RED HAT ENTERPRISE LINUX

Volume Groups:
Assign Physical Volumes

Assign physical volumes to volume groups

CIS126RH | RHEL System Administration 1
Mesa Community College

A volume group (VG) is the central storage pool in LVM — it aggregates the physical extents of one or more physical volumes into a unified capacity that logical volumes draw from. Creating a volume group, naming it meaningfully, and managing its membership over the lifecycle of a server are core RHEL administration tasks. This module covers vgcreate, vgextend, vgreduce, and the inspection commands vgs and vgdisplay. These skills are tested on the RHCSA exam.

Learning Objectives

1. Create a volume group — Use vgcreate to pool one or more physical volumes into a named volume group, setting the physical extent size
2. Inspect volume groups — Use vgs and vgdisplay to list volume group attributes, capacity, and member physical volumes
3. Extend a volume group — Use vgextend to add a new physical volume to an existing volume group to increase its total capacity
4. Reduce and remove a volume group — Use vgreduce to remove a PV from a VG and vgremove to decommission an entire volume group

What is a Volume Group?

A volume group is a named storage pool that aggregates the physical extents of one or more physical volumes into a single addressable capacity.

• A VG has a unique name chosen by the administrator
• All PEs from all member PVs contribute to the VG's total capacity
• The physical extent size (default 4 MiB) is set at VG creation and applies to all member PVs
• Logical volumes are carved from the VG's pool of free extents
• A VG can span multiple physical disks — extents are not tied to one disk
• The device mapper provides VG access at /dev/VGNAME/

Creating a Volume Group: vgcreate

vgcreate creates a new volume group and assigns one or more physical volumes to it in a single command.

# Create a VG named "datavg" from a single PV
$ sudo vgcreate datavg /dev/sdb1
  Volume group "datavg" successfully created

# Create a VG from multiple PVs at once
$ sudo vgcreate datavg /dev/sdb1 /dev/sdc1
  Volume group "datavg" successfully created

# Create with a custom physical extent size (default is 4 MiB)
$ sudo vgcreate -s 8m datavg /dev/sdb1
# -s 8m sets PE size to 8 MiB

# Verify the VG was created
$ sudo vgs
  VG     #PV #LV #SN Attr   VSize   VFree
  datavg   1   0   0 wz--n- <10.00g <10.00g
  rhel     1   2   0 wz--n- <19.00g       0

Inspecting Volume Groups: vgs

vgs provides a concise one-line-per-VG summary — the standard command to check VG capacity and status at a glance.

# Summary listing of all volume groups
$ sudo vgs
  VG     #PV #LV #SN Attr   VSize    VFree
  datavg   2   3   0 wz--n-  <20.00g   4.00g
  rhel     1   2   0 wz--n-  <19.00g       0

# Show a specific VG only
$ sudo vgs datavg

# Add the PV count and PE size columns
$ sudo vgs -o +vg_extent_size,pv_count

Detailed VG Inspection: vgdisplay

vgdisplay shows the complete details of a volume group — the verbose counterpart to vgs.

$ sudo vgdisplay datavg
  --- Volume group ---
  VG Name               datavg
  System ID
  Format                lvm2
  Metadata Areas        2
  Metadata Sequence No  5
  VG Access             read/write
  VG Status             resizable
  MAX LV                0
  Cur LV                3
  Open LV               3
  Max PV                0
  Cur PV                2
  Act PV                2
  VG Size               <20.00 GiB
  PE Size               4.00 MiB
  Total PE              5118
  Alloc PE / Size       4096 / <16.00 GiB
  Free  PE / Size       1022 / <4.00 GiB
  VG UUID               xYz789-AB01-cd23-EF45-gh67-IJ89-kl01MN

Listing VG Members

Several approaches show which physical volumes belong to a given volume group.

# pvs shows the VG each PV belongs to
$ sudo pvs
  PV         VG     Fmt  Attr PSize   PFree
  /dev/sda2  rhel   lvm2 a--  <19.00g      0
  /dev/sdb1  datavg lvm2 a--  <10.00g  <2.00g
  /dev/sdc1  datavg lvm2 a--  <10.00g  <2.00g
# Both /dev/sdb1 and /dev/sdc1 are in datavg

# vgdisplay shows PV count and VG UUID
$ sudo vgdisplay datavg | grep "Cur PV\|PV Name"

# pvdisplay shows which VG a specific PV is in
$ sudo pvdisplay /dev/sdb1 | grep "VG Name"
  VG Name               datavg

# vgscan discovers all VGs and their PVs
$ sudo vgscan
  Found volume group "datavg" using metadata type lvm2
  Found volume group "rhel" using metadata type lvm2

Extending a Volume Group: vgextend

vgextend adds one or more physical volumes to an existing volume group, immediately increasing its total capacity. No downtime is required.

