网络参考: http://molinux.blog.51cto.com/2536040/518441
1. LVM基本创建及管理
2. LVM快照
3. LVM与RAID的结合使用:基于RAID的LVM
LVM创建:
描述:
LVM是逻辑盘卷管理(LogicalVolumeManager)的简称,它是Linux环境下对磁盘分区进行管理的一种机制,LVM是建立在硬盘和分区之上的一个逻辑层,来提高磁盘分区管理的灵活性.
通过创建LVM,我们可以更轻松的管理磁盘分区,将若干个不同大小的不同形式的磁盘整合为一个整块的卷组,然后在卷组上随意的创建逻辑卷,既避免了大量不同规格硬盘的管理难题,也使逻辑卷容量的扩充缩减不再受限于磁盘规格;并且LVM的snapshot(快照)功能给数据的物理备份提供了便捷可靠的方式;
创建LVM过程;(如图)
1. 在物理设备上创建物理分区,每个物理分区称为一个PE
2. 使用fdisk工具创建物理分区卷标(修改为8e),形成PV(Physical Volume 物理卷)
3. 使用vgcreate 将多个PV添加到一个VG(Volume Group 卷组)中,此时VG成为一个大磁盘;
4. 在VG大磁盘上划分LV(Logical Volume 逻辑卷),将逻辑卷格式化后即可挂载使用;
各阶段可用的命令工具:(详细选项信息请be a man)
st1\:*{behavior:url(#ieooui) }
阶段 | 显示信息 | 创建 | 删除组员 | 扩大大小 | 缩减大小 |
PV | pvdisplay | pvcreat | pvremove | ----- | ----- |
VG | vgdisplay | vgcreat | vgremove | vgextend | vgreduce |
LV | lvdispaly | lvcreat | lvremove | lvextend | lvreduce |
创建示例:
1. 创建PV
- Device Boot Start End Blocks Id System
- /dev/sdb1 1 609 4891761 8e Linux LVM
- /dev/sdc1 1 609 4891761 8e Linux LVM
- /dev/sdd1 1 609 4891761 8e Linux LVM
- [root@bogon ~]# pvcreate /dev/sd[bcd]1
- Physical volume "/dev/sdb1" successfully created
- Physical volume "/dev/sdc1" successfully created
- Physical volume "/dev/sdd1" successfully created
查看PV信息:
- [root@bogon ~]# pvdisplay
- --- Physical volume ---
- PV Name /dev/sda2
- VG Name vol0
- PV Size 40.00 GB / not usable 2.61 MB
- Allocatable yes
- PE Size (KByte) 32768
- Total PE 1280
- Free PE 281
- Allocated PE 999
- PV UUID GxfWc2-hzKw-tP1E-8cSU-kkqY-z15Z-11Gacd
- "/dev/sdb1" is a new physical volume of "4.67 GB"
- --- NEW Physical volume ---
- PV Name /dev/sdb1
- VG Name
- PV Size 4.67 GB
- Allocatable NO
- PE Size (KByte) 0
- Total PE 0
- Free PE 0
- Allocated PE 0
- PV UUID 1rrc9i-05Om-Wzd6-dM9G-bo08-2oJj-WjRjLg
- "/dev/sdc1" is a new physical volume of "4.67 GB"
- --- NEW Physical volume ---
- PV Name /dev/sdc1
- VG Name
- PV Size 4.67 GB
- Allocatable NO
- PE Size (KByte) 0
- Total PE 0
- Free PE 0
- Allocated PE 0
- PV UUID RCGft6-l7tj-vuBX-bnds-xbLn-PE32-mCSeE8
- "/dev/sdd1" is a new physical volume of "4.67 GB"
- --- NEW Physical volume ---
- PV Name /dev/sdd1
- VG Name
- PV Size 4.67 GB
- Allocatable NO
- PE Size (KByte) 0
- Total PE 0
- Free PE 0
- Allocated PE 0
- PV UUID SLiAAp-43zX-6BC8-wVzP-6vQu-uyYF-ugdWbD
2. 创建VG
- [[root@bogon ~]# vgcreate myvg -s 16M /dev/sd[bcd]1
