>#NFS本地集群动态配置持久卷 [TOC] <br/> 关闭防火墙 ------------ <br/> ```bash systemctl stop firewalld systemctl disable firewalld yum install -y nfs-common nfs-utils ``` <br/> 创建共享目录 ------------ <br/> ```bash mkdir /nfsdata ``` <br/> 授权共享目录 ------------ <br/> ```bash chmod 666 /nfsdata ``` <br/> 编辑exports文件 ------------ <br/> ```bash vi /etc/exports /nfsdata *(rw,no_root_squash,no_all_squash,sync) ``` （不严谨规则） ```bash vi /etc/exports /nfsdata *(rw,async) ``` >把*改成网段 比如在本地虚拟机的 192.168.139.0/24 /nfsdata 192.168.139.0/24(rw,no_root_squash,no_all_squash,sync) 192.168.139.0/24 跟()之间不能有空格 切记 <br/> 配置生效 ------------ <br/> ```bash exportfs -r 或 exportfs -rv ``` <br/> 启动rpc和nfs（注意顺序） ------------ <br/> >必须先启动rpcbind服务，再启动nfs服务，这样才能让nfs服务在rpcbind服务上注册成功： ```bash systemctl start rpcbind systemctl start nfs systemctl enable rpcbind.service nfs ``` >在NFS服务器上创建成功共享目录 /nfsdata： 执行 showmount -e 判断NFS创建的共享目录是否成功 ```bash showmount -e (Export list for K8sNFS: /nfsdata 192.168.69.0/24) ``` >在k8s工作节点上安装NFS客户端 ```bash yum install -y nfs-utils ``` >判断k8s从节点是否可以挂载NFS共享的目录 ```bash showmount -e 192.168.139.128 (192.168.139.128 换成NFS的地址) Export list for 192.168.139.128: /nfsdata 192.168.139.0/24 ``` ```bash helm install nfs-client-provision --set nfs.server=192.168.139.128 --set nfs.path=/nfsdata ./nfs-client-provinsioner ``` https://github.com/kubernetes-incubator/external-storage/tree/master/nfs-client ## Ubuntu ```bash apt-get install nfs-kernel-server -y sudo ufw disable mkdir /nfsdata chmod 666 /nfsdata echo "/nfsdata *(async,insecure,no_root_squash,no_subtree_check,rw)" > /etc/exports cat /etc/exports /etc/init.d/nfs-kernel-server restart showmount -e ```