一、前言

kubeadm是Kubernetes官方推出的快速部署Kubernetes集群工具，kubeadm通过将Kubernetes相关服务容器化以简化部署过程。

在kubernetes 1.8以上的版本中，对于kube-proxy组件增加了除iptables模式和用户模式之外ipvs模式的支持。通过ipvs的NAT模式，对访问k8s service的请求进行虚IP到POD IP的转发。

本文在kubernetes v1.10.0的基础上enable IPVS模式的kube-proxy。

转载自https://blog.csdn.net/cloudvtech

二、准备工作

1.安装好CentOS操作系统

cat /etc/redhat-release 
CentOS Linux release 7.4.1708 (Core) 

2.预先设置必要配置

2.1 停止防火墙：systemctl stop firewalld; systemctl disable firewalld; setenforce 0

2.2 更改/etc/selinux/config： SELINUX=disabled

2.3 配置内核参数：/etc/sysctl.d/k8s.conf

net.bridge.bridge-nf-call-ip6tables = 1
net.bridge.bridge-nf-call-iptables = 1
net.ipv4.ip_forward = 1
vm.swappiness=0

sysctl -p /etc/sysctl.d/k8s.conf

载入ipvs内核模块：

modprobe ip_vs 

modprobe ip_vs_rr 

2.4 配置kubernetes软件源

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

3.安装 kubernetes软件

yum install -y kubelet kubeadm kubectl

3.1 根据需要配置kubelet的启动参数

vi /etc/systemd/system/kubelet.service.d/10-kubeadm.conf

[Service]
Environment="KUBELET_KUBECONFIG_ARGS=--bootstrap-kubeconfig=/etc/kubernetes/bootstrap-kubelet.conf --kubeconfig=/etc/kubernetes/kubelet.conf"
Environment="KUBELET_SYSTEM_PODS_ARGS=--pod-manifest-path=/etc/kubernetes/manifests --allow-privileged=true"
Environment="KUBELET_NETWORK_ARGS=--network-plugin=cni --cni-conf-dir=/etc/cni/net.d --cni-bin-dir=/opt/cni/bin"
Environment="KUBELET_DNS_ARGS=--cluster-dns=10.96.0.10 --cluster-domain=cluster.local"
#Environment="KUBELET_AUTHZ_ARGS=--authorization-mode=Webhook --client-ca-file=/etc/kubernetes/pki/ca.crt"
Environment="KUBELET_CADVISOR_ARGS=--cadvisor-port=0"
Environment="KUBELET_CGROUP_ARGS=--cgroup-driver=systemd"
Environment="KUBELET_CERTIFICATE_ARGS=--rotate-certificates=true --cert-dir=/var/lib/kubelet/pki"
Environment="KUBELET_PROXY_ARGS=--proxy-mode=ipvs --ipvs-min-sync-period=5s --ipvs-sync-period=5s --ipvs-scheduler=rr"

ExecStart=
ExecStart=/usr/bin/kubelet $KUBELET_KUBECONFIG_ARGS $KUBELET_SYSTEM_PODS_ARGS $KUBELET_NETWORK_ARGS $KUBELET_DNS_ARGS $KUBELET_AUTHZ_ARGS $KUBELET_CADVISOR_ARGS $KUBELET_CGROUP_ARGS $KUBELET_CERTIFICATE_ARGS $KUBELET_EXTRA_ARGS

3.2 编辑kubeadm的配置文件

4.关闭这个虚拟机，制作多个clone作为minion节点

转载自https://blog.csdn.net/cloudvtech

三、配置安装kubernetes master和minion节点

1.启动一个虚拟机作为master节点

2.编辑kubeadm配置文件 kubeadm.config

apiVersion: kubeadm.k8s.io/v1alpha1
kind: MasterConfiguration
api:
  advertiseAddress: 192.168.166.101
networking:
  podSubnet: 10.244.0.0/16
kubernetesVersion: "v1.10.0"
kubeProxy:
  config:
    mode: "ipvs"

3.使用kubeadm初始化master节点

kubeadm init   --config ./kubeadm.config [init] Using Kubernetes version: v1.10.0

......

