跳至主要內容

K8S与服务网格化

bincooo大约 3 分钟

基本信息

虚拟机： VMware Workstation Pro 15
系统： CentOS Linux release 7.9.2009 (Core) 64bit
  • k8s1(master) 192.168.170.131
  • k8s2(worker) 192.168.170.132
  • k8s3(worker) 192.168.170.133
docker： Docker Engine - Community 20.10.5
k8s：
  • kubelet 1.15.0
  • kubeadm 1.15.0
  • kubectl 1.15.0

准备工作

关闭交换空间

swapoff -a

避免开机启动交换空间

#注释swap行
vim /etc/fstab

修改主机名

hostnamectl set-hostname k8s1
vim /etc/hosts

192.168.170.130 k8s1

设置时区

timedatectl status
timedatectl set-timezone "Asia/Shanghai"

关闭防火墙

systemctl status firewalld
firewall-cmd --state
# 开启
# service firewalld start
# 重启
# serivce firewalld restart
# 停止
service firewalld stop

安装k8s组件

可参考官网提供的安装步骤 点我open in new window

1. 配置yum源

cat <<EOF > /etc/yum.repos.d/kubernetes.repo
[kubernetes]
name=Kubernetes
baseurl=https://mirrors.aliyun.com/kubernetes/yum/repos/kubernetes-el7-x86_64
enabled=1
gpgcheck=1
repo_gpgcheck=1
gpgkey=https://mirrors.aliyun.com/kubernetes/yum/doc/yum-key.gpg https://mirrors.aliyun.com/kubernetes/yum/doc/rpm-package-key.gpg
EOF

2. 设置selinex

setenforce 0

3. 安装组件

yum install -y kubelet-1.15.0 kubeadm-1.15.0 kubectl-1.15.0
# 启动kublet,并开机自启动
systemctl start kubelet
systemctl enable kubelet

kubeadm是k8s集群安装工具，能够快速搭建k8s集群。

  • 离线包上传suveng-k8s-image.tar.gz 下载open in new window 提取码: m6a1；版本为1.15.0

    可参考博客 点我open in new window

    解压
    tar -zxvf suveng-k8s-image.tar.gz
    导入镜像
    docker load -i suveng/k8s.gcr.io-kube-proxy.tar
docker load -i suveng/k8s.gcr.io-kube-apiserver.tar
docker load -i suveng/k8s.gcr.io-kube-controller-manager.tar
docker load -i suveng/k8s.gcr.io-kube-scheduler.tar
docker load -i suveng/k8s.gcr.io-coredns.tar
docker load -i suveng/k8s.gcr.io-etcd.tar
docker load -i suveng/k8s.gcr.io-pause.tar
docker load -i suveng/flannel.tar
    重新打标签
    docker tag suveng/k8s.gcr.io-kube-apiserver:v1.15.0 k8s.gcr.io/kube-apiserver:v1.15.0
docker tag suveng/k8s.gcr.io-kube-scheduler:v1.15.0 k8s.gcr.io/kube-scheduler:v1.15.0
docker tag suveng/k8s.gcr.io-kube-controller-manager:v1.15.0 k8s.gcr.io/kube-controller-manager:v1.15.0
docker tag suveng/k8s.gcr.io-kube-proxy:v1.15.0 k8s.gcr.io/kube-proxy:v1.15.0
docker tag suveng/k8s.gcr.io-etcd:3.3.10 k8s.gcr.io/etcd:3.3.10
docker tag suveng/k8s.gcr.io-pause:3.1 k8s.gcr.io/pause:3.1
docker tag suveng/k8s.gcr.io-coredns:1.3.1 k8s.gcr.io/coredns:1.3.1
    删除自己的标签
    docker rmi suveng/k8s.gcr.io-kube-apiserver:v1.15.0 
docker rmi suveng/k8s.gcr.io-kube-scheduler:v1.15.0 
docker rmi suveng/k8s.gcr.io-kube-controller-manager:v1.15.0 
docker rmi suveng/k8s.gcr.io-kube-proxy:v1.15.0 
docker rmi suveng/k8s.gcr.io-etcd:3.3.10 
docker rmi suveng/k8s.gcr.io-pause:3.1 
docker rmi suveng/k8s.gcr.io-coredns:1.3.1

