[TOC] ### **基本概念** 我们可能会遇到这样的情况，系统中有两块硬盘，一个100G，一个200G，但是我们需要在某个目录下，空间大小需要250G。如果按照之前的方法，先把硬盘分区，再把分区挂载到该目录下，很明显满足不了需求。此时，LVM（Logic Volume Management）技术就能够派上用场了。 首先，我们介绍三个重要的概念 * PV：物理卷（Physical Volume），一个物理卷可以由一个主分区或者逻辑分区、或者是一个没有分区的硬盘创建而来，但是不能使用扩展分区创建物理卷。 * VG：卷组（Volume Group），一个卷组由多个物理卷组成，这些物理卷可以是不同硬盘上的分区。卷组支持动态增加或删除物理卷，从而实现动态扩缩容。 * LV：逻辑卷（Logic Volume），可以从一个卷组中创建多个逻辑卷，逻辑卷可以格式化与挂载 ### **实践** ##### **查看磁盘情况** 介绍了上面三个概念，我们发现，一个100G和200G的硬盘，我们可以通过LVM技术创建一个250G逻辑卷，然后挂载到对应目录下，满足需求。 接下来，我们将通过实践来看看是如何使用LVM的。首先我们看一下前置条件：系统中有两个硬盘`/dev/sda`与`/dev/sdb`。`/dev/sda`有两个主分区`/dev/sda1`（50G）与`/dev/sda2`（787G），`/dev/sdb`有两个主分区`/dev/sdb1`（50G）与`/dev/sdb2`（173G）。我们的目标是，用`/dev/sda2`与`/dev/sdb2`创建一个卷组（787G+173G=960G），然后从卷组中创建一个大小为900G的逻辑卷，格式化该逻辑卷然后将其挂载到目录`/data2`下。 硬盘与分区情况如下： ``` [root@localhost ~]# lsblk NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT sda 8:0 1 837G 0 disk ├─sda1 8:1 1 50G 0 part / └─sda2 8:2 1 787G 0 part sdb 8:16 1 223.4G 0 disk ├─sdb1 8:17 1 50G 0 part └─sdb2 8:18 1 173.4G 0 part [root@localhost ~]# fdisk -l Disk /dev/sda: 898.7 GB, 898708275200 bytes, 1755289600 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk label type: dos Disk identifier: 0x0009613d Device Boot Start End Blocks Id System /dev/sda1 * 2048 104859647 52428800 83 Linux /dev/sda2 104859648 1755289599 825214976 83 Linux Disk /dev/sdb: 239.9 GB, 239902654464 bytes, 468559872 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk label type: dos Disk identifier: 0x93de47a2 Device Boot Start End Blocks Id System /dev/sdb1 2048 104859647 52428800 83 Linux /dev/sdb2 104859648 468559871 181850112 83 Linux ``` ##### **创建物理卷** 接下来，我们用分区`/dev/sda2`与`/dev/sdb2`创建对应的两个物理卷，然后pvdisplay一下发现多了两个pv ``` [root@localhost ~]# pvcreate /dev/sda2 Physical volume "/dev/sda2" successfully created. [root@localhost ~]# pvcreate /dev/sdb2 Physical volume "/dev/sdb2" successfully created. [root@localhost ~]# pvdisplay "/dev/sdb2" is a new physical volume of "<173.43 GiB" --- NEW Physical volume --- PV Name /dev/sdb2 VG Name PV Size <173.43 GiB Allocatable NO PE Size 0 Total PE 0 Free PE 0 Allocated PE 0 PV UUID mLapVo-tZSK-mpVY-ye4T-rf7w-kBUO-PTqsDd "/dev/sda2" is a new physical volume of "<786.99 GiB" --- NEW Physical volume --- PV Name /dev/sda2 VG Name PV Size <786.99 GiB Allocatable NO PE Size 0 Total PE 0 Free PE 0 Allocated PE 0 PV UUID kle4Ia-DdOY-zG40-8CqP-gqhw-kwLP-hOABne ``` ##### **创建卷组** 然后利用这两个物理卷创建一个卷组，可以看出，这个卷组的大小为960.41G ``` [root@localhost ~]# vgcreate myvg /dev/sda2 /dev/sdb2 Volume group "myvg" successfully created [root@localhost ~]# vgdisplay --- Volume group --- VG Name myvg System ID Format lvm2 Metadata Areas 2 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 2 Act PV 2 VG Size <960.41 GiB PE Size 4.00 MiB Total PE 245864 Alloc PE / Size 0 / 0 Free PE / Size 245864 / <960.41 GiB VG UUID 86QgV0-uxvo-nA8F-x2ae-HR2Q-GnXd-fut39o ``` ##### **创建逻辑卷** 接下来，我们在卷组中创建一个大小为900G的逻辑卷 ``` [root@localhost ~]# lvcreate -n mylv --size 900G myvg Logical volume "mylv" created. [root@localhost ~]# lvdisplay --- Logical volume --- LV Path /dev/myvg/mylv LV Name mylv VG Name myvg LV UUID 00dplX-uqHr-IpzA-tA6T-hf3d-TFnA-c5cnuk LV Write Access read/write LV Creation host, time localhost.localdomain, 2018-11-30 17:08:52 +0800 LV Status available # open 0 LV Size 900.00 GiB Current LE 230400 Segments 2 Allocation inherit Read ahead sectors auto - currently set to 256 Block device 253:0 ``` 当然，创建逻辑卷时，也可以用下面的方式指定逻辑卷的大小，见参考文献： ``` $ lvcreate -n mylv -l 50%FREE myvg # 大小为myvg中FREE空间的50% $ lvcreate -n mylv -l 50%VG myvg # 大小为myvg总空间的50% ``` ##### **格式化与挂载** 然后格式化逻辑卷，并挂载到`/data2`目录下 ``` [root@localhost ~]# mkfs.xfs -f -i attr=2 -l lazy-count=1,sectsize=4096 -b size=4096 -d sectsize=4096 -n ftype=1 /dev/myvg/mylv meta-data=/dev/myvg/mylv isize=512 agcount=4, agsize=58982400 blks = sectsz=4096 attr=2, projid32bit=1 = crc=1 finobt=0, sparse=0 data = bsize=4096 blocks=235929600, imaxpct=25 = sunit=0 swidth=0 blks naming =version 2 bsize=4096 ascii-ci=0 ftype=1 log =internal log bsize=4096 blocks=115200, version=2 = sectsz=4096 sunit=1 blks, lazy-count=1 realtime =none extsz=4096 blocks=0, rtextents=0 [root@localhost ~]# mkdir /data2 [root@localhost ~]# mount /dev/myvg/mylv /data2 ``` 接下来，我们来看一下分区与挂载的情况，发现/data2目录的大小为900G，设备的目录需要注意下，为`/dev/mapper/myvg-mylv` ``` [root@localhost ~]# df -hT Filesystem Type Size Used Avail Use% Mounted on /dev/sda1 xfs 50G 7.8G 43G 16% / devtmpfs devtmpfs 63G 0 63G 0% /dev tmpfs tmpfs 63G 0 63G 0% /dev/shm tmpfs tmpfs 63G 18M 63G 1% /run tmpfs tmpfs 63G 0 63G 0% /sys/fs/cgroup tmpfs tmpfs 13G 40K 13G 1% /run/user/0 /dev/mapper/myvg-mylv xfs 900G 33M 900G 1% /data2 ``` 我们看一下lsblk与fdisk命令的输出，发现lsblk中的sda2与sdb2下多了myvg-mylv，且类型为lvm，这说明myvg-mylv是一个逻辑卷，它所属的卷组所sda2与sdb2两个物理组成。另外，fdisk命令中并没有显示这个逻辑卷，看来fdisk只会显示分区的内容 ``` [root@localhost ~]# lsblk NAME MAJ:MIN RM SIZE RO TYPE MOUNTPOINT sda 8:0 1 837G 0 disk ├─sda1 8:1 1 50G 0 part / └─sda2 8:2 1 787G 0 part └─myvg-mylv 253:0 0 900G 0 lvm /data2 sdb 8:16 1 223.4G 0 disk ├─sdb1 8:17 1 50G 0 part └─sdb2 8:18 1 173.4G 0 part └─myvg-mylv 253:0 0 900G 0 lvm /data2 [root@localhost ~]# fdisk -l Disk /dev/sda: 898.7 GB, 898708275200 bytes, 1755289600 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk label type: dos Disk identifier: 0x0009613d Device Boot Start End Blocks Id System /dev/sda1 * 2048 104859647 52428800 83 Linux /dev/sda2 104859648 1755289599 825214976 83 Linux Disk /dev/sdb: 239.9 GB, 239902654464 bytes, 468559872 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes Disk label type: dos Disk identifier: 0x93de47a2 Device Boot Start End Blocks Id System /dev/sdb1 2048 104859647 52428800 83 Linux /dev/sdb2 104859648 468559871 181850112 83 Linux Disk /dev/mapper/myvg-mylv: 966.4 GB, 966367641600 bytes, 1887436800 sectors Units = sectors of 1 * 512 = 512 bytes Sector size (logical/physical): 512 bytes / 512 bytes I/O size (minimum/optimal): 512 bytes / 512 bytes ``` ##### **删除操作** 接下来，我们来删除这些逻辑卷、卷组和物理卷。首先先卸载逻辑卷 ``` umount /dev/mapper/myvg-mylv ``` 然后删除逻辑卷`/dev/myvg/mylv`，再lvdisplay一下发现lv没有了 ``` [root@localhost ~]# lvremove /dev/myvg/mylv Do you really want to remove active logical volume myvg/mylv? [y/n]: y Logical volume "mylv" successfully removed [root@localhost ~]# lvdisplay [root@localhost ~]# ``` 接着删除卷组myvg，再vgdisplay一下发现vg没有了 ``` [root@localhost ~]# vgremove myvg Volume group "myvg" successfully removed [root@localhost ~]# vgdisplay [root@localhost ~]# ``` 接着可以删除物理卷`/dev/sda2`和`/dev/sdb2`，再pvdisplay一下发现不存在pv了 ``` [root@localhost ~]# pvremove /dev/sda2 Labels on physical volume "/dev/sda2" successfully wiped. [root@localhost ~]# pvremove /dev/sdb2 Labels on physical volume "/dev/sdb2" successfully wiped. [root@localhost ~]# pvdisplay [root@localhost ~]# ``` ### **Reference** * https://www.tecmint.com/add-new-disks-using-lvm-to-linux/ * http://blog.51cto.com/11093860/2164405 * https://web.mit.edu/rhel-doc/5/RHEL-5-manual/Cluster_Logical_Volume_Manager/LV_create.html