If you want to create three Logical Volumes (LVs) of sizes 1TB, 1TB, and 5TB on a single Volume Group (VG) for each disk, here’s how you can do it step by step.
Assume:
Disks: /dev/nvme0n1, /dev/nvme1n1, /dev/nvme2n1
Volume Groups: vg_nvme0, vg_nvme1, vg_nvme2
Logical Volumes (LVs): lv1, lv2, lv3 for each VG
LV Sizes: 1TB, 1TB, and 5TB
1. Create Physical Volumes (PVs)
1 2 3 4
# Initialize each disk as a PV: sudo pvcreate /dev/nvme0n1 sudo pvcreate /dev/nvme1n1 sudo pvcreate /dev/nvme2n1
2. Create Volume Groups (VGs)
1 2 3 4
# Create a VG for each disk: sudo vgcreate vg_nvme0 /dev/nvme0n1 sudo vgcreate vg_nvme1 /dev/nvme1n1 sudo vgcreate vg_nvme2 /dev/nvme2n1
# Create directories for each LV and mount them: sudo mkdir -p /mnt/nvme0/lv1 /mnt/nvme0/lv2 /mnt/nvme0/lv3 sudo mkdir -p /mnt/nvme1/lv1 /mnt/nvme1/lv2 /mnt/nvme1/lv3 sudo mkdir -p /mnt/nvme2/lv1 /mnt/nvme2/lv2 /mnt/nvme2/lv3
sudo mount /dev/vg_nvme0/lv1 /mnt/nvme0/lv1 sudo mount /dev/vg_nvme0/lv2 /mnt/nvme0/lv2 sudo mount /dev/vg_nvme0/lv3 /mnt/nvme0/lv3
sudo mount /dev/vg_nvme1/lv1 /mnt/nvme1/lv1 sudo mount /dev/vg_nvme1/lv2 /mnt/nvme1/lv2 sudo mount /dev/vg_nvme1/lv3 /mnt/nvme1/lv3
sudo mount /dev/vg_nvme2/lv1 /mnt/nvme2/lv1 sudo mount /dev/vg_nvme2/lv2 /mnt/nvme2/lv2 sudo mount /dev/vg_nvme2/lv3 /mnt/nvme2/lv3
7. Persist the Mounts
1 2 3 4 5 6 7 8 9 10 11 12
# Add the mounts to /etc/fstab for automatic remounting on boot: echo '/dev/vg_nvme0/lv1 /mnt/nvme0/lv1 xfs defaults 0 0' | sudo tee -a /etc/fstab echo '/dev/vg_nvme0/lv2 /mnt/nvme0/lv2 xfs defaults 0 0' | sudo tee -a /etc/fstab echo '/dev/vg_nvme0/lv3 /mnt/nvme0/lv3 xfs defaults 0 0' | sudo tee -a /etc/fstab
echo '/dev/vg_nvme1/lv1 /mnt/nvme1/lv1 xfs defaults 0 0' | sudo tee -a /etc/fstab echo '/dev/vg_nvme1/lv2 /mnt/nvme1/lv2 xfs defaults 0 0' | sudo tee -a /etc/fstab echo '/dev/vg_nvme1/lv3 /mnt/nvme1/lv3 xfs defaults 0 0' | sudo tee -a /etc/fstab
echo '/dev/vg_nvme2/lv1 /mnt/nvme2/lv1 xfs defaults 0 0' | sudo tee -a /etc/fstab echo '/dev/vg_nvme2/lv2 /mnt/nvme2/lv2 xfs defaults 0 0' | sudo tee -a /etc/fstab echo '/dev/vg_nvme2/lv3 /mnt/nvme2/lv3 xfs defaults 0 0' | sudo tee -a /etc/fstab
The difference is in the last number (0 vs 2) in the fstab entry, which represents the filesystem check (fsck) pass number. Here’s what these numbers mean:
Last field (6th field) in fstab:
0 = No filesystem check will be done at boot time
1 = Filesystem will be checked first (typically used for root filesystem /)
2 = Filesystem will be checked after pass 1 filesystems (typically used for other filesystems)
So:
defaults 0 0 means the filesystem will never be automatically checked during boot
defaults 0 2 means the filesystem will be checked during boot, but after the root filesystem
