Red Hat System Administration II

Objectives

After completing this section, you should be able to:

Extending and Reducing a Volume Group

You can add more disk space to a volume group by adding additional physical volumes. This is called extending the volume group. Then, you can assign the new physical extents from the additional physical volumes to logical volumes.

You can remove unused physical volumes from a volume group. This is called reducing the volume group. First, use the pvmove command to move data from extents on one physical volume to extents on other physical volumes in the volume group. In this way, a new disk can be added to an existing volume group, data can be moved from an older or slower disk to a new disk, and the old disk removed from the volume group. You can perform these actions while the logical volumes in the volume group are in use.

Extending a Volume Group

To extend a volume group, perform the following steps:

Prepare the physical device and create the physical volume.

As with creating a new volume group, you must create and prepare a new partition for use as a physical volume if there are none prepared already.

[root@host ~]# parted -s /dev/vdb mkpart primary 1027MiB 1539MiB
[root@host ~]# parted -s /dev/vdb set 3 lvm on
[root@host ~]# pvcreate /dev/vdb3

A PV only needs to be created if there are no PVs free to extend the VG.

Extend the volume group.

Use vgextend to add the new physical volume to the volume group. Use the VG name and PV device name as arguments to vgextend.

[root@host ~]# vgextend vg01 /dev/vdb3

This extends the vg01 VG by the size of the /dev/vdb3 PV.

Verify that the new space is available.

Use vgdisplay to confirm the additional physical extents are available. Inspect the Free PE / Size in the output. It should not be zero.

[root@host ~]# vgdisplay vg01
  --- Volume group ---
  VG Name               vg01
...output omitted...
  Free  PE / Size       178 / 712.00 MiB
...output omitted...

Reducing a Volume Group

To reduce a volume group, perform the following steps:

Move the physical extents.

Use pvmove PV_DEVICE_NAME to relocate any physical extents from the physical volume you want to remove to other physical volumes in the volume group. The other physical volumes must have a sufficient number of free extents to accommodate this move. This is only possible if there are enough free extents in the VG and if all of those come from other PVs.

[root@host ~]# pvmove /dev/vdb3

This command moves the PEs from /dev/vdb3 to other PVs with free PEs in the same VG.

Reduce the volume group.

Use vgreduce VG_NAME PV_DEVICE_NAME to remove a physical volume from a volume group.

[root@host ~]# vgreduce vg01 /dev/vdb3

This removes the /dev/vdb3 PV from the vg01 VG and it can now be added to another VG. Alternatively, pvremove can be used to permanently stop using the device as a PV.

Extending a Logical Volume and XFS File System

One benefit of logical volumes is the ability to increase their size without experiencing downtime. Free physical extents in a volume group can be added to a logical volume to extend its capacity, which can then be used to extend the file system it contains.

Extending a Logical Volume

To extend a logical volume, perform the following steps:

Verify that the volume group has space available.

Use vgdisplay to verify that there are sufficient physical extents available.

[root@host ~]# vgdisplay vg01
  --- Volume group ---
  VG Name               vg01
...output omitted...
  Free  PE / Size       178 / 712.00 MiB
...output omitted...

Inspect the Free PE / Size in the output. Confirm that the volume group has sufficient free space for the LV extension. If insufficient space is available, then extend the volume group appropriately. See Sectionthe section called “Extending and Reducing a Volume Group”.

Extend the logical volume.

Use lvextendLV_DEVICE_NAME to extend the logical volume to a new size.

[root@host ~]# lvextend -L +300M /dev/vg01/lv01

This increases the size of the logical volume lv01 by 300 MiB. Notice the plus sign (+) in front of the size, which means add this value to the existing size; otherwise, the value defines the final size of the LV.

As with lvcreate, different methods exist to specify the size: the -l option expects the number of physical extents as the argument. The -L option expects sizes in bytes, mebibytes, gibibytes, and similar.

The following list provides some examples of extending LVs.

Extend the file system.

Use xfs_growfs mountpoint to expand the file system to occupy the extended LV. The target file system must be mounted when you use the xfs_growfs command. You can continue to use the file system while it is being resized.

[root@host ~]# xfs_growfs /mnt/data

Verify the new size of the mounted file system.

[root@host ~]# df -h /mountpoint

Extending a Logical Volume and ext4 File System

The steps for extending an ext4-based logical volume are essentially the same as for an XFS-based LV, except for the step that resizes the file system. Review Sectionthe section called “Extending a Logical Volume and XFS File System”.

Verify that the volume group has space available.

Use vgdisplay VGNAME to verify that the volume group has a sufficient number of physical extents available.

Extend the logical volume.

Use lvextend -l +extents /dev/vgname/lvname to extend the logical volume /dev/vgname/lvname by the extents value.

Extend the file system.

Use resize2fs /dev/vgname/lvname to expand the file system to occupy the new extended LV. The file system can be mounted and in use while the extension command is running. You can include the -p option to monitor the progress of the resize operation.

[root@host ~]# resize2fs /dev/vg01/lv01

Extend a logical volume and swap space

Logical volumes formatted as swap space can be extended as well, however the process is different than the one for extending a file system, such as ext4 or XFS. Logical volumes formatted with a file system can be extended dynamically with no downtime. Logical volumes formatted with swap space must be taken offline in order to extend them.

Verify the volume group has space available.

Use vgdisplay vgname to verify that a sufficient number of free physical extents are available.

Deactivate the swap space.

Use swapoff -v /dev/vgname/lvname to deactivate the swap space on the logical volume.

Extend the logical volume.

lvextend -l +extents /dev/vgname/lvname extends the logical volume /dev/vgname/lvname by the extents value.

Format the logical volume as swap space.

mkswap /dev/vgname/lvname formats the entire logical volume as swap space.

Activate the swap space.

Use swapon -va /dev/vgname/lvname to activate the swap space on the logical volume.