4. 运行上述初始化返回要求运行的命令来配置kubectl

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

5.安装flannel CNI插件

wget https://raw.githubusercontent.com/coreos/flannel/master/Documentation/kube-flannel.yml

kubectl apply -f  kube-flannel.yml

这将在每个node上运行flannel的daemonset

6.在各个节点上运行join命令

kubeadm join 192.168.166.101:6443 --token dtvlg9.798fbe57awyzjzfx --discovery-token-ca-cert-hash sha256:a3406c544acb65954339ab9d6b9801673cb646dd17fa162fbf8aa1cd239b2b99

7.查看安装结果

Note，要reset整个cluster可以在master和每个node上运行如下命令，然后在运行初始化步骤：

kubeadm reset

rm -rf /etc/kubernetes/pki/ca.crt
rm -rf /etc/kubernetes/kubelet.conf 
rm -rf /etc/kubernetes/manifests

要切换kube-proxy的模式，必须修改kubeadm.config，然后使用kubeadm reset和再init整个cluster

转载自https://blog.csdn.net/cloudvtech

四、安装nginx deployment

1.编辑deployment文件

nginx.yml 

apiVersion: apps/v1beta2 # for versions before 1.8.0 use apps/v1beta1
kind: Deployment
metadata:
  name: nginx-deployment
spec:
  selector:
    matchLabels:
      app: nginx
  replicas: 2
  template:
    metadata:
      labels:
        app: nginx
    spec:
      containers:
      - name: nginx
        image: nginx:1.7.9
        ports:
        - containerPort: 80

2.部署

kubectl apply -f nginx.yml 

deployment.apps "nginx-deployment" created

3.暴露nginx service

kubectl expose deployment/nginx-deployment

4.查看ipvs，将数据NAT到后端

而cluster service的VIP会被绑定到每个node的kube-ipvs0虚拟网卡上

node1的NIC

5: kube-ipvs0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN 
    link/ether a2:e0:de:f0:46:e4 brd ff:ff:ff:ff:ff:ff
    inet 10.96.0.1/32 brd 10.96.0.1 scope global kube-ipvs0
       valid_lft forever preferred_lft forever
    inet 10.96.0.10/32 brd 10.96.0.10 scope global kube-ipvs0
       valid_lft forever preferred_lft forever
    inet 10.99.190.144/32 brd 10.99.190.144 scope global kube-ipvs0
       valid_lft forever preferred_lft forever

node2的NIC

5: kube-ipvs0: <BROADCAST,NOARP> mtu 1500 qdisc noop state DOWN 
    link/ether 9e:65:33:4f:c0:7a brd ff:ff:ff:ff:ff:ff
    inet 10.96.0.1/32 brd 10.96.0.1 scope global kube-ipvs0
       valid_lft forever preferred_lft forever
    inet 10.96.0.10/32 brd 10.96.0.10 scope global kube-ipvs0
       valid_lft forever preferred_lft forever
    inet 10.99.190.144/32 brd 10.99.190.144 scope global kube-ipvs0
       valid_lft forever preferred_lft forever

而且这些ipvs使用的NIC的状态是DOWN，只是作为绑定VIP使用

使用ipvs之后，iptables已经没有load balance的信息了

这里需要说明的是由于所有从主机经由IPVS调度进入POD的数据包都会经过MASQUERADE，所以数据包实际上经过了DNAT(by IPVS)和SNAT(by iptables MASQUERADE)，数据包在POD返回的时候一定会返回到原来node所在的IPVS。所以经由flannel的帮助，kubernnetes的IPVS解决方案完全满足IPVS NAT模式需要director和backend在同一个LAN的需求，事实上，这种模式更加类似于IPVS的FULLNAT模式。

通过cluster IP访问nginx service之后的ipvs统计信息

五、配置nodePort

1.node port配置文件nodeport.yaml 

kind: Service
apiVersion: v1
metadata:
  name: nginx-nodeport
spec:
  type: NodePort
  ports:
    - port: 80
      nodePort: 30080
  selector:
      app: nginx

2.apply

kubectl apply -f nodeport.yaml 

3.在k8s-node1上查看ipvs

node

六、相关代码

https://github.com/kubernetes/kubernetes/blob/master/pkg/proxy/ipvs/proxier.go

......