Best practices:
Use 0 1 for the root filesystem (/)
Use 0 2 for other important filesystems that should be checked
Use 0 0 for pseudo-filesystems (like proc, sysfs) or filesystems that don’t need checking (like swap)
wipe disk and create lvm
assum dev is /dev/nvme2n1
To erase all partitions on the device /dev/nvme2n1 and create multiple logical volumes (LVs) using the LVM framework, follow these steps:
1. Verify Device and Backup Data
1 2
# Ensure you are working on the correct device. Erasing partitions will delete all data on the device. sudo lsblk -o NAME,SIZE,TYPE,MOUNTPOINT /dev/nvme2n1
2. Erase Existing Partitions
1 2 3 4 5
# Clear the Partition Table, To wipe the partition table completely: sudo wipefs -a /dev/nvme2n1 # Verify the Disk is Clean, Check that no partitions remain sudo lsblk /dev/nvme2n1
3. Create Physical Volume (PV)
1 2 3 4 5
# Convert the entire disk into an LVM physical volume: sudo pvcreate /dev/nvme2n1 # Verify the PV sudo pvdisplay
4. Create Volume Group (VG)
1 2 3 4 5
# Create a volume group that spans the entire disk: sudo vgcreate vg_nvme2n1 /dev/nvme2n1 # Verify the VG: sudo vgdisplay
# Format each logical volume with your desired file system (e.g., XFS): sudo mkfs.xfs /dev/vg_nvme2n1/lv1 sudo mkfs.xfs /dev/vg_nvme2n1/lv2 sudo mkfs.xfs /dev/vg_nvme2n1/lv3
7. Mount Logical Volumes
1 2 3 4 5 6 7 8
# Create mount points and mount the LVs: sudo mkdir -p /mnt/nvme2n1/lv1 /mnt/nvme2n1/lv2 /mnt/nvme2n1/lv3
sudo mount /dev/vg_nvme2n1/lv1 /mnt/nvme2n1/lv1 sudo mount /dev/vg_nvme2n1/lv2 /mnt/nvme2n1/lv2 sudo mount /dev/vg_nvme2n1/lv3 /mnt/nvme2n1/lv3 # Verify the mounts: df -h
8. Make the Mounts Persistent
1 2 3 4
# Add entries to /etc/fstab to ensure the LVs are mounted on reboot: echo '/dev/vg_nvme2n1/lv1 /mnt/nvme2n1/lv1 xfs defaults 0 2' | sudo tee -a /etc/fstab echo '/dev/vg_nvme2n1/lv2 /mnt/nvme2n1/lv2 xfs defaults 0 2' | sudo tee -a /etc/fstab echo '/dev/vg_nvme2n1/lv3 /mnt/nvme2n1/lv3 xfs defaults 0 2' | sudo tee -a /etc/fstab
wipe existed lv
1. Check What’s Using the LV
First, identify what is still using the LV:
1
sudo lsof | grep /dev/mapper/<lv-name>
Also, check active processes using the device:
1
sudo fuser -m /dev/mapper/<lv-name>
If any process is using the LV, stop it:
1
sudo kill -9 <PID>
2. Unmount If Mounted
Check if the LV is mounted:
1
mount | grep /dev/mapper/<lv-name>
If it is mounted, unmount it:
1
sudo umount -l /dev/mapper/<lv-name>
Use -l (lazy unmount) to force unmount if needed.
3. Disable the LV
Before removing the LV, deactivate it:
1
sudo lvchange -an /dev/<vg-name>/<lv-name>
Now try to remove it:
1
sudo dmsetup remove /dev/mapper/<lv-name>
4. Forcefully Remove LV, VG, and PV
If the above steps don’t work, forcefully remove everything:
