glusterfs 无法启动。
查看日志/var/log/glusterfs/etc-glusterfs-glusterd.vol.log 报如下错误：
[2017-03-10 05:26:09.228295] I [glusterfsd.c:1910:main] 0-/usr/sbin/glusterd: Started running /usr/sbin/glusterd version 3.4.2 (/usr/sbin/glusterd -p /var/run/glusterd.pid)
[2017-03-10 05:26:09.230929] I [glusterd.c:961:init] 0-management: Using /var/lib/glusterd as working directory
[2017-03-10 05:26:09.231616] I [socket.c:3480:socket_init] 0-socket.management: SSL support is NOT enabled
[2017-03-10 05:26:09.231628] I [socket.c:3495:socket_init] 0-socket.management: using system polling thread
[2017-03-10 05:26:09.232036] W [rdma.c:4197:__gf_rdma_ctx_create] 0-rpc-transport/rdma: rdma_cm event channel creation failed (No such device)
[2017-03-10 05:26:09.232047] E [rdma.c:4485:init] 0-rdma.management: Failed to initialize IB Device
[2017-03-10 05:26:09.232052] E [rpc-transport.c:320:rpc_transport_load] 0-rpc-transport: 'rdma' initialization failed
[2017-03-10 05:26:09.232085] W [rpcsvc.c:1389:rpcsvc_transport_create] 0-rpc-service: cannot create listener, initing the transport failed
[2017-03-10 05:26:09.819847] E [glusterd-store.c:1333:glusterd_restore_op_version] 0-management: wrong op-version (30600) retrieved
[2017-03-10 05:26:09.819896] E [glusterd-store.c:2510:glusterd_restore] 0-management: Failed to restore op_version
[2017-03-10 05:26:09.819916] E [xlator.c:390:xlator_init] 0-management: Initialization of volume 'management' failed, review your volfile again
[2017-03-10 05:26:09.819930] E [graph.c:292:glusterfs_graph_init] 0-management: initializing translator failed
[2017-03-10 05:26:09.819941] E [graph.c:479:glusterfs_graph_activate] 0-graph: init failed
[2017-03-10 05:26:09.820294] W [glusterfsd.c:1002:cleanup_and_exit] (-->/usr/sbin/glusterd(main+0x3df) [0x7f1524b238df] (-->/usr/sbin/glusterd(glusterfs_volumes_init+0xb0) [0x7f1524b266d0]
(-->/usr/sbin/glusterd(glusterfs_process_volfp+0x103) [0x7f1524b265e3]))) 0-: received signum (0), shutting down
报错是无法初始化glusterfs，可能之前已经安装过glusterfs。
解决方法：
将 /var/lib/glusterd/ 目录下所有文件清空，再重新启动
/etc/init.d/glusterfs-server start
恢复正常。