# Current VG is nearly full
$ sudo vgs datavg
  VG     #PV #LV #SN Attr   VSize   VFree
  datavg   1   2   0 wz--n- <10.00g 500.00m

# Step 1: Prepare the new disk and create a PV
$ sudo parted --script /dev/sdc mklabel gpt mkpart primary 1MiB 100%
$ sudo partprobe /dev/sdc
$ sudo pvcreate /dev/sdc1
  Physical volume "/dev/sdc1" successfully created.

# Step 2: Add the PV to the existing VG
$ sudo vgextend datavg /dev/sdc1
  Volume group "datavg" successfully extended

# Confirm the VG now has more free space
$ sudo vgs datavg
  VG     #PV #LV #SN Attr   VSize   VFree
  datavg   2   2   0 wz--n- <20.00g <10.50g

Reducing a Volume Group: vgreduce

vgreduce removes a physical volume from a volume group. The PV must have no allocated extents before it can be removed.

# Check the PV's allocated extents first
$ sudo pvs /dev/sdc1
  PV         VG     Fmt  Attr PSize   PFree
  /dev/sdc1  datavg lvm2 a--  <10.00g  <10.00g
# PFree = PSize — no extents are allocated on this PV

# Remove the PV from the VG
$ sudo vgreduce datavg /dev/sdc1
  Removed "/dev/sdc1" from volume group "datavg"

# If the PV has allocated extents, move data off first
$ sudo pvmove /dev/sdc1       # move data to other PVs in the VG
$ sudo vgreduce datavg /dev/sdc1  # then remove

# Confirm the PV is no longer in the VG
$ sudo pvs
  PV         VG     Fmt  Attr PSize   PFree
  /dev/sda2  rhel   lvm2 a--  <19.00g      0
  /dev/sdb1  datavg lvm2 a--  <10.00g  <2.00g
  /dev/sdc1         lvm2 ---  <10.00g <10.00g   ← no VG

Renaming a Volume Group: vgrename

A volume group can be renamed without affecting its logical volumes or data, though the device paths for logical volumes will change.

# Rename a VG — logical volumes must not be in use
$ sudo vgrename datavg storagevg
  Volume group "datavg" successfully renamed to "storagevg"

# After renaming, device paths change:
# /dev/datavg/data  becomes  /dev/storagevg/data

# Update /etc/fstab if the old name was used in mount entries
$ sudo vim /etc/fstab
# Change /dev/datavg/data to /dev/storagevg/data
# Or use /dev/mapper/storagevg-data (device mapper name)

# Update initramfs if the system VG is renamed
$ sudo dracut -f
$ sudo grub2-mkconfig -o /boot/grub2/grub.cfg

Removing a Volume Group: vgremove

vgremove decommissions an entire volume group. All logical volumes must be removed first.

# Attempt to remove a VG that still has LVs — fails
$ sudo vgremove datavg
  Logical volume datavg/data is still active.
Do you really want to remove volume group "datavg" containing 1 logical volumes? [y/n]: n

# Step 1: Unmount filesystems and deactivate LVs
$ sudo umount /data
$ sudo lvremove datavg/data
  Logical volume "data" successfully removed

# Step 2: Remove the VG
$ sudo vgremove datavg
  Volume group "datavg" successfully removed

# Step 3: Uninitialise the PVs if they will be reused
$ sudo pvremove /dev/sdb1 /dev/sdc1

VG Activation and Deactivation

A volume group can be deactivated (making all its LVs inaccessible) and reactivated. This is used before maintenance operations like pvmove.

# Check VG activation status
$ sudo vgs -o vg_name,vg_attr
  VG     Attr
  datavg wz--n-    ← active
  rhel   wz--n-    ← active

# Deactivate a VG (makes all LVs inaccessible)
$ sudo vgchange -an datavg
  0 logical volume(s) in volume group "datavg" now active

# Activate a VG
$ sudo vgchange -ay datavg
  2 logical volume(s) in volume group "datavg" now active

# Activate all VGs (typically done at boot)
$ sudo vgchange -ay

Migrating a VG Between Systems

LVM supports moving an entire volume group from one system to another by exporting, physically moving the disks, and importing.

