kubernetes启用ipvs模式 （以kubeadm安装的集群为例）
启用ipvs而不使用iptables的原因
- ipvs 可以更快地重定向流量，并且在同步代理规则时具有更好的性能。此外，ipvs 为负载均衡算法提供了更
- 多选项，例如：
- rr ：轮询调度
- lc ：最小连接数
- dh ：目标哈希
- sh ：源哈希
- sed ：最短期望延迟
- nq ： 不排队调度
每个节点都进行如下操作：
安装ipvs相关软件包
yum -y install ipvsadm ipset
modprobe ip_vs
lsmod | grep ip_vs
修改内核参数
vi /etc/sysctl.conf 添加
net.ipv4.ip_forward = 1
net.bridge.bridge-nf-call-iptables = 1
net.bridge.bridge-nf-call-ip6tables = 1
# 应用生效
sysctl -p
修改kube-proxy模式
kubectl edit configmap -n kube-system kube-proxy
mode: "ipvs"
nodePortAddresses: null
如果要修改调度算法使用
scheduler: ""
默认为空
[root@k8s-master ~]# ipvsadm -L -n
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
依次重启/删除kube-proxy的pod
Kubeadm不仅仅将k8s集群搭建简化了，还采用了容器化的方式去部署k8s组件，这里只需要知道采用了容器化部署，因为帮我们封装好了。唯一一个采用的不是容器化部署的是kubelete，这个使用的是传统的systemd去管理的，在宿主机进行管理。其他的都是采用容器化部署，也就是通过启动容器帮你拉起组件。
[root@k8s-master ~]# kubectl get pod -n kube-system -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES 1/1 Running 6 2d12h 192.168.179.102 k8s-master <none> <none>
kube-proxy-7wgls 1/1 Running 2 2d11h 192.168.179.103 k8s-node1 <none> <none>
kube-proxy-vkt7g 1/1 Running 3 2d11h 192.168.179.104 k8s-node2 <none> <none>
kube-proxy-xth6p 1/1 Running 2 2d12h 192.168.179.102 k8s-master <none> <none>
删除pod之后 会新起一个新的pod
[root@k8s-master ~]# kubectl delete pod kube-proxy-7wgls -n kube-system
pod "kube-proxy-7wgls" deleted
[root@k8s-master ~]# kubectl delete pod kube-proxy-vkt7g -n kube-system
pod "kube-proxy-vkt7g" deleted
[root@k8s-master ~]# kubectl delete pod kube-proxy-xth6p -n kube-system
pod "kube-proxy-xth6p" deleted
等容器重启完毕再去看看
[root@k8s-master ~]# ipvsadm -L -n
IP Virtual Server version 1.2.1 (size=4096)
Prot LocalAddress:Port Scheduler Flags
-> RemoteAddress:Port Forward Weight ActiveConn InActConn
TCP 127.0.0.1:32567 rr
-> 10.244.169.137:80 Masq 1 0 0
TCP 172.17.0.1:31332 rr
-> 10.244.169.139:80 Masq 1 0 0
TCP 172.17.0.1:31383 rr
-> 10.244.169.138:8443 Masq 1 0 0
TCP 172.17.0.1:32567 rr
-> 10.244.169.137:80 Masq 1 0 0
TCP 192.168.179.102:31332 rr
-> 10.244.169.139:80 Masq 1 0 0
TCP 192.168.179.102:31383 rr
-> 10.244.169.138:8443 Masq 1 0 0
TCP 192.168.179.102:32567 rr
-> 10.244.169.137:80 Masq 1 0 0
TCP 10.96.0.1:443 rr
-> 192.168.179.102:6443 Masq 1 0 0
TCP 10.96.0.10:53 rr
-> 10.244.36.73:53 Masq 1 0 0
-> 10.244.36.75:53 Masq 1 0 0
TCP 10.96.0.10:9153 rr
-> 10.244.36.73:9153 Masq 1 0 0
-> 10.244.36.75:9153 Masq 1 0 0
TCP 10.98.137.105:80 rr
-> 10.244.169.137:80 Masq 1 0 0
TCP 10.99.50.2:80 rr
-> 10.244.169.139:80 Masq 1 0 0
TCP 10.99.68.0:443 rr
-> 192.168.179.103:4443 Masq 1 0 0
TCP 10.106.187.194:443 rr
-> 10.244.169.138:8443 Masq 1 0 0
TCP 10.108.138.194:8000 rr
-> 10.244.36.74:8000 Masq 1 0 0
TCP 127.0.0.1:31332 rr
-> 10.244.169.139:80 Masq 1 0 0
TCP 127.0.0.1:31383 rr
-> 10.244.169.138:8443 Masq 1 0 0
UDP 10.96.0.10:53 rr
-> 10.244.36.73:53 Masq 1 0 0
-> 10.244.36.75:53 Masq 1 0 0
TCP 172.17.0.1:31332 rr
-> 10.244.169.139:80 Masq 1 0 0
[root@k8s-master ~]# kubectl get svc
NAME TYPE CLUSTER-IP EXTERNAL-IP PORT(S) AGE
kubernetes ClusterIP 10.96.0.1 <none> 443/TCP 2d12h
nginx NodePort 10.99.50.2 <none> 80:31332/TCP 2d11h
[root@k8s-master ~]# kubectl get pod -o wide
NAME READY STATUS RESTARTS AGE IP NODE NOMINATED NODE READINESS GATES
nginx-6799fc88d8-mkq8g 1/1 Running 0 10m 10.244.169.139 k8s-node2 <none> <none>