  • centos7用户还需要设置路由

    yum install -y bridge-utils.x86_64

# 加载br_netfilter模块，使用lsmod查看开启的模块
modprobe  br_netfilter
cat <<EOF >  /etc/sysctl.d/k8s.conf
net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
EOF

# 重新加载所有配置文件
sysctl --system

    后面添加k8s 从节点时会报出一个warn错误：The recommended driver is "systemd".

    可通过配置去除这个警告

    vi /etc/docker/daemon.json

    "exec-opts": ["native.cgroupdriver=systemd"]

    sudo systemctl daemon-reload
sudo systemctl restart docker

​

  • 安装主节点(master)

    # master节点初始化,配置网络,应用flannel网络,注意这里的flannel网络的pod内网地址默认是10.244.0.0/16,在master初始化时指定内网地址
kubeadm init --apiserver-advertise-address <master_ip> \ 
  --pod-network-cidr 10.244.0.0/16 --kubernetes-version 1.15.0 \
  | tee kubeadm-init.log
kubectl get node

    根据日志信息执行脚本： kubeadm-init.log里有脚本提示

    k8s1(192.168.170.131)主机

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

  • 安装从节点(worker)： kubeadm-init.log里有脚本提示 (先安装CNI通讯组件)

    k8s2(192.168.170.132)主机、k8s3(192.168.170.133)主机

    kubeadm join 192.168.170.131:6443 --token abcdef.0123456789abcdef \
  --discovery-token-ca-cert-hash \
  sha256:f7ef568da1265047be6170fdbae7ccbea79f1f8b4762d44f4d6eab425319d4e5

# 主节点查看 (一秒刷新)
watch -n 1 kubectl get pod -A

    使从节点可访问整个集群 api-resources (可选：一般只会在主节点操作pod)

    mkdir $HOME/.kube
# 从master拷贝文件到worker
scp root@192.168.170.131:/etc/kubernetes/admin.conf $HOME/.kube/config
# 授权
sudo chown $(id -u):$(id -g) $HOME/.kube/config

4. CNI 通讯

CNI(Container Network Interface)是容器平台(docker、k8s、mesos ...) 的一套容器网络标准；

它的解决方案有flannel、calico、weave这些都是CNI的实现；

这里提供flannel和calico的安装步骤：(二选一即可)

flannel安装
# 下载配置
curl -o kube-flannel.yml \
  https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml
# 配置网络
kubectl apply -f kube-flannel.yml

# 检测master节点
kubectl get node
watch kubectl get pods --all-namespaces

# 查看kubelet的日志
journalctl -fu kubelet

解决k8s的coredns一直处于的crashloopbackoff问题

# 查看日志
kubectl logs -n kube-system coredns-5c98db65d4-bxmhq
# error: no such file or directory

参考博客 点我open in new window

解决方式(仅供参考)

可能是由iptables规则的错乱或者缓存导致的

systemctl stop kubelet
systemctl stop docker
iptables --flush
iptables -tnat --flush
systemctl start kubelet
systemctl start docker
calico安装 点我open in new window
kubectl create -f https://docs.projectcalico.org/archive/v3.18/manifests/tigera-operator.yaml

kubectl get node 获取节点时

NAME   STATUS     ROLES    AGE    VERSION
k8s1   Ready      master   7d1h   v1.15.0
k8s2   NotReady   <none>   7d     v1.15.0
k8s3   NotReady   <none>   7d     v1.15.0

可以尝试刷新iptables试试

systemctl stop kubelet
iptables --flush
iptables -tnat --flush
systemctl start kubelet
Copyright © 2025 bincooo