通过 Kubeadm联网部署K8S

最新推荐文章于 2024-08-05 01:14:29 发布

汀桦坞

最新推荐文章于 2024-08-05 01:14:29 发布

阅读量205

点赞数

分类专栏：云计算文章标签： docker linux 运维

本文链接：https://blog.csdn.net/wiborgite/article/details/120046733

版权

云计算专栏收录该内容

31 篇文章 1 订阅

订阅专栏

环境信息

角色	IP	系统
master	10.4.7.152	Ubuntu18.04
node	10.4.7.162	Ubuntu18.04

1. 操作系统配置（两个节点均需执行）

操作节点：master，node

禁用swap

swapoff -a  #临时 
sed -ri 's/.*swap.*/#&/' /etc/fstab   #将/etc/fstab文件的swap行用#注释（永久禁用）

关闭防火墙

sudo ufw disable  #关闭防火墙
sudo ufw status  #查看防火墙状态 Status: inactive

关闭SELinux (如果没有安装SELinux，跳过)

getenforce #查看selinux状态
sudo apt install selinux-utils
sudo setenforce 0 #临时关闭
sed -i 's/enforcing/disabled/' /etc/selinux/config  #永久关闭 修改/etc/selinux/config文件中设置SELINUX=disabled，然后重启服务器。

修改内核参数

cat <<EOF > /etc/sysctl.d/k8s.conf
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
net.ipv4.ip_forward = 1
EOF
# 执行以下命令使之生效
sysctl --system

加载内核模块

modprobe br_netfilter
lsmod | grep br_netfilter

2. 安装Docker（两个节点均需执行）

操作节点：master，node

为了下一步安装KubeEdge，Docker必须安装19.3以上的版本。

# Docker旧版本卸载
sudo apt-get remove docker docker-engine docker-ce docker.io

# 安装Docker
curl -sSL https://get.daocloud.io/docker | sh

# 在 /etc/docker/daemon.json 文件中添加如下信息：
 { "registry-mirrors": ["https://hub-mirror.c.163.com"] } 
使上述配置生效：

sudo systemctl daemon-reload
sudo systemctl restart docker

#安装完成 检验是否安装成功
sudo docker version

3. 安装kubectl，kubelet，kubeadm（两个节点均需执行）

操作节点：master，node

#打开apt源文件：
sudo vim /etc/apt/sources.list
#添加如下：
deb https://mirrors.aliyun.com/kubernetes/apt kubernetes-xenial main
#添加public key
curl https://mirrors.aliyun.com/kubernetes/apt/doc/apt-key.gpg | sudo apt-key add - 
#然后更新apt源：
sudo apt-get update
#安装kube(可以apt-cache madison 查看可安装的版本列表 )
sudo apt-get install -y kubelet=1.19.8-00 kubeadm=1.19.8-00 kubectl=1.19.8-00
#安装之后查看版本
kubelet --version

4. 在master节点通过kubeadm部署Kubernetes

操作节点：master

查询需要的镜像

root@master-152:~# kubeadm config images list --kubernetes-version v1.19.8
k8s.gcr.io/kube-apiserver:v1.19.8
k8s.gcr.io/kube-controller-manager:v1.19.8
k8s.gcr.io/kube-scheduler:v1.19.8
k8s.gcr.io/kube-proxy:v1.19.8
k8s.gcr.io/pause:3.2
k8s.gcr.io/etcd:3.4.13-0
k8s.gcr.io/coredns:1.7.0

Kubernetes安装

kubeadm init \
--apiserver-advertise-address=10.4.7.152 \
--image-repository registry.aliyuncs.com/google_containers \
--kubernetes-version v1.19.8 \
--service-cidr=10.96.0.0/12 \
--pod-network-cidr=10.244.0.0/16

kubeadm init选项说明:
–apiserver-advertise-address #APIserver所公布的其正在监听的 IP 地址。如果未设置，则使用默认网络接口。
–image-repository #选择拉取 control plane images 的镜像repo (default “k8s.gcr.io”)
–kubernetes-version #选择kubernetes版本.(default “stable-1”)
–service-cidr #指定service的IP 范围.(default “10.96.0.0/12”)
–pod-network-cidr #指定pod的网络，control plane会自动将网络发布到其他节点的node，让其上启动的容器使用此网络。

安装成功输出如下：

Your Kubernetes control-plane 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) $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 10.4.7.152:6443 --token 7iddpr.00c75zjjoh78gpbi \
    --discovery-token-ca-cert-hash sha256:c5dad4cf76016b5e82e95ba4e69f53559835759b684672de7ea32b8548ef1184

