GlusterFS简介
参考：https://blog.51cto.com/13447608/2386830?source=dra
GlusterFS安装
如下所示三台机器，hostname分别是node-1/node-2/node-3，在/etc/hosts添加3台机器的主机名解析，关闭selinux、防火墙、NetworkManager等。
| node-1 | 10.0.0.1 |
| node-2 | 10.0.0.2 |
| node-3 | 10.0.0.3 |
安装glusterfs(三台都要执行)
#安装glusterfs yum install centos-release-gluster yum install -y glusterfs glusterfs-server glusterfs-fuse glusterfs-rdma #启动gluster服务 systemctl start glusterd.service systemctl enable glusterd.service
在node1上执行，将node-2、node-3加入集群
gluster peer probe node-2 gluster peer probe node-3
在node-1上看集群状态
[root@node-1 db]# gluster peer status Number of Peers: 2 Hostname: node-2 Uuid: d8d232f-e5fb-4752-b40d-b0103afrf8ea State: Peer in Cluster (Connected) Hostname: node-3 Uuid: ff3gg732-ae12-4b6c-4rfb-65fd7886588c State: Peer in Cluster (Connected)
在node-1上执行，创建3副本的复制卷。glusterfs使用现有Linux文件系统创建glusterfs文件系统。
#创建卷[root@node-1 ~]# gluster volume create app-data replica 3 transport tcp node-1:/gfs-data/ node-2:/gfs-data/ node-3:/gfs-data/ force volume create: app-data: success: please start the volume to access data
# 列出卷[root@k8sm0 ~]# gluster volume listapp-data# 查看卷信息[root@node-1 ~]# gluster volume infoVolume Name: app-dataType: ReplicateVolume ID: rr2068b2-0da6-431f-8f93-3d7b2f331gdeStatus: CreatedSnapshot Count: 0Number of Bricks: 1 x 3 = 3Transport-type: tcpBricks:Brick1: node-1:/gfs-dataBrick2: node-2:/gfs-dataBrick3: node-3:/gfs-dataOptions Reconfigured:transport.address-family: inetnfs.disable: onperformance.client-io-threads: off# 启动卷[root@node-1 ~]# gluster volume start app-datavolume start: app-data: success[root@node-1 ~]# gluster volume quota app-data enable volume quota : success[root@node-1 ~]# gluster volume quota app-data limit-usage / 10GBvolume quota : success# 查看卷状态[root@node-1 app]# gluster volume statusStatus of volume: app-dataGluster process TCP Port RDMA Port Online Pid------------------------------------------------------------------------------Brick node-1:/gfs-data 49152 0 Y 6283Brick node-2:/gfs-data 49152 0 Y 6216Brick node-3:/gfs-data 49152 0 Y 6232#挂载[root@node-1 ~]# mkdir /gfs-share && mount -t glusterfs localhost:k8s-data /gfs-share[root@node-1 ~]# echo "localhost:app-data /gfs-share/ glusterfs defaults 0 0" >> /etc/fstab