GlusterFS之故障节点更换
1. 目标
原有的glusterfs集群,采用2*2的分布式复制卷
现在有两个节点发生了硬盘故障,而且两个节点是同样的文件,所以造成了部分数据的丢失。
2. 恢复gluster集群
1. 准备工作
安装系统,安装gluster软件等。略
172.16.1.98 node1
172.16.1.99 node2
2. 启动gluster软件
3. 将原来的节点所包含的brick删除
gluster volume remove-brick movie_rep_1 replica 2 node1:/gluster_data/vol_rep_1 node2:/gluster_data/vol_rep_1 force
4. 将故障节点删除
gluster peer detach node1 force
gluster peer detach node2 force
5. 将两个新节点加入集群
gluster peer probe node1
gluster peer probe node2
6. 添加两个节点的brick
gluster volume add-brick vol_rep_1 replica 2 node2:/gluster_data/vol_rep_1 node5:/gluster_data/vol_rep_1
7. 查看该卷的信息
gluster volume info vol_rep_1
Volume Name: vol_rep_1 Type: Distributed-Replicate Volume ID: 294166af-db23-4253-963b-7f0e6b8898bf Status: Started Snapshot Count: 0 Number of Bricks: 2 x 2 = 4 Transport-type: tcp Bricks: Brick1: node3:/gluster_data/vol_rep_1 Brick2: node4:/gluster_data/vol_rep_1 Brick3: node2:/gluster_data/vol_rep_1 Brick4: node5:/gluster_data/vol_rep_1 Options Reconfigured: nfs.disable: on network.tcp-window-size: 1048576 performance.read-ahead: on performance.cache-size: 512MB performance.read-ahead-page-count: 4 performance.io-thread-count: 32 cluster.self-heal-daemon: enable network.inode-lru-limit: 200000 performance.md-cache-timeout: 600 performance.cache-invalidation: on performance.stat-prefetch: on features.cache-invalidation-timeout: 600 features.cache-invalidation: on
8. 查看数据恢复的状态
[root@node4 vol_rep_1]# gluster volume heal vol_rep_1 info Brick node3:/gluster_data/vol_rep_1 Status: Connected Number of entries: 0 Brick node4:/gluster_data/vol_rep_1 Status: Connected Number of entries: 0 Brick node2:/gluster_data/vol_rep_1 Status: Connected Number of entries: 0 Brick node1:/gluster_data/vol_rep_1 Status: Connected Number of entries: 0
9. 如果有数据没有同步,可以强制修复卷
gluster volume heal vol_rep_1 full
 
                    
                     
                    
                 
                    
                
 
                
            
         
         浙公网安备 33010602011771号
浙公网安备 33010602011771号