一、環境規劃
GlusterFS服務端:10.100.0.41/10.100.0.44
GlusterFS用戶端:10.100.0.43
二、所需軟體包
glusterfs-server-3.4.2-1.el6.x86_64
glusterfs-3.4.2-1.el6.x86_64
三、安裝
3.1、服務端安裝
#wget -P /etc/yum.repos.d http://download.gluster.org/pub/gluster/glusterfs/3.6/3.6.2/CentOS/glusterfs-epel.repo
#service glusterd start
#chkconfig glusterd on
3.2、配置存儲目錄
GlusterFS預設不能使用根分區作為GlusterFS的存儲目錄,本例中使用第二塊磁盤來做
# fdisk /dev/sdb
Device contains neither a valid DOS partition table, nor Sun, SGI or OSF disklabel
Building a new DOS disklabel with disk identifier 0x423f8af8.
Changes will remain in memory only, until you decide to write them.
After that, of course, the previous content won't be recoverable.
Warning: invalid flag 0x0000 of partition table 4 will be corrected by w(rite)
WARNING: DOS-compatible mode is deprecated. It's strongly recommended to
switch off the mode (command 'c') and change display units to
sectors (command 'u').
Command (m for help): n
Command action
e extended
p primary partition (1-4)
p
Partition number (1-4): 1
First cylinder (1-78325, default 1):
Using default value 1
Last cylinder, +cylinders or +size{K,M,G} (1-78325, default 78325):
Using default value 78325
Command (m for help): w
The partition table has been altered!
Calling ioctl() to re-read partition table.
Syncing disks.
# mkfs.ext4 /dev/sdb1
mke2fs 1.41.12 (17-May-2010)
Filesystem label=
OS type: Linux
Block size=4096 (log=2)
Fragment size=4096 (log=2)
Stride=0 blocks, Stripe width=0 blocks
39321600 inodes, 157286382 blocks
7864319 blocks (5.00%) reserved for the super user
First data block=0
Maximum filesystem blocks=4294967296
4800 block groups
32768 blocks per group, 32768 fragments per group
8192 inodes per group
Superblock backups stored on blocks:
32768, 98304, 163840, 229376, 294912, 819200, 884736, 1605632, 2654208,
4096000, 7962624, 11239424, 20480000, 23887872, 71663616, 78675968,
102400000
Writing inode tables: done
Creating journal (32768 blocks): done
Writing superblocks and filesystem accounting information: done
This filesystem will be automatically checked every 35 mounts or
180 days, whichever comes first. Use tune2fs -c or -i to override.
# mkdir /gfs
# mount /dev/sdb1 /gfs
# df -h
Filesystem Size Used Avail Use% Mounted on
/dev/sda2 30G 1.8G 27G 7% /
tmpfs 495M 0 495M 0% /dev/shm
/dev/sda1 194M 28M 157M 15% /boot
/dev/sda4 65G 180M 62G 1% /home
/dev/sdb1 591G 198M 561G 1% /gfs
3.3、加入随機啟動
#echo "/dev/sdb1 /gfs ext4 defaults 1 2" >> /etc/fstab
3.4、在第一台GlusterFS server中加入存儲節點。即(10.100.0.44)
# gluster peer probe 10.100.0.44 # 加入新節點
.......
3.5、檢視節點資訊
# gluster peer status
Number of Peers: 1
Hostname: 10.100.0.44
Uuid: c674ccbb-072b-4f26-97a2-facd15add645
State: Peer in Cluster (Connected)
3.5、建立存儲卷。即叢集磁盤
#gluster volume create gfsv replica 2 10.100.0.41:/gfs 10.100.0.44:/gfs
Creation of volume gfsv has been successful. Please start the volume to access data.
3.6、啟動存儲卷
# gluster volume start gfsv # 啟動卷
3.7、檢視存儲卷資訊
# gluster volume info
Volume Name: gfsv
Type: Replicate
Volume ID: be627a08-1d36-40f5-add0-1277be47c328
Status: Started
Number of Bricks: 1 x 2 = 2
Transport-type: tcp
Bricks:
Brick1: 10.100.0.41:/gfs/data
Brick2: 10.100.0.44:/gfs/data
3.8、配置通路ACL規則
# gluster volume set gfs auth.allow 10.100.0.* # 授權通路
四、用戶端安裝配置
4.1、安裝用戶端
用戶端需要fuse支援
# yum -y install glusterfs glusterfs-fuse glusterfs-api glusterfs-devel glusterfs-libs
4.2、挂載glusterfs存儲
#mount -t glusterfs -o rw 10.100.0.41:gfsv /home/gfs/data/
五:存儲卷模式
(1)distribute volume:分布式卷,檔案通過hash算法分布到brick server上,這種卷是glusterfs的基礎和最大特點;
(2)stripe volume:條帶卷,類似RAID0,條帶數=brick server數量,檔案分成資料塊以Round Robin方式分布到brick server上,并發粒度是資料塊,大檔案性能高;
(3)replica volume:鏡像卷,類似RAID1,鏡像數=brick server數量,是以brick server上檔案資料相同,構成n-way鏡像,可用性高;
(4)distribute stripe volume:分布式條帶卷,brick server數量是條帶數的倍數,兼具distribute和stripe卷的特點;
(5)distribute replica volume:分布式鏡像卷,brick server數量是鏡像數的倍數,兼具distribute和replica卷的特點;
六、 平衡布局是很有必要的,因為布局結構是靜态的,當新的bricks加入現有卷,新建立的檔案會分布到舊的bricks中,是以需要平衡布局結構,使新加入 的bricks生效。布局平衡隻是使新布局生效,并不會在新的布局移動老的資料,如果你想在新布局生效後,重新平衡卷中的資料,還需要對卷中的資料進行平 衡。
gluster volume rebalance VOLNAME fix-layout start
gluster volume rebalance VOLNAME migrate-data start
也可以兩步合一步同時操作
gluster volume rebalance VOLNAME start
你也可以删除/停止卷
gluster volume stop/delete VOLNAME
你可以在在平衡過程中檢視平衡資訊:
你也可以暫停平衡,再次啟動平衡的時候會從上次暫停的地方繼續開始平衡。
gluster volume rebalance VOLNAME stop
本文轉自 亮公子 51CTO部落格,原文連結:http://blog.51cto.com/iyull/1946587