按照上面提示，复制config文件

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

此时可以看到主节点状态，状态是NotReady

root@master-152:~# kubectl get nodes
NAME         STATUS      ROLES    AGE     VERSION
master-152   NotReady    master   6m36s   v1.19.8

flannel部署

#下载flannel部署文件到本地
wget https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml
#flannel部署
kubectl apply -f kube-flannel.yml
#flannel部署成功后，查看节点状态
root@master-152:~# kubectl get nodes
NAME         STATUS   ROLES    AGE     VERSION
master-152   Ready    master   6m36s   v1.19.8

5. 将Node节点添加到K8s集群中

操作节点：node

在node节点运行kubeadm join添加该节点到集群，运行命令后需要等待几分钟，后台在下载镜像。

kubeadm join 10.4.7.152:6443 --token mfymh2.rxvxyp2579coacbv \
--discovery-token-ca-cert-hash sha256:9c5b395069f1b327bf4da4f91345674e1f316adf1a6af98a4ea7c466ebacaf68

如果token过期或者忘记，在master节点重新获取。

kubeadm token create --print-join-command
kubeadm join 10.4.7.152:6443 --token k66jak.2c6yw8p50jj6g1e9     --discovery-token-ca-cert-hash sha256:9c5b395069f1b327bf4da4f91345674e1f316adf1a6af98a4ea7c466ebacaf68

查看节点状态

kubectl get nodes
NAME         STATUS   ROLES    AGE     VERSION
master-152   Ready    master   6m36s   v1.19.8
node-162     Ready    <none>   2m33s   v1.19.8

二. kubeadm离线安装K8s

角色	IP	系统
master	10.4.7.152	Ubuntu18.04

1. 系统初始化

禁用swap

swapoff -a  #临时 
sed -ri 's/.*swap.*/#&/' /etc/fstab   #将/etc/fstab文件的swap行用#注释（永久禁用）

关闭防火墙

sudo ufw disable  #关闭防火墙
sudo ufw status  #查看防火墙状态 Status: inactive

关闭SELinux (如果没有安装SELinux，跳过)

getenforce #查看selinux状态
sudo apt install selinux-utils
sudo setenforce 0 #临时关闭
sed -i 's/enforcing/disabled/' /etc/selinux/config  #永久关闭 修改/etc/selinux/config文件中设置SELINUX=disabled，然后重启服务器。

修改内核参数

cat <<EOF > /etc/sysctl.d/k8s.conf
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
net.ipv4.ip_forward = 1
EOF
# 执行以下命令使之生效
sysctl --system

加载内核模块

modprobe br_netfilter
lsmod | grep br_netfilter

2. Docker部署

为了下一步安装KubeEdge，Docker必须安装19.3以上的版本。

离线安装包下载
网址：https://download.docker.com/linux/ubuntu/dists/xenial/pool/stable/amd64/
下载3个安装包：containerd.io ，docker-ce-cli，docker-ce
选择19.3以上的版本，比如：
docker-ce_20.10.0_3-0ubuntu-xenial_amd64.deb
docker-ce-cli_20.10.0_3-0ubuntu-xenial_amd64.deb
containerd.io_1.4.3-1_amd64.deb
Docker离线安装包安装

dpkg -i containerd.io_1.4.3-1_amd64.deb
dpkg -i docker-ce-cli_20.10.0~3-0~ubuntu-xenial_amd64.deb
dpkg -i docker-ce_20.10.0~3-0~ubuntu-xenial_amd64.deb

检验是否安装成功

docker version

3. kubectl，kubelet，kubeadm部署

安装包下载

以下操作在可联网的Ubuntu机器。

打开apt源文件：

vim /etc/apt/sources.list

将下行内容添加到文件末尾
deb https://mirrors.aliyun.com/kubernetes/apt kubernetes-xenial main

#添加public key
curl https://mirrors.aliyun.com/kubernetes/apt/doc/apt-key.gpg | sudo apt-key add - 
#然后更新apt源：
sudo apt-get update

#下载安装包到本地
sudo apt-get download -y kubelet=1.19.8-00 kubeadm=1.19.8-00 kubectl=1.19.8-00
#下载依赖项安装包到本地
sudo apt-get download cri-tools=1.13.0-01 socat=1.7.3.2-2ubuntu2 conntrack kubernetes-cni=0.8.7-00