# On the source system — prepare and export
$ sudo umount /data                 # unmount all LVs
$ sudo vgchange -an datavg          # deactivate
$ sudo vgexport datavg              # mark as exported
  Volume group "datavg" successfully exported

# Move the disks to the destination system

# On the destination system — import and activate
$ sudo pvscan                        # discover the new disks
$ sudo vgimport datavg              # import the VG
  Volume group "datavg" successfully imported
$ sudo vgchange -ay datavg          # activate
$ sudo mount /dev/datavg/data /data  # mount filesystems

Volume Group Commands Quick Reference

Complete LVM Setup Workflow

The full sequence from raw disk to usable filesystem, connecting all three modules.

1. Partition the disk
   sudo parted --script /dev/sdb mklabel gpt mkpart primary 1MiB 100%
sudo partprobe /dev/sdb
2. Create the physical volume
   sudo pvcreate /dev/sdb1
3. Create the volume group
   sudo vgcreate datavg /dev/sdb1
4. Create a logical volume
   sudo lvcreate -L 8G -n data datavg
5. Create a filesystem
   sudo mkfs.xfs /dev/datavg/data
6. Mount and add to fstab
   sudo mkdir /data && sudo mount /dev/datavg/data /data

Common Mistakes

Knowledge Check

Answer these before moving to the next slide.

1. Write the command to create a volume group named appvg using the physical volumes /dev/sdb1 and /dev/sdc1.
2. After creating appvg, write the command to confirm its total size and free space.
3. The VG appvg is running low on space. A new PV has been created on /dev/sdd1. Write the command to add it to appvg.
4. What does the VFree column in vgs output represent, and why is it important?
5. You need to remove /dev/sdc1 from appvg. It has 3 GB of allocated extents. Write the commands in the correct order.
6. What must be done before running vgremove appvg to safely decommission the volume group?

Knowledge Check — Answers

1. sudo vgcreate appvg /dev/sdb1 /dev/sdc1
   Both PVs must have been initialised with pvcreate before this command.
2. sudo vgs appvg — the output shows VSize (total capacity) and VFree (unallocated space). Alternatively, sudo vgdisplay appvg for full details.
3. sudo vgextend appvg /dev/sdd1 — the PV must already exist (created with pvcreate). The VG capacity increases immediately after this command.
4. VFree is the amount of unallocated space in the volume group — the space available for creating new logical volumes or extending existing ones. When VFree reaches 0, no new LVs can be created and no existing LV can be extended without first adding a new PV to the VG with vgextend.
5. 1. sudo pvmove /dev/sdc1 — move all 3 GB of allocated data to other PVs in appvg
   2. sudo vgreduce appvg /dev/sdc1 — remove /dev/sdc1 from appvg
   3. Optionally: sudo pvremove /dev/sdc1 — uninitialise the PV if no longer needed
6. Before vgremove: unmount all filesystems on the VG's logical volumes, then remove all logical volumes with sudo lvremove appvg/LVNAME for each LV in the VG. Only after all LVs are removed can vgremove complete successfully without prompting for data destruction confirmation.

Key Takeaways

1. Create a volume group with vgcreate VGNAME PV.... PVs must be initialised with pvcreate first. The VG gets a name and the PE size (default 4 MiB) is set at creation. Verify immediately with vgs or vgdisplay.
2. Use vgs for a summary; vgdisplay for full detail. VFree is the key capacity planning figure — it shows how much space remains for new or extended logical volumes. When VFree = 0, run vgextend.
3. Extend with vgextend VGNAME PV — no downtime required. The new PV's extents join the pool immediately. This is the standard way to expand storage on a live server. Prepare the PV with pvcreate first.
4. Reduce safely: pvmove → vgreduce; remove safely: lvremove → vgremove. A PV with allocated extents cannot be removed — pvmove first. A VG with active LVs cannot be safely removed — lvremove each LV first. Skipping either step risks data loss.

Graded Lab

• Using the PVs created in the previous lab (/dev/sdb1), create a volume group named labvg. Verify with sudo vgs and sudo vgdisplay labvg.
• Record the PE Size, Total PE, and VFree from the vgdisplay output. Calculate: how many 4 GiB logical volumes could be created in this VG?
• Prepare a second PV on /dev/sdb2 (create the partition with parted, run partprobe, then pvcreate). Extend labvg to include this PV with sudo vgextend labvg /dev/sdb2.
• Run sudo vgs labvg to confirm the VG now shows two PVs and increased capacity. Run sudo pvs to confirm both PVs show labvg in the VG column.
• Remove /dev/sdb2 from labvg: confirm it has no allocated extents with pvs, then run sudo vgreduce labvg /dev/sdb2. Verify with pvs that /dev/sdb2 shows an empty VG column.
• Run sudo vgscan to see the system-wide VG summary. Note that labvg appears and shows its member PV count.

Next: Getting Help from Local Documentation