.47   k8s-node2     <none>           <none>
calico-node-wztnw                          1/1     Running   0          3m47s   192.168.100.45   k8s-master1   <none>           <none>
coredns-6d56c8448f-772vz                   1/1     Running   0          9m34s   10.245.169.130   k8s-node2     <none>           <none>
coredns-6d56c8448f-g5c7k                   1/1     Running   0          9m50s   10.245.36.65     k8s-node1     <none>           <none>
etcd-k8s-master1                           1/1     Running   0          11h     192.168.100.45   k8s-master1   <none>           <none>
kube-apiserver-k8s-master1                 1/1     Running   0          11h     192.168.100.45   k8s-master1   <none>           <none>
kube-controller-manager-k8s-master1        1/1     Running   0          11h     192.168.100.45   k8s-master1   <none>           <none>
kube-proxy-7w9tk                           1/1     Running   0          11h     192.168.100.47   k8s-node2     <none>           <none>
kube-proxy-fx99k                           1/1     Running   0          11h     192.168.100.46   k8s-node1     <none>           <none>
kube-proxy-rrqlf                           1/1     Running   0          11h     192.168.100.45   k8s-master1   <none>           <none>
kube-scheduler-k8s-master1                 1/1     Running   0          11h     192.168.100.45   k8s-master1   <none>           <none>

# Nginx 也自动恢复了
kubectl get pods -o wide
NAME                     READY   STATUS    RESTARTS   AGE   IP             NODE        NOMINATED NODE   READINESS GATES
nginx-6799fc88d8-tvrbw   1/1     Running   0          11h   10.245.36.66   k8s-node1   <none>           <none>
# 查看端口
kubectl get pod,svc
NAME                         READY   STATUS    RESTARTS   AGE
pod/nginx-6799fc88d8-tvrbw   1/1     Running   0          11h

NAME                 TYPE        CLUSTER-IP   EXTERNAL-IP   PORT(S)        AGE
service/kubernetes   ClusterIP   10.96.0.1    <none>        443/TCP        11h
service/nginx        NodePort    10.96.48.7   <none>        80:31882/TCP   11h

八、部署 Dashboard

https://gitcode.net/mirrors/kubernetes/dashboard?utm_source=csdn_github_accelerator

1
kubectl apply -f https://raw.githubusercontent.com/kubernetes/dashboard/v2.7.0/aio/deploy/recommended.yaml
1
$ wget https://raw.githubusercontent.com/kubernetes/dashboard/v2.0.3/aio/deploy/recommended.yaml

默认 Dashboard 只能集群内部访问，修改 Service 为 NodePort 类型，暴露到外部：

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
$ vim recommended.yaml
...
kind: Service
apiVersion: v1
metadata:
  labels:
    k8s-app: kubernetes-dashboard
  name: kubernetes-dashboard
  namespace: kubernetes-dashboard
spec:
  ports:
    - port: 443
      targetPort: 8443
      nodePort: 30001
  selector:
    k8s-app: kubernetes-dashboard
  type: NodePort
...
$ kubectl apply -f recommended.yaml
$ kubectl get pods -n kubernetes-dashboard
NAME                                         READY   STATUS    RESTARTS   AGE
dashboard-metrics-scraper-6b4884c9d5-gl8nr   1/1     Running   0          13m
kubernetes-dashboard-7f99b75bf4-89cds        1/1     Running   0          13m

访问地址：https://NodeIP:30001

创建 service account 并绑定默认 cluster-admin 管理员集群角色：

1
2
3
4
5
6
# 创建用户
$ kubectl create serviceaccount dashboard-admin -n kube-system
# 用户授权
$ kubectl create clusterrolebinding dashboard-admin --clusterrole=cluster-admin --serviceaccount=kube-system:dashboard-admin
# 获取用户Token
$ kubectl describe secrets -n kube-system $(kubectl -n kube-system get secret | awk '/dashboard-admin/{print $1}')

使用输出的 token 登录 Dashboard

参考资料1
参考资料2
参考资料3