- Volume group "myvg" successfully created
- ##-s 在创建时指定PE块的大小,默认是4M。
- 查看系统上VG状态
- [root@bogon ~]# vgdisplay
- --- Volume group ---
- VG Name myvg
- System ID
- Format lvm2
- Metadata Areas 3
- Metadata Sequence No 1
- VG Access read/write
- VG Status resizable
- MAX LV 0
- Cur LV 0
- Open LV 0
- Max PV 0
- Cur PV 3
- Act PV 3
- VG Size 13.97 GB ##VG总大小
- PE Size 16.00 MB ##默认的PE块大小是4M
- Total PE 894 ##总PE块数
- Alloc PE / Size 0 / 0 ##已经使用的PE块数目
- Free PE / Size 894 / 13.97 GB ##可用的PE数目及磁盘大小
- VG UUID RJC6Z1-N2Jx-2Zjz-26m6-LLoB-PcWQ-FXx3lV
3. 在VG中划分出LV:
- [root@bogon ~]# lvcreate -L 256M -n data1 myvg
- Logical volume "data1" created
- ## -L指定LV大小
- ## -n 指定lv卷名称
- [root@bogon ~]# lvcreate -l 20 -n test myvg
- Logical volume "test" created
- ## -l 指定LV大小占用多少个PE块;上面大小为:20*16M=320M
- [root@bogon ~]# lvdisplay
- --- Logical volume ---
- LV Name /dev/myvg/data
- VG Name myvg
- LV UUID d0SYy1-DQ9T-Sj0R-uPeD-xn0z-raDU-g9lLeK
- LV Write Access read/write
- LV Status available
- # open 0
- LV Size 256.00 MB
- Current LE 16
- Segments 1
- Allocation inherit
- Read ahead sectors auto
- - currently set to 256
- Block device 253:2
- --- Logical volume ---
- LV Name /dev/myvg/test
- VG Name myvg
- LV UUID os4UiH-5QAG-HqOJ-DoNT-mVeT-oYyy-s1xArV
- LV Write Access read/write
- LV Status available
- # open 0
- LV Size 320.00 MB
- Current LE 20
- Segments 1
- Allocation inherit
- Read ahead sectors auto
- - currently set to 256
- Block device 253:3
4. 然后就可以格式化LV,挂载使用:
- [root@bogon ~]# mkfs -t ext3 -b 2048 -L DATA /dev/myvg/data
- [root@bogon ~]# mount /dev/myvg/data /data/
- ## 拷贝进去一些文件,测试后面在线扩展及缩减效果:
- [root@bogon data]# cp /etc/* /data/
5. 扩大LV容量:(2.6kernel+ext3filesystem)
lvextend命令可以增长逻辑卷
resize2fs可以增长filesystem在线或非在线
(我的系统内核2.6.18,ext3文件系统,查man文档,现在只有2.6的内核+ext3文件系统才可以在线增容量)
首先逻辑扩展:
- [root@bogon data]# lvextend -L 500M /dev/myvg/data
- Rounding up size to full physical extent 512.00 MB
- Extending logical volume data to 512.00 MB
- Logical volume data successfully resized
- ##-L 500M :指扩展到500M,系统此时会找最近的柱面进行匹配;
- ##-L +500M :值在原有大小的基础上扩大500M;
- ##-l [+]50 类似上面,但是以Pe块为单位进行扩展;
然后文件系统物理扩展:
- [root@bogon data]# resize2fs -p /dev/myvg/data
- resize2fs 1.39 (29-May-2006)
- Filesystem at /dev/myvg/data is mounted on /data; on-line resizing required
- Performing an on-line resize of /dev/myvg/data to 262144 (2k) blocks.
