GlusterFS 分布式文件系统的使用入门-管理GlusterFS卷
作者：尹正杰
版权声明：原创作品，谢绝转载！否则将追究法律责任。
一.卷的扩容
您可以根据需要在群集联机且可用时扩展卷。例如，您可能希望向分布式卷添加块，从而增加分发并增加GlusterFS卷的容量。
1>.添加扩展卷
[root@node101 ~]# df -h | grep gluster
node101.yinzhengjie.org.cn:/test-volume 23G 312M 22G % /mnt/gluster1
node101.yinzhengjie.org.cn:/replicated-volume 12G 156M 11G % /mnt/gluster2
node101.yinzhengjie.org.cn:/raid0-volume 23G 312M 22G % /mnt/gluster3
node101.yinzhengjie.org.cn:/distributed-replication-volume 12G 156M 11G % /mnt/gluster4
node101.yinzhengjie.org.cn:/my-distributed-replication-volume 12G 156M 11G % /mnt/gluster10
[root@node101 ~]#
[root@node101 ~]# mkdir /home/yinzhengjie/glusterfs/file8 #我们这里创建一个目录，用来模拟设备的挂载点
[root@node101 ~]#
[root@node101 ~]# gluster volume add-brick test-volume node101.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file8 #我们将块存储挂载到分布式卷中
volume add-brick: success
[root@node101 ~]#
[root@node101 ~]#
[root@node101 ~]# gluster volume info test-volume Volume Name: test-volume
Type: Distribute
Volume ID: d73f1306--4fea-8fe2-a37771b471d5
Status: Started
Snapshot Count:
Number of Bricks:
Transport-type: tcp
Bricks:
Brick1: node101.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file1
Brick2: node102.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file1
Brick3: node101.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file8 #这块磁盘就是咱们刚刚加上去的，注意生产环境操作基本上也是这样，只不过不是我们随手创建一个目录，而是把存储社保挂载到咱们指定到挂载点上而已！
Options Reconfigured:
transport.address-family: inet
nfs.disable: on
[root@node101 ~]#
2>.重新均衡扩展卷
往卷中添加一块设备后，需要咱们手动均衡一下，否则数据不会往新设备上存储哟，具体操作如下：
[root@node101 ~]# gluster volume rebalance test-volume start #重新均衡的适合需要等待一定的时间！
volume rebalance: test-volume: success: Rebalance on test-volume has been started successfully. Use rebalance status command to check status of the rebalance process.
ID: 7bd59481-af36-4fcc-9b04-9d421e4d6c0f
[root@node101 ~]#
3>.往卷中添加数据
[root@node101 ~]# ls /home/yinzhengjie/glusterfs/file8/
[root@node101 ~]#
[root@node101 ~]#
[root@node101 ~]# touch /mnt/gluster1/{..}.txt
[root@node101 ~]#
[root@node101 ~]# ls /mnt/gluster1/
.txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt blog.txt
.txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt
.txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt
.txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt
[root@node101 ~]#
[root@node101 ~]#
[root@node101 ~]# ls /home/yinzhengjie/glusterfs/file8/
.txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt
[root@node101 ~]#
[root@node101 ~]#
[root@node101 ~]# ls /home/yinzhengjie/glusterfs/file1/
.txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt
.txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt blog.txt
[root@node101 ~]#
[root@node101 ~]#
[root@node101 ~]# ssh node102.yinzhengjie.org.cn
Last login: Mon Feb :: from node101.yinzhengjie.org.cn
[root@node102 ~]#
[root@node102 ~]# ls /home/yinzhengjie/glusterfs/file1/
.txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt
.txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt
.txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt .txt
[root@node102 ~]#
二.卷的缩容
您可以根据需要在群集联机且可用时缩小卷。例如，您可能需要删除由于硬件或网络故障而在分布式卷中无法访问的块。
[root@node101 ~]# gluster volume info test-volume Volume Name: test-volume
Type: Distribute
Volume ID: d73f1306--4fea-8fe2-a37771b471d5
Status: Started
Snapshot Count:
Number of Bricks:
Transport-type: tcp
Bricks:
Brick1: node101.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file1
Brick2: node102.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file1
Brick3: node101.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file8
Options Reconfigured:
performance.client-io-threads: on
transport.address-family: inet
nfs.disable: on
[root@node101 ~]#
[root@node101 ~]# gluster volume remove-brick test-volume node101.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file8 start
Running remove-brick with cluster.force-migration enabled can result in data corruption. It is safer to disable this option so that files that receive writes during migration are not migrated.
Files that are not migrated can then be manually copied after the remove-brick commit operation.
Do you want to continue with your current cluster.force-migration settings? (y/n) y #这里会有一个提示，提示咱们可能存在数据丢失的风险，我输入y即可！由此可见官方并不推荐我们对GlusterFS进行缩容操作，对扩容操作倒是很支持呢！
volume remove-brick start: success
ID: 8e121850-c23b-4be3-94c4-c854148e5077
[root@node101 ~]#
[root@node101 ~]#
[root@node101 ~]# ls /home/yinzhengjie/glusterfs/file8/ #很显然这个目录对数据已经不再了，当然里面对数据会被转移走，这就是为什么执行上面的那条命令会等很长一段时间的缘由！
[root@node101 ~]#
[root@node101 gluster1]# gluster volume remove-brick test-volume node101.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file8 status
Node Rebalanced-files size scanned failures skipped status run time in h:m:s
--------- ----------- ----------- ----------- ----------- ----------- ------------ --------------
localhost 0Bytes completed :: #大家看着一行的status的值，如果为completed时，则说明可以执行下面的一条命令了，官方网站是这么说的！
[root@node101 gluster1]#
[root@node101 gluster1]#
[root@node101 gluster1]# gluster volume remove-brick test-volume node101.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file8 commit #提交我们一处该块磁盘的操作
volume remove-brick commit: success
Check the removed bricks to ensure all files are migrated.
If files with data are found on the brick path, copy them via a gluster mount point before re-purposing the removed brick.
[root@node101 gluster1]#
[root@node101 gluster1]# gluster volume info test-volume #再次查我们会发现node101.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file8这个文块设备不见啦！ Volume Name: test-volume
Type: Distribute
Volume ID: d73f1306--4fea-8fe2-a37771b471d5
Status: Started
Snapshot Count:
Number of Bricks:
Transport-type: tcp
Bricks:
Brick1: node101.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file1
Brick2: node102.yinzhengjie.org.cn:/home/yinzhengjie/glusterfs/file1
Options Reconfigured:
performance.client-io-threads: on
transport.address-family: inet
nfs.disable: on
[root@node101 gluster1]#
博主推荐阅读：https://docs.gluster.org/en/latest/Administrator%20Guide/Managing%20Volumes/