练习一：创建卷组
准备3块10G的空闲分区，将类型ID修改为8e （LVM）
[root@localhost~]# fdisk /dev/sdb
进入交互模式通过新建n-p-分区号-开始位置-结束位置（分区大小）
交互模式t修改类型-分区号-类型为8e
DeviceBoot Start End Blocks Id System
/dev/sdb1 1 1217 9775521 8e Linux LVM
W保存后退出
使用其中2块分区组建名为myvg的卷组，查看此卷组信息
先检查有哪些物理卷
[root@localhost~]# pvscan
Nomatching physical volumes found
将两块空闲分区转换成物理卷
例：[root@localhost~]# pvcreate /dev/sdb1
Writingphysical volume data to disk "/dev/sdb1"
Physicalvolume "/dev/sdb1" successfully created
再检查有哪些物理卷，查看其中一个物理卷的详细信息
[root@localhost~]# pvscan
PV/dev/sdb1 lvm2 [9.32 GB]
PV/dev/sdb2 lvm2 [9.32GB]
Total:2 [18.65 GB] / in use: 0 [0 ] / in noVG: 2 [18.65 GB]
[root@localhost~]# pvdisplay /dev/sdb1
"/dev/sdb1" is anew physical volume of "9.32 GB"
--- NEW Physical volume ---
PV Name /dev/sdb1
VG Name
PV Size 9.32 GB
Allocatable NO
PE Size (KByte) 0
Total PE 0
Free PE 0
Allocated PE 0
PV UUID 9QuHkE-pXKI-tlWM-vJdv-2qmt-Sd3A-p8Sbwq
先查看有哪些卷组
[root@localhost~]# vgdisplay
Novolume groups found
将两个物理卷整编成卷组myvg
[root@localhost~]# vgcreatemyvg /dev/sdb1 /dev/sdb2
Volumegroup "myvg" successfully created
再查看有哪些卷组，并查看卷组myvg的详细信息
[root@localhost~]# vgdisplay
---Volume group ---
VGName myvg
SystemID
Format lvm2
MetadataAreas 2
MetadataSequence No 1
VG Access read/write
VGStatus resizable
MAXLV 0
CurLV 0
OpenLV 0
MaxPV 0
CurPV 2
ActPV 2
VGSize 18.64 GB
PESize 4.00 MB
TotalPE 4772
AllocPE / Size 0 / 0
Free PE / Size 4772 / 18.64 GB
VGUUID oSPZlv-Gt6D-gTQA-Gmw6-OsRd-TRqD-gcfbr0
练习二：创建/使用/扩展逻辑卷
划分一个16G的逻辑卷，名称为lvmox，查看逻辑卷信息
[root@localhost~]# lvcreate -L 16G -n lvmoxmyvg
Logicalvolume "lvmox" created
[root@localhost~]# lvdisplay
---Logical volume ---
LVName /dev/myvg/lvmox
VGName myvg
LVUUID r22EGe-Cvg5-D1Qf-Q6lt-s3SJ-XuL1-gIALQD
LVWrite Access read/write
LVStatus available
#open 0
LVSize 16.00 GB
CurrentLE 4096
Segments 2
Allocation inherit
Readahead sectors auto
-currently set to 256
Blockdevice 253:0
将此逻辑卷格式化为ext3文件系统，并挂载到/mbox目录
格式化该逻辑卷:
[root@localhost~]# mkfs.ext3 /dev/myvg/lvmox
挂载
[root@localhost~]# mkdir /mbox
[root@localhost~]# mount /dev/myvg/lvmox /mbox/
通过mount命令查看：
/dev/mapper/myvg-lvmoxon /mbox type ext3 (rw)
进入/mbox目录，测试读写操作
写入：[root@localhostmbox]#ifconfig> 121.txt
[root@localhostmbox]#ls
121.txtlost+found
读取：[root@localhostmbox]# cat 121.txt
eth0 Link encap:EthernetHWaddr00:0C:29:19:BB:76
将逻辑卷从16G扩展为24G，确保df识别的大小准确
先扩展卷组（增加一个10G物理卷），再扩展逻辑卷
[root@localhostmbox]#vgextendmyvg /dev/sdb3
Nophysical volume label read from /dev/sdb3
Writingphysical volume data to disk "/dev/sdb3"
Physicalvolume "/dev/sdb3" successfully created
Volumegroup "myvg" successfully extended
扩展逻辑卷：
[root@localhostmbox]#lvextend -L +8G /dev/myvg/lvmox
Extendinglogical volume lvmox to 24.00 GB
Logicalvolume lvmox successfully resized
resize2fs识别新文件系统的大小
[root@localhostmbox]#resize2fs /dev/myvg/lvmox
创建一个大小为250M的逻辑卷lvtest
[root@localhostmbox]#vgchange -s 1M myvg
Volumegroup "myvg" successfully changed
查看：
[root@localhostmbox]#vgdisplay
---Volume group ---
VGName myvg
SystemID
Format lvm2
MetadataAreas 3
MetadataSequence No 5
VGAccess read/write
VGStatus resizable
MAXLV 0
CurLV 1
OpenLV 1
MaxPV 0
CurPV 3
ActPV 3
VGSize 27.96 GB
PESize 1.00 MB
TotalPE 28632
AllocPE / Size 24576 / 24.00 GB
Free PE / Size 4056 / 3.96 GB
VGUUID oSPZlv-Gt6D-gTQA-Gmw6-OsRd-TRqD-gcfbr0
练习三：逻辑卷综合应用
删除上一练习建立的卷组myvg
保证没有使用或者挂载的时候删除
[root@localhost~]# vgremovemyvg
Doyou really want to remove volume group "myvg" containing 1 logicalvolumes? [y/n]: y
