Why This Error Happens

This usually occurs when the new disk added by vgextend has a different logical block size from disks already in the VG.

Example:

• Existing VG disks use 512-byte logical sectors
• Newly added device uses 4K logical sectors

Then vgextend may fail with this error.

Solution 1: Enable mixed block sizes in lvm.conf

Edit /etc/lvm/lvm.conf and set:

1

allow_mixed_block_sizes=1

Reference discussion:
https://serverfault.com/questions/1150643/is-it-safe-to-use-allow-mixed-block-sizes-1-in-lvm-when-using-ext4-with-4k-blo

Solution 2: Make block sizes consistent

Adjust device logical block sizes so all disks in the VG use a consistent setting.

Check logical block size

lsblk

1

lsblk -td

PHY-SEC is physical sector size, LOG-SEC is logical sector size.

sysfs

1
2

cat /sys/class/block/sda/queue/physical_block_size
cat /sys/class/block/sda/queue/logical_block_size

fdisk

1

sudo fdisk -l /dev/sda

Look at Sector size (logical/physical).

smartctl (NVMe)

1

sudo smartctl -a /dev/nvme0n1

Check Supported LBA Sizes; + indicates currently selected LBA format.

Change LBA / logical block size

Warning: formatting or changing LBA settings can destroy data.

For SATA HDD (if supported)

Check supported sector sizes:

1

sudo hdparm -I /dev/sdX | grep 'Sector size:'

If supported, switch to 4K (example):

1

hdparm --set-sector-size 4096 --please-destroy-my-drive /dev/sdX

Verify:

1

hdparm -I /dev/sdX | grep 'Sector size:'

For NVMe SSD

Check supported LBA formats:

1

sudo nvme id-ns -H /dev/nvme0n1 | grep "Relative Performance"

Or:

1

smartctl -c /dev/nvme0n1

Format to target LBA format (example):

1

nvme format --lbaf=1 /dev/nvme0n1

lbaf corresponds to the format ID. This operation will erase data and requires reboot/reinitialization steps afterward.