- The filesystem on /dev/myvg/data is now 262144 blocks long.
- ##据上面信息,系统自动识别并进行了在线扩展;
- 查看状态:
- [root@bogon ~]# lvdisplay
- --- Logical volume ---
- LV Name /dev/myvg/data
- VG Name myvg
- LV UUID d0SYy1-DQ9T-Sj0R-uPeD-xn0z-raDU-g9lLeK
- LV Write Access read/write
- LV Status available
- # open 1
- LV Size 512.00 MB
- Current LE 32
- Segments 1
- Allocation inherit
- Read ahead sectors auto
- - currently set to 256
- Block device 253:2
- ##此时查看挂载目录,文件应该完好;
6. 缩减LV容量:
缩减容量是一件危险的操作;缩减必须在离线状态下执行;并且必须先强制检查文件系统错误,防止缩减过程损坏数据;
- [root@bogon ~]# umount /dev/myvg/data
- [root@bogon ~]# e2fsck -f /dev/myvg/data
- e2fsck 1.39 (29-May-2006)
- Pass 1: Checking inodes, blocks, and sizes
- Pass 2: Checking directory structure
- Pass 3: Checking directory connectivity
- Pass 4: Checking reference counts
- Pass 5: Checking group summary information
- DATA: 13/286720 files (7.7% non-contiguous), 23141/573440 blocks
先缩减物理大小:
- [root@bogon ~]# resize2fs /dev/myvg/data 256M
- resize2fs 1.39 (29-May-2006)
- Resizing the filesystem on /dev/myvg/data to 131072 (2k) blocks.
- The filesystem on /dev/myvg/data is now 131072 blocks long.
再缩减逻辑大小:
- [root@bogon ~]# lvreduce -L 256M /dev/myvg/data
- WARNING: Reducing active logical volume to 256.00 MB
- THIS MAY DESTROY YOUR DATA (filesystem etc.)
- Do you really want to reduce data? [y/n]: y
- Reducing logical volume data to 256.00 MB
- Logical volume data successfully resized
查看状态、重新挂载:
- [root@bogon ~]# lvdisplay
- --- Logical volume ---
- LV Name /dev/myvg/data
- VG Name myvg
- LV UUID d0SYy1-DQ9T-Sj0R-uPeD-xn0z-raDU-g9lLeK
- LV Write Access read/write
- LV Status available
- # open 0
- LV Size 256.00 MB
- Current LE 16
- Segments 1
- Allocation inherit
- Read ahead sectors auto
- - currently set to 256
- Block device 253:2
- 重新挂载:
- [root@bogon ~]# mount /dev/myvg/data /data/
- [root@bogon data]# df /data/
- Filesystem 1K-blocks Used Available Use% Mounted on
- /dev/mapper/myvg-data
- 253900 9508 236528 4% /data
7. 扩展VG,向VG中添加一个PV:
- [root@bogon data]# pvcreate /dev/sdc2
- Physical volume "/dev/sdc2" successfully created
- [root@bogon data]# vgextend myvg /dev/sdc2
- Volume group "myvg" successfully extended
- [root@bogon data]# pvdisplay
- --- Physical volume ---
- PV Name /dev/sdc2
- VG Name myvg
- PV Size 4.67 GB / not usable 9.14 MB
- Allocatable yes
- PE Size (KByte) 16384
- Total PE 298
- Free PE 298
- Allocated PE 0