Doyou really want to remove active logical volume lvmox? [y/n]: y
Logicalvolume "lvmox" successfully removed
Volumegroup "myvg" successfully removed
使用其中2个物理卷组成卷组vgnsd，另一个物理卷组成卷组vgdata
[root@localhost~]# vgcreatevgnsd /dev/sdb1 /dev/sdb2
Volumegroup "vgnsd" successfully created
[root@localhost~]# vgcreatevgdata /dev/sdb3
Volumegroup "vgdata" successfully created
从卷组vgnsd中创建一个20G的逻辑卷lvhome
[root@localhost~]# lvcreate -L 16G -n lvhomevgnsd
Logicalvolume "lvhome" created
从卷组vgdata中创建一个4G的逻辑卷lvswap
[root@localhost~]# lvcreate -L 4G -n lvswapvgdata
Logicalvolume "lvswap" created
将/home目录迁移到逻辑卷lvhome上
[root@localhost~]# mkfs.ext3 /dev/vgnsd/lvhome
[root@localhost~]# mkdir /1
[root@localhost~]# mv /home/* /1
[root@localhost~]# mount /dev/vgnsd/lvhome /home
/dev/mapper/vgnsd-lvhomeon /home type ext3 (rw)
将逻辑卷lvswap扩展到交换空间
格式化逻辑卷lvswap：
[root@localhost~]# mkswap /dev/vgdata/lvswap
Settingup swapspace version 1, size = 4294963 kB
[root@localhost~]# swapon /dev/vgdata/lvswap
[root@localhost~]# swapon -s
Filename Type Size Used Priority
/dev/sda3partition 200804 0 -1
/dev/mapper/vgdata-lvswap partition 4194296 0 -2
为第5、6步配置开机自动挂载，重启后验证
通过以下加载
[root@localhost ~]# vim/etc/fstab
练习四：创建软RAID阵列
添加4块大小均为20GB的空磁盘
将其中第一块、第二块磁盘划分为单个主分区
把上述分区的类型ID改成fd
Device Boot Start End Blocks Id System
/dev/sdb1 1 2610 20964793+ fd Linux raid autodetect
2）阵列创建练习
创建RAID0设备/dev/md0、RAID1设备/dev/md1
[root@localhost ~]# mdadm -C/dev/md0 -l0 -n2 /dev/sdb1 /dev/sdc1
mdadm: array /dev/md0started.
[root@localhost ~]# mdadm -C/dev/md1 -l1 -n2 /dev/sdd /dev/sde
mdadm: array /dev/md1started.
b）查看这两个阵列的容量及成员盘个数 【-Q、-D】
[root@localhost ~]# mdadm -D/dev/md0
/dev/md0:
Version : 0.90
Creation Time : Wed Jun 4 19:04:41 2014
Raid Level : raid0
Array Size : 41929344 (39.99GiB 42.94 GB)
Raid Devices : 2
Total Devices : 2
Preferred Minor : 0
Persistence : Superblock ispersistent
Update Time : Wed Jun 4 19:04:41 2014
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
Chunk Size : 64K
UUID :923d3722:10437de4:f871f97a:b358ef7b
Events : 0.1
Number Major Minor RaidDevice State
0 8 17 0 active sync /dev/sdb1
1 8 33 1 active sync /dev/sdc1
[root@localhost ~]# mdadm -D/dev/md1
/dev/md1:
Version : 0.90
Creation Time : Wed Jun 4 19:05:15 2014
Raid Level : raid1
Array Size : 20971456 (20.00GiB 21.47 GB)
Used DevSize : 20971456(20.00 GiB 21.47 GB)
Raid Devices : 2
Total Devices : 2
Preferred Minor : 1
Persistence : Superblock ispersistent
Update Time : Wed Jun 4 19:06:59 2014
State : clean
Active Devices : 2
Working Devices : 2
Failed Devices : 0
Spare Devices : 0
UUID :1a6e3772:e4b55604:dbe09f01:b78a3faa
Events : 0.4
Number Major Minor RaidDevice State
0 8 48 0 active sync /dev/sdd
1 8 64 1 active sync /dev/sde
解散并删除阵列设备/dev/md0和/dev/md1 【-S】
[root@localhost~]# mdadm -S /dev/md0
mdadm:stopped /dev/md0
[root@localhost~]# rm -rf /dev/md0
创建RAID5软阵列设备 /dev/md0
第一个成员盘用分区来做
其余三个成员盘用整块磁盘来做
用fdisk分别查看第一块、第二块磁盘的分区表
[root@localhost ~]# mdadm -C/dev/md0 -l5 -n4 /dev/sdb1 /dev/sd[c-e]
mdadm: /dev/sdb1 appears tobe part of a raid array:
level=raid0 devices=2ctime=Wed Jun 4 19:04:41 2014
mdadm: /dev/sdd appears tobe part of a raid array:
level=raid1 devices=2ctime=Wed Jun 4 19:05:15 2014
mdadm: /dev/sde appears tobe part of a raid array:
level=raid1 devices=2ctime=Wed Jun 4 19:05:15 2014
Continue creating array? y
mdadm: array /dev/md0started.
