请确认你已经完成了准备工作中的内容，这里我们来看一下如何一步步搭建一个非高可用的 Kube-OVN。

部署相关 CRD

Kube-OVN 创建了 Subnet 和 IP 两种 CRD 资源方便网络的管理

  1. kubectl apply -f https://raw.githubusercontent.com/alauda/kube-ovn/release-1.2/yamls/crd.yaml

查看 CRD

  1. [root@node2 yamls]# kubectl get crd
  2. NAME                 CREATED AT
  3. ips.kubeovn.io       2019-08-06T09:08:44Z
  4. subnets.kubeovn.io   2019-08-06T09:08:44Z

部署 OVN

Kube-OVN 底层依赖 Open vSwitch 社区提供的 OVS 和 OVN，这里我们通过 yaml 来进行部署。

给部署 ovndb 的机器打标签

ovndb 需要将数据存在宿主机硬盘来持久化状态，选择一个节点增加标签

  1. kubectl label node <Node on which to deploy OVN DB> kube-ovn/role=master

查看标签状态

  1. [root@node2 yamls]# kubectl get no --show-labels
  2. NAME    STATUS   ROLES    AGE     VERSION   LABELS
  3. node1   Ready    master   7d21h   v1.15.1   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kube-ovn/role=master,kubernetes.io/arch=amd64,kubernetes.io/hostname=node1,kubernetes.io/os=linux,node-role.kubernetes.io/master=
  4. node2   Ready    master   7d21h   v1.15.1   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=node2,kubernetes.io/os=linux,node-role.kubernetes.io/master=
  5. node3   Ready    <none>   7d21h   v1.15.1   beta.kubernetes.io/arch=amd64,beta.kubernetes.io/os=linux,kubernetes.io/arch=amd64,kubernetes.io/hostname=node3,kubernetes.io/os=linux

部署 OVN/OVS

部署 yaml

  1. kubectl apply -f https://raw.githubusercontent.com/alauda/kube-ovn/release-1.2/yamls/ovn.yaml

查看 pod 状态

  1. [root@node2 yamls]# kubectl get pod -n kube-system
  2. NAME                           READY   STATUS    RESTARTS   AGE
  3. ovn-central-6d5d566cc8-vktvm   1/1     Running   0          32s
  4. ovs-ovn-47wxk                  1/1     Running   0          32s
  5. ovs-ovn-l8t6g                  1/1     Running   0          32s
  6. ovs-ovn-v225f                  1/1     Running   0          32s

安装 Kube-OVN Controller 及 CNIServer

Kube-OVN Controller 和 CNIServer 中有大量可配参数，这里为了快速上手，我们不做更改。默认配置下 Kube-OVN 会使用 10.16.0.0/16 作为默认子网，100.64.0.1/16 作为主机和 Pod 通信子网，使用 Kubernetes 中的 Node 主网卡作为 Pod 流量通信使用网卡，并开启流量镜像功能。

部署 yaml

  1. kubectl apply -f https://raw.githubusercontent.com/alauda/kube-ovn/release-1.2/yamls/kube-ovn.yaml

观察部署状态：

  1. [root@node2 yamls]# kubectl get pod -n kube-system
  2. NAME                                   READY   STATUS    RESTARTS   AGE
  3. kube-ovn-cni-29784                     1/1     Running   0          17s
  4. kube-ovn-cni-5572f                     1/1     Running   0          17s
  5. kube-ovn-cni-hdffl                     1/1     Running   0          17s
  6. kube-ovn-controller-65b5cb8574-262vg   1/1     Running   0          17s
  7. kube-ovn-controller-65b5cb8574-ss94l   1/1     Running   0          17s
  8. ovn-central-6d5d566cc8-vktvm           1/1     Running   0          10m
  9. ovs-ovn-47wxk                          1/1     Running   0          10m
  10. ovs-ovn-l8t6g                          1/1     Running   0          10m
  11. ovs-ovn-v225f                          1/1     Running   0          10m

观察自动创建的 Subnet:

  1. [root@node2 yamls]# kubectl get subnet
  2. NAME          PROTOCOL   CIDR            PRIVATE   NAT
  3. join          IPv4       100.64.0.0/16   false     false
  4. ovn-default   IPv4       10.16.0.0/16    false     true

安装 Kubectl 插件

为了更好的对网络质量进行监控以及对故障进行排查，Kube-OVN 提供了 kubectl 的插件，推荐安装。

  1. 下载 kubectl-ko 文件
  1. wget https://raw.githubusercontent.com/alauda/kube-ovn/release-1.2/dist/images/kubectl-ko
  1. 将文件复制到 $PATH 下的某个目录
  1. mv kubectl-ko /usr/local/bin/kubectl-ko
  1. 给 kubectl-ko增加可执行权限
  1. chmod +x /usr/local/bin/kubectl-ko
  1. 检查插件状态
  1. [root@kube-ovn01 ~]# kubectl plugin list
  2. The following compatible plugins are available:
  3.  
  4. /usr/local/bin/kubectl-ko
  1. 对网络质量进行检查
  1. [root@kube-ovn01 ~]# kubectl ko diagnose all

好了！你已经完成了 Kube-OVN 的安装，让我们运行一个 Pod 来观察一下网络。

  1. kubectl run nginx --image=nginx

观察容器地址

  1. [root@node2 yamls]# kubectl get pod -o wide
  2. NAME                     READY   STATUS    RESTARTS   AGE     IP          NODE    NOMINATED NODE   READINESS GATES
  3. nginx-74d5899f46-x5ghb   1/1     Running   0          2m58s   10.16.0.2   node1   <none>           <none>

进入容器进行一些简单网络操作

  1. [root@node2 yamls]# kubectl exec -it nginx-74d5899f46-x5ghb sh
  2. / # ping -c 1 10.16.0.1
  3. PING 10.16.0.1 (10.16.0.1): 56 data bytes
  4. 64 bytes from 10.16.0.1: seq=0 ttl=254 time=0.379 ms
  5.  
  6. --- 10.16.0.1 ping statistics ---
  7. 1 packets transmitted, 1 packets received, 0% packet loss
  8. round-trip min/avg/max = 0.379/0.379/0.379 ms
  9. / # ping -c 1 100.64.0.1
  10. PING 100.64.0.1 (100.64.0.1): 56 data bytes
  11. 64 bytes from 100.64.0.1: seq=0 ttl=254 time=0.447 ms
  12.  
  13. --- 100.64.0.1 ping statistics ---
  14. 1 packets transmitted, 1 packets received, 0% packet loss
  15. round-trip min/avg/max = 0.447/0.447/0.447 ms
  16. / # ping -c 1 114.114.114.114
  17. PING 114.114.114.114 (114.114.114.114): 56 data bytes
  18. 64 bytes from 114.114.114.114: seq=0 ttl=87 time=40.669 ms
  19.  
  20. --- 114.114.114.114 ping statistics ---
  21. 1 packets transmitted, 1 packets received, 0% packet loss
  22. round-trip min/avg/max = 40.669/40.669/40.669 ms