- PV UUID hrveTu-2JUH-aSgT-GKAJ-VVv2-Hit0-PyoOOr
8. 缩减VG,取出VG中的某个PV:
移除某个PV时,需要先转移该PV上数据到其他PV,然后再将该PV删除;
移出指定PV中的数据:
- [root@bogon data]# pvmove /dev/sdc2
- No data to move for myvg
- ##如果sdc2上面有数据,则会花一段时间移动,并且显示警告信息,再次确认后才会执行;
- ##如上,提示该分区中没有数据;
移除PV:
- [root@bogon data]# vgreduce myvg /dev/sdc2
- Removed "/dev/sdc2" from volume group "myvg"
- ##若发现LVM中磁盘工作不太正常,怀疑是某一块磁盘工作由问题后就可以用该方法移出问题盘
- ##上数据,然后删掉问题盘;
LVM快照:
描述:
在一个非常繁忙的服务器上,备份大量的数据时,需要停掉大量的服务,否则备份下来的数据极容易出现不一致状态,而使备份根本不能起效;这时快照就起作用了;
原理:
逻辑卷快照实质是访问原始数据的另外一个路径而已;快照保存的是做快照那一刻的数据状态;做快照以后,任何对原始数据的修改,会在修改前拷贝一份到快照区域,所以通过快照查看到的数据永远是生成快照那一刻的数据状态;但是对于快照大小有限制,做快照前需要估算在一定时间内数据修改量大小,如果在创建快照期间数据修改量大于快照大小了,数据会溢出照成快照失效崩溃;
快照不是永久的。如果你卸下LVM或重启,它们就丢失了,需要重新创建。
创建快照:
- [root@bogon ~]# lvcreate -L 500M -p r -s -n datasnap /dev/myvg/data
- Rounding up size to full physical extent 512.00 MB
- Logical volume "datasnap" created
- ## -L –l 设置大小
- ## -p :permission,设置生成快照的读写权限,默认为RW;r为只读
- ##-s 指定lvcreate生成的是一个快照
- ##-n 指定快照名称
- 挂载快照到指定位置:
- [root@bogon ~]# mount /dev/myvg/datasnap /backup/
- mount: block device /dev/myvg/datasnap is write-protected, mounting read-only
- 然后备份出快照中文件即可,备份后及时删除快照:
- [root@bogon ~]# ls /backup/
- inittab lost+found
基于RAID的LVM的建立:
描述:
基于RAID的LVM,可以在底层实现RAID对数据的冗余或是提高读写性能的基础上,可以在上层实现LVM的可以灵活管理磁盘的功能;
如图:
建立过程:
1. 建立LinuxRAID形式的分区:
- Device Boot Start End Blocks Id System
- /dev/sdb1 1 62 497983+ fd Linux raid autodetect
- /dev/sdc1 1 62 497983+ fd Linux raid autodetect
- /dev/sdd1 1 62 497983+ fd Linux raid autodetect
- /dev/sde1 1 62 497983+ fd Linux raid autodetect
2. 创建RAID-5:
- [root@nod1 ~]# mdadm -C /dev/md0 -a yes -l 5 -n 3 -x 1 /dev/sd{b,c,d,e}1
- mdadm: array /dev/md0 started.
- # RAID-5有一个spare盘,三个活动盘。
- # 查看状态,发现跟创建要求一致,且创建正在进行中:
- [root@nod1 ~]# cat /proc/mdstat
- Personalities : [raid6] [raid5] [raid4]
- md0 : active raid5 sdd1[4] sde1[3](S) sdc1[1] sdb1[0]
- 995712 blocks level 5, 64k chunk, algorithm 2 [3/2] [UU_]
- [=======>.............] recovery = 36.9% (184580/497856) finish=0.2min speed=18458K/sec
- unused devices: <none>
- # 查看RAID详细信息:
- [root@nod1 ~]# mdadm -D /dev/md0
- /dev/md0:
- Version : 0.90
- Creation Time : Wed Apr 6 04:27:46 2011
- Raid Level : raid5
- Array Size : 995712 (972.54 MiB 1019.61 MB)
- Used Dev Size : 497856 (486.27 MiB 509.80 MB)
- Raid Devices : 3
- Total Devices : 4
- Preferred Minor : 0
- Persistence : Superblock is persistent
- Update Time : Wed Apr 6 04:36:08 2011