练习五：格式化并使用阵列
将RAID5阵列/dev/md0格式化成EXT3文件系统
[root@localhost~]# mkfs.ext3 /dev/md0
将阵列设备/dev/md0挂载到/mymd目录
[root@localhost~]# mkdir /mymd
[root@localhost~]# mount /dev/md0 /mymd/
Mount查看
/dev/md0on /mymd type ext3 (rw)
进入/mymd目录，测试读写
写入：[root@localhostmymd]#ls> 12.txt
[root@localhostmymd]#ls
12.txtlost+found
读取：[root@localhostmymd]#cat 12.txt
12.txt
lost+found
练习六：RAID5阵列的故障测试
通过VMware设置拔掉阵列/dev/md0的最后一个成员
[root@localhostmymd]#mdadm -D /dev/md0
/dev/md0:
Version: 0.90
CreationTime : Wed Jun 4 19:10:30 2014
RaidLevel : raid5
ArraySize : 62894016 (59.98 GiB 64.40 GB)
UsedDevSize : 20964672 (19.99 GiB 21.47 GB)
RaidDevices : 4
TotalDevices : 4
PreferredMinor : 0
Persistence: Superblock is persistent
UpdateTime : Wed Jun 4 19:16:02 2014
State: clean, degraded
ActiveDevices : 3
WorkingDevices : 3
FailedDevices : 1
SpareDevices : 0
Layout: left-symmetric
ChunkSize : 64K
UUID: 8a0dd0eb:2fdf8913:00f9e8e9:972e8b80
Events: 0.14
Number Major Minor RaidDevice State
0 8 17 0 active sync /dev/sdb1
1 8 32 1 active sync /dev/sdc
2 8 48 2 active sync /dev/sdd
3 0 0 3 removed
4 8 64 - faulty spare /dev/sde
2）再次访问/mymd，测试读写
3）RAID5阵列的故障盘替换
将已失效的成员盘标记为失败
[root@localhostmymd]# mdadm/dev/md0 -f /dev/sde
mdadm: set /dev/sde faultyin /dev/md0
移除已失效的成员盘
[root@localhostmymd]# mdadm/dev/md0 -r /dev/sde
mdadm: hot removed /dev/sde
重新添加一个完好的成员盘（与其他成员盘大小一致）
[root@localhostmymd]#mdadm /dev/md0 -a /dev/sde
mdadm:added /dev/sde
观察阵列状态信息，查看修复过程
[root@localhostmymd]#watch cat /proc/mdstat
Every2.0s: cat /proc/mdstat Wed Jun 4 19:22:10 2014
Personalities: [raid0] [raid1] [raid6] [raid5] [raid4]
md0: active raid5 sde[4] sdd[2] sdc[1] sdb1[0]
62894016blocks level 5, 64k chunk, algorithm 2 [4/3] [UUU_]
[================>....] recovery = 82.3% (17257344/20964672) fi
nish=0.3minspeed=196343K/sec
unuseddevices: <none>
练习七：保存、重组阵列
查询当前正在运行的阵列设置
[root@localhostmymd]#mdadm -vDs
ARRAY/dev/md0 level=raid5 num-devices=4 metadata=0.90UUID=8a0dd0eb:2fdf8913:00f9e8e9:972e8b80
devices=/dev/sdb1,/dev/sdc,/dev/sdd,/dev/sde
保存正在运行的阵列设置为/etc/mdadm.conf
[root@localhostmymd]#mdadm -vDs> /etc/mdadm.conf
解散并删除阵列/dev/md0
[root@localhost~]# umount /dev/md0
[root@localhost~]# mdadm -S /dev/md0
mdadm:stopped /dev/md0
[root@localhost~]# rm -rf /dev/md0
重组阵列/dev/md0，并挂载测试
[root@localhost~]# mdadm -A /dev/md0
mdadm:/dev/md0 has been started with 4 drives.
[root@localhost~]# mount /dev/md0 /mymd/
/dev/md0on /mymd type ext3 (rw)