I. Concept and Origin

LVM: Logical Volume Manager

Ordinary disk partition management can not change its size after logical partition is well divided. When a logical partition can not store a file, the file can not be stored across multiple partitions because of the limitation of the upper file system, so it can not be placed on multiple disks.

When a partition space is exhausted, the solution is usually to use symbolic links or tools to adjust the partition size (tools may cause our system to crash), but this has not fundamentally solved the problem.

With the emergence of LVM, the problem is solved. Users can easily adjust the size of partitions without downtime.

II. Working Principle

LVM works by abstractly encapsulating the underlying physical hard disk, and then presenting it to the upper application in the form of logical volumes.

In the traditional disk management mechanism, the upper application accesses the file system directly to read and write the underlying physical hard disk. In LVM, it encapsulates the underlying hard disk. When we operate the underlying physical disk through software, it no longer operates directly on partitions, but on the underlying physical hard disk through logical volumes.

The greatest feature of LVM is that it can dynamically manage disks, because the size of logical volumes can be dynamically adjusted, and data will not be lost. Logical volume technology greatly improves the logic of disk management.

3. Noun Interpretation

The physical media: This refers to the storage device of the system: hard disk, such as: / dev/hda, / dev/sda, etc., is the storage unit of the lowest layer of the storage system.

Physical Volume: Physical Volume refers to a hard disk partition or a device (such as RAID) that logically has the same function as a disk partition. It is the basic storage logic block of LVM, but compared with the basic physical storage media (such as partition, disk, etc.), it contains management parameters related to LVM.

Volume Group: A LVM volume group is similar to a physical hard disk in a non-LVM system and consists of physical volumes. You can create one or more LVM partitions (logical volumes) on a volume group, which consists of one or more physical volumes.

Logical Volume: Logical Volume of LVM is similar to hard disk partitioning in non-LVM systems, on which file systems (such as / home or / usr, etc.) can be built.

PE (Physical Extent): Each physical volume is divided into basic units called PE(Physical Extents). PE with a unique number is the smallest unit that can be addressed by LVM. The size of PE is configurable and defaults to 4MB.

LE (Logical Extent): Logical volumes are also divided into addressable basic units called LE(Logical Extents). In the same volume group, LE size and PE are the same and one to one.

IV. LVM Creates Logical Concepts

We use LVM to dynamically manage the disks, so that the hard disks are presented to the upper services in the form of logical volumes. So the purpose of our operation is to create LV, replace our previous partition with LV, we format the logical volume, and then mount it to use. The steps are as follows:

 

1. After formatting the created LV file system and mounting it with mount, we can use logical volumes as usual as partitions.
2. The working principle of LVM is summarized.

(a) Physical disks are formatted into PV s, and space is divided into PE's.

(b) Adding different PVs to the same VG, and putting all the PE of different PVs into the PE pool of the VG

LV is created based on PE, which is several times the size of PE. The PE that makes up LV may come from different disks.

(d) LV can be mounted after formatting the file system.

(e) Extension and reduction of LV is actually to increase or decrease the number of PE, and the process will not lose data.

So, if we want to expand LV, we first add a disk, then format it into PV, add the PV to VG, and then we can take out the free PE from VG to expand LV.

 

V. Order parsing

(a) PV management

pvcreate Create PV

pvscan Scan and list all PV s

pvdisplay Lists PV attribute information

pvremove Remove PV

Data in pvmove Mobile PV

　　　

(b) VG management

vgcreate Create VG

- s # Specifies the size of the PE ___________

- l) Maximum number of logical volumes allowed to be created on # volume groups

- p) Maximum number of physical volumes allowed to be added in volume groups

Scanning and listing all VG s

List VG attribute information in vgdisplay

vgremove Remove VG

vgreduce Remove PV from VG

vgextend # Add PV to VG

vgchange # Sets whether the VG is active

　　　　

(c) LV management

lvcreate Create LV

- L + size # Specify LV size

- n # Specify the LV name

Example: vg0-lv1 will be created in lvcreate-L 5G-n LV vg0 and/dev/mapper

lvscan Scan and list all LV s

lvdisplay # lists LV attribute information

lvremove Remove lv

lvreduce(lvresize) Reduce lv capacity

lvextend(lvresize) Increase lv capacity

lvresize Cover lv Capacity

　　　　　　

6. Creating Logical Volumes

1. Format the physical disk into PV using the pvcreate command, and use the pvcreate command.

Virtual two 2G physical hard disks, which can be viewed by fdisk-l command

(b) First format / dev/sdb, / dev/sdc as PV

(c) Use pvs to view simple information and pvdisplay to view detailed information

 

 

2. Create volume group VG and add PV to VG through vgcreate command

3. Create LV based on VG through lvcreate command

[root@centos7 ~]# lvcreate -n xiaoliziLV -L 2G pool　　　　　　　　　　　　# Create LV logical volume name xiaoliziLV size 2G
  Logical volume "xiaoliziLV" created.
[root@centos7 ~]# lvs　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　# Simple display of LV information
  LV         VG   Attr       LSize Pool Origin Data%  Meta%  Move Log Cpy%Sync Convert
  xiaoliziLV pool -wi-a----- 2.00g  
[root@centos7 ~]# lvdisplay 　　　　　　　　　　　　　　　　　　　　　　　　　　# Detailed display of LV information
  --- Logical volume ---　　
  LV Path                /dev/pool/xiaoliziLV
  LV Name                xiaoliziLV
  VG Name                pool
  LV UUID                xhCdbm-JZvA-MIMW-4p0K-0V4O-Gl60-9X4mNr
  LV Write Access        read/write
  LV Creation host, time centos7, 2019-04-19 17:40:39 +0800
  LV Status              available
  # open                 0
  LV Size                2.00 GiB
  Current LE             512
  Segments               2
  Allocation             inherit
  Read ahead sectors     auto
  - currently set to     8192
  Block device           253:0

　　[root@centos7 ~]# ls /dev/mapper/
　　　　control pool-xiaoliziLV

　　[root@centos7 ~]# mkfs.xfs /dev/mapper/pool-xiaoliziLV 　　　　　　　　　# Format file system
　　　　meta-data=/dev/mapper/pool-xiaoliziLV isize=512 agcount=4, agsize=131072 blks
　　　　= sectsz=512 attr=2, projid32bit=1
　　　　= crc=1 finobt=0, sparse=0
　　　　data = bsize=4096 blocks=524288, imaxpct=25
　　　　= sunit=0 swidth=0 blks
　　　　naming =version 2 bsize=4096 ascii-ci=0 ftype=1
　　　　log =internal log bsize=4096 blocks=2560, version=2
　　　　= sectsz=512 sunit=0 blks, lazy-count=1
　　　　realtime =none extsz=4096 blocks=0, rtextents=0

　[root@centos7 ~]# mkdir /lvtest　　　　　　　　　　　　　　　　　　　　　　　　# Create mount points
　[root@centos7 ~]# mount /dev/mapper/pool-xiaoliziLV /lvtest/
　[root@centos7 ~]# df -h
　　　　Filesystem Size Used Avail Use% Mounted on
　　　　/dev/sda3 98G 1.6G 97G 2% /
　　　　devtmpfs 901M 0 901M 0% /dev
　　　　tmpfs 911M 0 911M 0% /dev/shm
　　　　tmpfs 911M 9.6M 902M 2% /run
　　　　tmpfs 911M 0 911M 0% /sys/fs/cgroup
　　　　/dev/sda1 1014M 135M 880M 14% /boot
　　　　tmpfs 183M 0 183M 0% /run/user/0
　　　　/dev/mapper/pool-xiaoliziLV 2.0G 33M 2.0G 2% /lvtest

 

7. Expanding a Logical Volume

In the traditional hard disk management mode, if one partition size is not enough (when the hard disk still has space), we need to expand, then we can only transfer all the information of this partition to other places, re-partition after uninstalling, and finally upload the new partition.

If in the production environment, such operation is unimaginable, time-consuming and laborious. The use of LVM is much more convenient, as shown in the figure:

 

 

As can be seen from the figure above, the so-called expansion is to increase the number of PE in LV, the increase depends on the number of PE in your VG.

[Note] Logical volume expansion can be done online without uninstalling the logical volume. This kind of management is very necessary for companies that provide services to the outside world in 7*24 hours.

In the first case, there is enough space left in the VG (1.99G free can be seen), which increases the capacity of the LV by 1G. The steps are as follows:

[root@centos7 ~]# vgdisplay 
  --- Volume group ---
  VG Name               pool
  System ID             
  Format                lvm2
  Metadata Areas        2
  Metadata Sequence No  2
  VG Access             read/write
  VG Status             resizable
  MAX LV                0
  Cur LV                1
  Open LV               1
  Max PV                0
  Cur PV                2
  Act PV                2
  VG Size               3.99 GiB
  PE Size               4.00 MiB
  Total PE              1022
  Alloc PE / Size       512 / 2.00 GiB
  Free  PE / Size       510 / 1.99 GiB
  VG UUID               e55Yks-0Phr-VarP-xAF8-8W4B-vBFg-VWqPm3
   
[root@centos7 ~]# lvextend -L +1G /dev/pool/xiaoliziLV 　　　　　　　　　　　　　　# Expanded Logic Volume 1 G，Use lvextend command

  Size of logical volume pool/xiaoliziLV changed from 2.00 GiB (512 extents) to 3.00 GiB (768 
extents).
  Logical volume pool/xiaoliziLV successfully resized.
[root@centos7 ~]# lvdisplay 　　　　　　　　　　　　　　　　　　　　　　　　　　　　　　# View Logical Volume Information (Discovery) LV It's 3. G )

--- Logical volume ---
  LV Path                /dev/pool/xiaoliziLV
  LV Name                xiaoliziLV
  VG Name                pool
  LV UUID                xhCdbm-JZvA-MIMW-4p0K-0V4O-Gl60-9X4mNr
  LV Write Access        read/write
  LV Creation host, time centos7, 2019-04-19 17:40:39 +0800
  LV Status              available
  # open                 1
  LV Size                3.00 GiB
  Current LE             768
  Segments               2
  Allocation             inherit
  Read ahead sectors     auto
  - currently set to     8192
  Block device           253:0


[root@centos7 ~]# df -hT
Filesystem                  Type      Size  Used Avail Use% Mounted on
/dev/sda3                   xfs        98G  1.6G   97G   2% /
devtmpfs                    devtmpfs  901M     0  901M   0% /dev
tmpfs                       tmpfs     911M     0  911M   0% /dev/shm
tmpfs                       tmpfs     911M  9.6M  902M   2% /run
tmpfs                       tmpfs     911M     0  911M   0% /sys/fs/cgroup
/dev/sda1                   xfs      1014M  135M  880M  14% /boot
tmpfs                       tmpfs     183M     0  183M   0% /run/user/0
/dev/mapper/pool-xiaoliziLV xfs       2.0G   33M  2.0G   2% /lvtest

#  As we can see above, it's just LV The size of the real file system remains unchanged. That's because we're creating it. LV After that, it will be formatted into a file system immediately. After that, the modification of the logical volume will not directly affect the file system. An updated command will be used at this time. resize2fs/xfs_growfs

[root@centos7 ~]# resize2fs /dev/mapper/pool-xiaoliziLV 　　　　　　
　　resize2fs 1.42.9 (28-Dec-2013)
　　resize2fs: Bad magic number in super-block while trying to open /dev/mapper/pool-xiaoliziLV　　　　　　# This error is due to the previous filesystem. xfs Yes, so the extension command here is xfs_growfs
　　Couldn't find valid filesystem superblock.

[root@centos7 ~]# xfs_growfs /dev/mapper/pool-xiaoliziLV　　　　　　　　　　　　　　　　　　　　　　　　　　　　　# Update file system
meta-data=/dev/mapper/pool-xiaoliziLV isize=512 agcount=4, agsize=131072 blks
= sectsz=512 attr=2, projid32bit=1
= crc=1 finobt=0 spinodes=0
data = bsize=4096 blocks=524288, imaxpct=25
= sunit=0 swidth=0 blks
naming =version 2 bsize=4096 ascii-ci=0 ftype=1
log =internal bsize=4096 blocks=2560, version=2
= sectsz=512 sunit=0 blks, lazy-count=1
realtime =none extsz=4096 blocks=0, rtextents=0
data blocks changed from 524288 to 786432

[root@centos7 ~]# df -h　　　　　　　　　　　　　　　　　　# Capacity expansion
Filesystem Size Used Avail Use% Mounted on
/dev/sda3 98G 1.6G 97G 2% /
devtmpfs 901M 0 901M 0% /dev
tmpfs 911M 0 911M 0% /dev/shm
tmpfs 911M 9.6M 902M 2% /run
tmpfs 911M 0 911M 0% /sys/fs/cgroup
/dev/sda1 1014M 135M 880M 14% /boot
tmpfs 183M 0 183M 0% /run/user/0
/dev/mapper/pool-xiaoliziLV 3.0G 33M 3.0G 2% /lvtest

In the second case, there is not enough space left in the VG (1 G can be seen idle), which increases the capacity of the LV by 16 G. The steps are as follows:

7. Reducing a Logic Volume

Reducing logical volumes is actually replacing a specified number of PE from LV back to VG.

 

[Note] The shrinking operation of logical volumes must first uninstall the logical volumes, otherwise the files in the logical volumes may be damaged. At this time, we will reduce the LV to 10G, the steps are as follows:

1.adopt umount Unload mounted logical volumes

[root@onion ~]# umount /mnt/
[root@onion ~]# df -hT
Filesystem     Type   Size  Used Avail Use% Mounted on
/dev/sda3      ext4    48G  1.5G   44G   4% /
tmpfs          tmpfs  491M     0  491M   0% /dev/shm
/dev/sda1      ext4   190M   35M  146M  19% /boot

2.narrow LV Size: As we mentioned earlier, yes. LV After expansion, it needs to be implemented. resize2fs/xfs_growfs Update our file system, but on LV When performing scaling-down operations(xfs File system does not support scaling)，Reduction must be performed first

[root@onion ~]# resize2fs /dev/pool/mylv 10G   # Check the file system or e2fsck-f/dev/pool/mylv
resize2fs 1.41.12 (17-May-2010)
Please run 'e2fsck -f /dev/pool/mylv' first.
[root@onion ~]# e2fsck -f /dev/pool/mylv
e2fsck 1.41.12 (17-May-2010)
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
/dev/pool/mylv: 12/1245184 files (0.0% non-contiguous), 95937/4980736 blocks

3.After checking, use it again resize2fs To shrink the file system

[root@onion ~]# resize2fs /dev/pool/mylv 10G
resize2fs 1.41.12 (17-May-2010)
Resizing the filesystem on /dev/pool/mylv to 2621440 (4k) blocks.
The filesystem on /dev/pool/mylv is now 2621440 blocks long.

4.Contrary to the order of expansion, we first narrowed down the file system, then narrowed down. LV，Use lvreduce command

[root@onion ~]# lvreduce -L -9G /dev/pool/mylv 
  WARNING: Reducing active logical volume to 10.00 GiB.
  THIS MAY DESTROY YOUR DATA (filesystem etc.)
Do you really want to reduce pool/mylv? [y/n]: y

Note: Here's a warning that our files may be damaged. This will happen when we need to reduce the file size and actual operation.( resize2fs)When the size of the file is not the same, for example, if I shrink one more file here G The size of the logical volume, this time there will be file corruption
  Size of logical volume pool/mylv changed from 19.00 GiB (4864 extents) to 10.00 GiB (2560 extents).
  Logical volume mylv successfully resized.

[root@onion ~]# lvs
  LV   VG   Attr       LSize  Pool Origin Data%  Meta%  Move Log Cpy%Sync Convert
  mylv pool -wi-a----- 10.00g

[Note: Reducing logical volumes is a very dangerous operation. A little negligence may damage our logical volumes, causing the loss or damage of files, so we generally do not want to reduce logical volumes.

5.After shrinking, mount the logical volume on/mnt Next, see if the file inside exists.

 

VIII. Reducing Volume Group

When we have enough space in VG and a lot of spare space, we want to take down the hard disk which is not used in VG for its use. At this point, we need to perform the operation of reducing volume groups.

　　

1.use first pvdisplay View each PV You can see the information/dev/sdd Space is unused
[root@onion ~]# pvdisplay 
  --- Physical volume ---
  PV Name               /dev/sdb
  VG Name               pool
  PV Size               8.00 GiB / not usable 4.00 MiB
  Allocatable           yes (but full)
  PE Size               4.00 MiB
  Total PE              2047
  Free PE               0
  Allocated PE          2047
  PV UUID               SfW2Wo-NeZM-KXSX-UqQj-VWHV-NAL6-UQb3lw
   
  --- Physical volume ---
  PV Name               /dev/sdc
  VG Name               pool
  PV Size               8.00 GiB / not usable 4.00 MiB
  Allocatable           yes 
  PE Size               4.00 MiB
  Total PE              2047
  Free PE               1534
  Allocated PE          513
  PV UUID               F8Gq56-M9ff-i9e9-TEvF-1BOL-qsiJ-PTuApu
   
  --- Physical volume ---
  PV Name               /dev/sdd
  VG Name               pool
  PV Size               8.00 GiB / not usable 4.00 MiB
  Allocatable           yes 
  PE Size               4.00 MiB
  Total PE              2047
  Free PE               2047
  Allocated PE          0
  PV UUID               IT2wzW-bVBQ-NU7C-9EbU-3uie-W1LF-Vx3clj

2.Whether to uninstall the logical volume after determining the disk

[root@onion ~]# umount /mnt/
[root@onion ~]# df -hT
Filesystem     Type   Size  Used Avail Use% Mounted on
/dev/sda3      ext4    48G  1.5G   44G   4% /
tmpfs          tmpfs  491M     0  491M   0% /dev/shm
/dev/sda1      ext4   190M   35M  146M  19% /boot

3.Take this PV Remove and view from the specified volume.

[root@onion ~]# vgreduce pool /dev/sdd 
  Removed "/dev/sdd" from volume group "pool"
[root@onion ~]# vgdisplay 
  --- Volume group ---
  VG Name               pool
  System ID             
  Format                lvm2
  Metadata Areas        2
  Metadata Sequence No  9
  VG Access             read/write
  VG Status             resizable
  MAX LV                0
  Cur LV                1
  Open LV               0
  Max PV                0
  Cur PV                2
  Act PV                2
  VG Size               15.99 GiB
  PE Size               4.00 MiB
  Total PE              4094
  Alloc PE / Size       2560 / 10.00 GiB
  Free  PE / Size       1534 / 5.99 GiB
  VG UUID               YUlucV-5pVP-W73A-DAsR-EIlw-jnVf-l5Y9M1

4.See PV The situation (found or found)/dev/sdd)

[root@onion ~]# pvs
  PV         VG   Fmt  Attr PSize PFree
  /dev/sdb   pool lvm2 a--u 8.00g    0 
  /dev/sdc   pool lvm2 a--u 8.00g 5.99g
  /dev/sdd        lvm2 ---- 8.00g 8.00g

5.remove/dev/sdd this PV

[root@onion ~]# pvremove /dev/sdd 
  Labels on physical volume "/dev/sdd" successfully wiped
[root@onion ~]# pvs
  PV         VG   Fmt  Attr PSize PFree
  /dev/sdb   pool lvm2 a--u 8.00g    0 
  /dev/sdc   pool lvm2 a--u 8.00g 5.99g

6.Remount to see if the file has changed

[root@onion ~]# mount /dev/pool/mylv /mnt
[root@onion ~]# ll /mnt/
total 16
-rw-r--r--. 1 root root     0 Jul 15 10:56 koala
drwx------. 2 root root 16384 Jul 15 10:46 lost+found

9. Delete Logical Volume

For unwanted logical volumes, you can delete them.

[Note] The creation of LV s is sequential and the deletion is strictly sequential.

 

Uninstall - > Delete Logical Volumes - > Delete Volume Groups - > Delete Physical Volumes

1. Unload the logical volume using the umount command
2. Delete logical volumes using the lvremove command
3. Delete volume groups, using the vgremove command
4. Delete the physical volume, using the pvremove command