#查看下载的安装包
ll
total 66904
drwxr-xr-x 2 root     root         4096 Aug  5 10:51 ./
drwxr-xr-x 6 lixingli lixingli     4096 Aug  5 07:13 ../
-rw-r--r-- 1 root     root        30580 Apr 16  2018 conntrack_1%3a1.4.4+snapshot20161117-6ubuntu2_amd64.deb
-rw-r--r-- 1 root     root      8775008 Jan  2  2021 cri-tools_1.13.0-01_amd64.deb
-rw-r--r-- 1 root     root      7763176 Feb 18 20:03 kubeadm_1.19.8-00_amd64.deb
-rw-r--r-- 1 root     root      8352916 Feb 18 20:03 kubectl_1.19.8-00_amd64.deb
-rw-r--r-- 1 root     root     18226180 Feb 18 20:03 kubelet_1.19.8-00_amd64.deb
-rw-r--r-- 1 root     root     24995436 Jan  2  2021 kubernetes-cni_0.8.7-00_amd64.deb
-rw-r--r-- 1 root     root       341772 Apr  4  2018 socat_1.7.3.2-2ubuntu2_amd64.deb

安装包安装

复制安装包到目标机器

（注意安装顺序）


dpkg -i kubernetes-cni_0.8.7-00_amd64.deb
dpkg -i conntrack_1%3a1.4.4+snapshot20161117-6ubuntu2_amd64.deb
dpkg -i socat_1.7.3.2-2ubuntu2_amd64.deb
dpkg -i cri-tools_1.13.0-01_amd64.deb
dpkg -i kubectl_1.19.8-00_amd64.deb
dpkg -i kubelet_1.19.8-00_amd64.deb
dpkg -i kubeadm_1.19.8-00_amd64.deb

#检查是否安装成功
keadm version
kubectl version
kubelet --version

4. Master节点启动

镜像列表查询

root@master-152:~# kubeadm config images list --kubernetes-version v1.19.8 --image-repository registry.aliyuncs.com/google_containers
registry.aliyuncs.com/google_containers/kube-apiserver:v1.19.8
registry.aliyuncs.com/google_containers/kube-controller-manager:v1.19.8
registry.aliyuncs.com/google_containers/kube-scheduler:v1.19.8
registry.aliyuncs.com/google_containers/kube-proxy:v1.19.8
registry.aliyuncs.com/google_containers/pause:3.2
registry.aliyuncs.com/google_containers/etcd:3.4.13-0
registry.aliyuncs.com/google_containers/coredns:1.7.0

–kubernetes-version #kubernetes版本号
–image-repository #选择拉取的镜像repo (default “k8s.gcr.io”)

镜像打包

在已经安装K8s的master机器上，执行以下操作，打包镜像。打包完毕后，将镜像copy到目标机器。

docker save -o kube-apiserver-1-19-8.tar registry.aliyuncs.com/google_containers/kube-apiserver:v1.19.8
docker save -o kube-controller-manager-1-19-8.tar registry.aliyuncs.com/google_containers/kube-controller-manager:v1.19.8
docker save -o kube-scheduler-1-19-8.tar registry.aliyuncs.com/google_containers/kube-scheduler:v1.19.8
docker save -o kube-proxy-1-19-8.tar registry.aliyuncs.com/google_containers/kube-proxy:v1.19.8
docker save -o pause-3-2.tar registry.aliyuncs.com/google_containers/pause:3.2
docker save -o etcd-3-4-13-0.tar registry.aliyuncs.com/google_containers/etcd:3.4.13-0
docker save -o coredns-1-7-0.tar registry.aliyuncs.com/google_containers/coredns:1.7.0

#查看已打包的镜像
ll
drwxr-xr-x 2 root     root          4096 Aug  6 02:59 ./
drwxr-xr-x 7 lixingli lixingli      4096 Aug  6 02:58 ../
-rw-rw-r-- 1 lixingli lixingli  45365760 Jul 26 08:29 coredns-1-7-0.tar
-rw-rw-r-- 1 lixingli lixingli 254679040 Jul 26 08:27 etcd-3-4-13-0.tar
-rw-rw-r-- 1 lixingli lixingli 120077824 Jul 26 07:35 kube-apiserver-1-19-8.tar
-rw-rw-r-- 1 lixingli lixingli 112070144 Jul 26 07:46 kube-controller-manager-1-19-8.tar
-rw-rw-r-- 1 lixingli lixingli 119683072 Jul 26 07:36 kube-proxy-1-19-8.tar
-rw-rw-r-- 1 lixingli lixingli  47775232 Jul 26 07:45 kube-scheduler-1-19-8.tar
-rw-rw-r-- 1 lixingli lixingli    692736 Jul 26 08:28 pause-3-2.tar

镜像加载

docker load -i kube-apiserver-1-19-8.tar
docker load -i coredns-1-7-0.tar
docker load -i etcd-3-4-13-0.tar
docker load -i kube-controller-manager-1-19-8.tar
docker load -i kube-proxy-1-19-8.tar
docker load -i kube-scheduler-1-19-8.tar
docker load -i pause-3-2.tar

#查看加载的镜像
docker images
REPOSITORY                                                        TAG        IMAGE ID       CREATED         SIZE
registry.aliyuncs.com/google_containers/kube-proxy                v1.19.8    ea03182b84a2   5 months ago    118MB
registry.aliyuncs.com/google_containers/kube-apiserver            v1.19.8    9ba91a90b7d1   5 months ago    119MB
registry.aliyuncs.com/google_containers/kube-controller-manager   v1.19.8    213ae7795128   5 months ago    111MB
registry.aliyuncs.com/google_containers/kube-scheduler            v1.19.8    919a3f36437d   5 months ago    46.5MB
registry.aliyuncs.com/google_containers/etcd                      3.4.13-0   0369cf4303ff   11 months ago   253MB
registry.aliyuncs.com/google_containers/coredns                   1.7.0      bfe3a36ebd25   13 months ago   45.2MB
registry.aliyuncs.com/google_containers/pause                     3.2        80d28bedfe5d   17 months ago   683kB

k8s安装

kubeadm init \
--apiserver-advertise-address=10.4.7.153 \
--image-repository registry.aliyuncs.com/google_containers \
--kubernetes-version v1.19.8 \
--service-cidr=10.96.0.0/12 \
--pod-network-cidr=10.244.0.0/16

安装成功，屏幕输出如下：

Your Kubernetes control-plane 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) $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 10.4.7.153:6443 --token yirah5.tcwkxhm9ui1kd8bg \
    --discovery-token-ca-cert-hash sha256:4e99dac259dd662932056b26fe8ff85208ad22ef8fe9700b86e7418af6bf92f6

按照上面提示，复制config文件

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

此时可以看到主节点状态，状态是NotReady

root@master-152:~# kubectl get nodes 
NAME         STATUS      ROLES    AGE   VERSION
master-152   NotReady    master   20m   v1.19.8

5. flannel部署

安装包下载

在已经安装K8s的master机器上，执行以下操作，打包镜像，然后复制到目标机器。

#下载flannel部署文件到本地
wget https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml

#打包flannel镜像
docker save -o flannel-0-14-0.tar  quay.io/coreos/flannel:v0.14.

#查看下载项目
ll
total 67324
drwxrwxr-x 2 lixingli lixingli     4096 Aug  6 06:43 ./
drwxr-xr-x 7 lixingli lixingli     4096 Aug  6 02:58 ../
-rw------- 1 root     root     68921344 Aug  6 06:42 flannel-0-14-0.tar
-rw-rw-r-- 1 lixingli lixingli     4813 Jul 25 14:30 kube-flannel.yml

flannel部署

#加载镜像
docker load -i flannel-v-0-14-0.tar
#查看镜像
docker images | grep flannel
quay.io/coreos/flannel                     v0.14.0    8522d622299c   2 months ago    67.9MB
#flannel部署
kubectl apply -f kube-flannel.yml

#查看节点状态，状态变为Ready
root@master-152:~# kubectl get nodes 
NAME         STATUS   ROLES    AGE   VERSION
master-152   Ready    master   20m   v1.19.8
#查看pods状态，所有pods处于Running状态
root@master-152:~# kubectl get pods -A
NAMESPACE     NAME                               READY   STATUS    RESTARTS   AGE
kube-system   coredns-6d56c8448f-shkd8           1/1     Running   0          22m
kube-system   coredns-6d56c8448f-t6gnk           1/1     Running   0          22m
kube-system   etcd-lixingli                      1/1     Running   0          22m
kube-system   kube-apiserver-lixingli            1/1     Running   0          22m
kube-system   kube-controller-manager-lixingli   1/1     Running   0          22m
kube-system   kube-flannel-ds-cmpqv              1/1     Running   0          2m24s
kube-system   kube-proxy-pd2bt                   1/1     Running   0          22m
kube-system   kube-scheduler-lixingli            1/1     Running   0          22m