What is LVM
Logical volume manager (LVM) introduces an extra layer between the physical disks and the file system allowing file systems to be :
– resized and moved easily and online without requiring a system-wide outage.
– Using discontinuous space on disk
– meaningful names to volumes, rather than the usual cryptic device names.
– span multiple physical disks
Linux LVM is very similar to HP-UX LVM and provides many other advanced features like snapshots, cluster support (GFS2, OCFS and Lustre).
LVM comprises of few conceptual layers such as physical volume, logical volume and file systems.
The conceptual layers are in turn made up of smaller units like Physical extents(in case of Physical volumes) and Logical extents (in case of Logical Volumes).
Physical Volume (PV)
Each Physical Volume can be a disk partition, whole disk, meta-device, or a loopback file. Use the command pvcreate to initialize storage for use by LVM. Initializing a block device as physical volume places a label at the start of the device.
Volume Group (VG)
A Volume Group gathers together a collection of Logical Volumes and Physical Volumes into one administrative unit. Volume group is divided into fixed size physical extents. The command vgcreate creates a new volume group using the block special device Physical Volume path previously configured for LVM with pvcreate.
– VGs are made up of PVs, which in turn are made up of physical extents (PEs). The size of PE can differe in different VGs and is defined at the time of creating VG.
– The default size of PE is 4MB, but you can change it to the value you want at the time of VG creation.
– Generally, larger the PE size, better the performance (though less granular control of LV).
Logical Volume (LV)
A Logical Volume is the conceptual equivalent of a disk partition in a non-LVM system. Logical volumes are block devices which are created from the physical extents present in the same volume group. You can use command lvcreate to create a logical volume in an existing volume group.
File systems are built on top of logical volumes. The command mkfs can be used to create file system on top of a logical volume. Once the file system is created we can mount the logical volume as per our need.
Lets Get Started
In the example below we would :
1. Create 3 Physical volumes from 3 physical disks (/dev/sdb, /dev/sdc, /dev/sdd).
2. Create Volume group from these 3 PVs (/dev/vg01).
3. Create a Lgical Volume in this VG (/dev/vg01/lvol01).
4. Create a File system on this LV and mount it (/data01).
Create Physical Volumes
The pvcreate command is used to initialize the PV for use by LVM. Before creating the PV, make sure the disk is visible in the OS. To scan the block devices to be used as PVs, use the lvmdiskscan command.
# lvmdiskscan ....... /dev/sdb [ 2.00 GiB] /dev/sdc [ 2.00 GiB] /dev/sdd [ 2.00 GiB] 3 disks 19 partitions 0 LVM physical volume whole disks 0 LVM physical volumes
Initialize the block devices :
# pvcreate /dev/sdb /dev/sdc /dev/sdd Physical volume "/dev/sdb" successfully created Physical volume "/dev/sdc" successfully created Physical volume "/dev/sdd" successfully created
Display physical volumes
Use the commands pvdisplay, pvs and pvscan to display the PVs we just created.
# pvdisplay "/dev/sdb" is a new physical volume of "2.00 GiB" --- NEW Physical volume --- PV Name /dev/sdb VG Name PV Size 2.00 GiB Allocatable NO PE Size 0 Total PE 0 Free PE 0 Allocated PE 0 PV UUID Mt3F7z-a2AV-28Vn-uXe2-QejE-Z6tP-UMlQGM "/dev/sdc" is a new physical volume of "2.00 GiB" --- NEW Physical volume --- PV Name /dev/sdc VG Name PV Size 2.00 GiB Allocatable NO PE Size 0 Total PE 0 Free PE 0 Allocated PE 0 PV UUID 5m1Fuc-yTRn-I2vG-bMfU-6SE7-53EA-s8VQjt "/dev/sdd" is a new physical volume of "2.00 GiB" --- NEW Physical volume --- PV Name /dev/sdd VG Name PV Size 2.00 GiB Allocatable NO PE Size 0 Total PE 0 Free PE 0 Allocated PE 0 PV UUID 1x3e2A-C0Lt-DrUA-tPSM-lsMu-sn70-qg1j8p
# pvscan PV /dev/sdb lvm2 [2.00 GiB] PV /dev/sdc lvm2 [2.00 GiB] PV /dev/sdd lvm2 [2.00 GiB] Total: 3 [6.00 GiB] / in use: 0 [0 ] / in no VG: 3 [6.00 GiB]
# pvs PV VG Fmt Attr PSize PFree /dev/sdb lvm2 a-- 2.00g 2.00g /dev/sdc lvm2 a-- 2.00g 2.00g /dev/sdd lvm2 a-- 2.00g 2.00g
Create a Volume Group
Use the vgcreate command to create the new Volume Group vg01 using the 3 PVs we just created. We can specify the extents with -s option and maximum number of PVs and LVs in the VG by using the options -p and -l respectively. All these option are optional and need not be necessarily used.
# vgcreate vg01 /dev/sdb /dev/sdc /dev/sdd Volume group "vg01" successfully created
The optional options that are used with vgcreate command are :
|-s||Physical extent size|
|-p||Max number of PVs|
|-l||Max number of LVs|
|–alloc||allocation policy (either contiguous, anywhere, or cling)|
Displaying the VG information
The commands vgs and vgdisplay can be used to display the information about the VG we just created :
# vgs vg01 VG #PV #LV #SN Attr VSize VFree vg01 3 0 0 wz--n- 5.99g 5.99g
# vgdisplay vg01 --- Volume group --- VG Name vg01 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 5.99 GiB PE Size 4.00 MiB Total PE 1533 Alloc PE / Size 0 / 0 Free PE / Size 1533 / 5.99 GiB VG UUID Cw7GGz-NH3o-Sax2-5jPv-buZS-938T-tmNKFa
Activating and deactivating VGs
The vgchange command can be used to activate/deactivate a volume group.
To deactivate a VG :
# vgchange -a n vg01 0 logical volume(s) in volume group "vg01" now active
To activate a VG :
# vgchange -a y vg01 1 logical volume(s) in volume group "vg01" now active
Create Logical Volume
The Logical volume can now be created in the VG using the lvcreate command.
– If you do not specify the LV name in the command, by default the LV is given the name lvol#.
– Normally if you do not specify which PV to span the LV, Logical volume will be created on the PV on a next-free basis.
– To create a logical volume lvol01 of size 5 GB :
# lvcreate -L 5G -n lvol01 vg01 Logical volume "lvol01" created
Creating a striped volume
To create a striped volume spanning all the 3 PVs we created :
# lvcreate -L 5G -I 4096 -i 3 -n lvol01 vg01 Rounding size (1280 extents) up to stripe boundary size (1281 extents) Logical volume "lvol01" created
I - PVs to span while creating striped volume i - stripe unit
Creating mirrored volume
To create a 3 way mirrored volume spanning the 3 PVs (sdb, sdc, sdd) :
# lvcreate -L 1G -m 2 -n lvol01 vg01 Logical volume "lvol01" created
We can also specify which devices to be used while creating the mirrored LV. In our case as we had only 3 PVs in the VG, the LV gets created by default on these 3 PVs.
Displaying the LV information
The commands lvdisplay, lvs and lvscan can be used to display the information about the LV we just created.
# lvs /dev/vg01/lvol01 LV VG Attr LSize Pool Origin Data% Move Log Cpy%Sync Convert lvol01 vg01 mwi-a-m-- 1.00g lvol01_mlog 100.00
# lvdisplay /dev/vg01/lvol01 --- Logical volume --- LV Path /dev/vg01/lvol01 LV Name lvol01 VG Name vg01 LV UUID ptlmAV-mO42-fWiJ-e2Ml-r9kj-PFcC-MOexxw LV Write Access read/write LV Creation host, time localhost.localdomain, 2014-10-22 09:04:25 -0700 LV Status available # open 0 LV Size 1.00 GiB Current LE 256 Mirrored volumes 3 Segments 1 Allocation inherit Read ahead sectors auto - currently set to 256 Block device 253:4
# lvscan ACTIVE '/dev/vg01/lvol01' [1.00 GiB] inherit
Creating File system
The final step is to create a file system on the new LV we just created and mount it on a directory to be able to access it and store data in it. The command mkfs can be used to create file system on top of the LV.
# mkfs.ext4 /dev/vg01/lvol01 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 65536 inodes, 262144 blocks 13107 blocks (5.00%) reserved for the super user First data block=0 Maximum filesystem blocks=268435456 8 block groups 32768 blocks per group, 32768 fragments per group 8192 inodes per group Superblock backups stored on blocks: 32768, 98304, 163840, 229376 Writing inode tables: done Creating journal (8192 blocks): done Writing superblocks and filesystem accounting information: done This filesystem will be automatically checked every 37 mounts or 180 days, whichever comes first. Use tune2fs -c or -i to override.
The logical volume can be mounted, once the file system is created. Make sure to add an entry to /etc/fstab, so that it is mounted automatically when the system boots.
# mkdir /data01 # mount /dev/vg01/lvol01 /data01
# vi /etc/fstab /dev/vg01/lvol01 /data01 ext4 defaults 0 0
# df -h /data01 Filesystem Size Used Avail Use% Mounted on /dev/mapper/vg01-lvol01 1008M 34M 924M 4% /data01
The Graphical Tool to manage LVM
There is a cool graphical tool available (system-config-lvm) in case you want to use. If not already installed on the system, install it using yum:
# yum install system-config-lvm
To start the Graphical LVM administration tool, fire the command :