- State : clean
- Active Devices : 3
- Working Devices : 4
- Failed Devices : 0
- Spare Devices : 1
- Layout : left-symmetric
- Chunk Size : 64K
- UUID : 5663fc8e:68f539ee:3a4040d6:ccdac92a
- Events : 0.4
- Number Major Minor RaidDevice State
- 0 8 17 0 active sync /dev/sdb1
- 1 8 33 1 active sync /dev/sdc1
- 2 8 49 2 active sync /dev/sdd1
- 3 8 65 - spare /dev/sde1
3. 建立RAID配置文件:
- [root@nod1 ~]# mdadm -Ds > /etc/mdadm.conf
- [root@nod1 ~]# cat /etc/mdadm.conf
- ARRAY /dev/md0 level=raid5 num-devices=3 metadata=0.90 spares=1 UUID=5663fc8e:68f539ee:3a4040d6:ccdac92a
4. 基于刚建立的RAID设备创建LVM:
- # 将md0创建成为PV(物理卷):
- [root@nod1 ~]# pvcreate /dev/md0
- Physical volume "/dev/md0" successfully created
- # 查看物理卷:
- [root@nod1 ~]# pvdisplay
- "/dev/md0" is a new physical volume of "972.38 MB"
- --- NEW Physical volume ---
- PV Name /dev/md0
- VG Name #此时该PV不包含在任何VG中,故为空;
- PV Size 972.38 MB
- Allocatable NO
- PE Size (KByte) 0
- Total PE 0
- Free PE 0
- Allocated PE 0
- PV UUID PUb3uj-ObES-TXsM-2oMS-exps-LPXP-jD218u
- # 创建VG
- [root@nod1 ~]# vgcreate myvg -s 32M /dev/md0
- Volume group "myvg" successfully created
- # 查看VG状态:
- [root@nod1 ~]# vgdisplay
- --- Volume group ---
- VG Name myvg
- System ID
- Format lvm2
- Metadata Areas 1
- Metadata Sequence No 1
- VG Access read/write
- VG Status resizable
- MAX LV 0
- Cur LV 0
- Open LV 0
- Max PV 0
- Cur PV 1
- Act PV 1
- VG Size 960.00 MB
- PE Size 32.00 MB
- Total PE 30
- Alloc PE / Size 0 / 0
- Free PE / Size 30 / 960.00 MB
- VG UUID 9NKEWK-7jrv-zC2x-59xg-10Ai-qA1L-cfHXDj
- # 创建LV:
- [root@nod1 ~]# lvcreate -L 500M -n mydata myvg
- Rounding up size to full physical extent 512.00 MB
- Logical volume "mydata" created
- # 查看LV状态:
- [root@nod1 ~]# lvdisplay
- --- Logical volume ---
- LV Name /dev/myvg/mydata
- VG Name myvg
- LV UUID KQQUJq-FU2C-E7lI-QJUp-xeVd-3OpA-TMgI1D
- LV Write Access read/write
- LV Status available
- # open 0
- LV Size 512.00 MB
- Current LE 16
- Segments 1
- Allocation inherit
- Read ahead sectors auto
- - currently set to 512
- Block device 253:2
问题:(实验环境,很难做出判断,望有经验的多做指教) 在RAID上实现LVM后,不能再手动控制阵列的停止。 raid设备在建成以后,是不可以改变活动磁盘数目的。只能新增备份盘, 删除故障盘。故,若对存储需求估计失误,raid上层的LVM分区即使可以扩展, 但是仍可能无法满足存储需求; 如果整个raid不能满足容量需求,在试验中是可以实现将两个RAID同时加入 一个VG,从而实现RAID间的无缝衔接~但是这样做在实际生产中效果怎么样?是 否会因为架构太复杂造成系统负过大,或是对拖慢I/O响应速度?!这仍旧是个问 题; 基于RAID之上的LVM,若划分多个LV,多个LV之间的数据分布情况怎么样?! 会不会拖慢RAID本来带来的性能提升? RAID间的无缝衔接示意图: