How to resize a DRBD backing disk

Resizing a DRBD backing disk increases the block storage available to an existing replicated resource after the lower-level device has been enlarged on every node. It is useful when an LVM logical volume, virtual disk, storage LUN, or similar device behind a resource needs more capacity without rebuilding the replication pair.

DRBD grows the replicated device after it reexamines the lower-level devices. The safe growth path is to extend the backing device on each node first, run drbdadm resize for the resource, wait for any required synchronization to finish, and only then resize the filesystem through the DRBD device or its mounted filesystem.

Start with an existing, healthy resource and keep the same size on every lower-level device before DRBD is told to grow. Replace resource names, backing devices, DRBD minors, and mount points with the production values for each node, and avoid running filesystem tools directly against lower-level devices.

Steps to resize a DRBD backing disk:

Check that the DRBD resource is connected and has one writable primary before changing the backing devices.
```
$ sudo drbdadm status r0
r0 role:Primary
  disk:UpToDate
  node-b role:Secondary
    peer-disk:UpToDate
```
Do not resize during split-brain, disconnected replication, or unknown disk state unless a DRBD recovery plan already identifies the authoritative node.

Record the current DRBD device and backing device sizes on each node.

$ lsblk /dev/drbd0 /dev/vgdata/r0
NAME          MAJ:MIN RM  SIZE RO TYPE MOUNTPOINTS
drbd0         147:0    0  100G  0 disk /srv/data
vgdata-r0     253:0    0  100G  0 lvm

Run the same check on every node that owns a backing device for the resource.

Back up the DRBD metadata before enlarging the storage layer.
```
$ sudo drbdadm dump-md r0 > r0-drbd-metadata.txt
```
Keep the metadata dump outside the backing device being changed.
Related: How to back up DRBD metadata before a change
Extend the underlying backing device on the first node.
```
$ sudo lvextend --size +20G /dev/vgdata/r0
  Size of logical volume vgdata/r0 changed from 100.00 GiB to 120.00 GiB.
  Logical volume vgdata/r0 successfully resized.
```
This example uses LVM. Use the storage platform's own grow operation for partitions, SAN LUNs, virtual disks, ZVOLs, or cloud volumes.
Extend the matching backing device on the peer node before asking DRBD to grow.
```
$ sudo lvextend --size +20G /dev/vgdata/r0
  Size of logical volume vgdata/r0 changed from 100.00 GiB to 120.00 GiB.
  Logical volume vgdata/r0 successfully resized.
```
DRBD can refuse to reconnect to a peer whose lower-level device is still smaller than the resized resource.

Confirm that every backing device now shows the same larger size.

$ lsblk /dev/vgdata/r0
NAME       MAJ:MIN RM  SIZE RO TYPE MOUNTPOINTS
vgdata-r0 253:0    0  120G  0 lvm

Resize the DRBD resource from a node where the resource is up.
```
$ sudo drbdadm resize r0
```
Use the default resize path unless the added area is known to be identical or unimportant on every node. The lower-level --assume-clean option skips resynchronizing added space, and --assume-peer-has-space is only for peer devices that are not connected but have already been grown.
Watch the resource until every peer device is connected and up to date.
```
$ sudo drbdadm status r0
r0 role:Primary
  disk:UpToDate
  node-b role:Secondary
    replication:Connected peer-disk:UpToDate
```
A large grow can show a temporary synchronization state before the peer disk returns to UpToDate.
Related: How to verify DRBD synchronization state

Verify that the DRBD device reports the new size.

$ lsblk /dev/drbd0
NAME  MAJ:MIN RM  SIZE RO TYPE MOUNTPOINTS
drbd0 147:0    0  120G  0 disk /srv/data

Grow the filesystem through the DRBD device on the primary node.
```
$ sudo resize2fs /dev/drbd0
resize2fs 1.47.2 (1-Jan-2025)
Filesystem at /dev/drbd0 is mounted on /srv/data; on-line resizing required
old_desc_blocks = 13, new_desc_blocks = 15
The filesystem on /dev/drbd0 is now 31457280 (4k) blocks long.
```
Run filesystem resize tools against /dev/drbd0 or the mounted filesystem, not against the lower-level backing device.

For XFS, grow the mounted filesystem with sudo xfs_growfs /srv/data instead of resize2fs.

Confirm that the mounted filesystem exposes the larger capacity.

$ df --human-readable /srv/data
Filesystem      Size  Used Avail Use% Mounted on
/dev/drbd0      118G   41G   72G  37% /srv/data

Author: Mohd Shakir Zakaria
Mohd Shakir Zakaria is a cloud architect with deep roots in software development and open-source advocacy. Certified in AWS, Red Hat, VMware, ITIL, and Linux, he specializes in designing and managing robust cloud and on-premises infrastructures.