CloudStack and Ceph RBD Integration

Apache CloudStack supports Ceph RBD as a primary storage zone, enabling virtual machine disks to be stored directly in a Ceph cluster. When integrated with Rook-Ceph, CloudStack can provision VM disks from the same Ceph cluster used for Kubernetes workloads, creating a unified storage tier.

Benefits of this integration:

• VM live migration without shared NFS
• Snapshot-based VM templates
• Thin-provisioned VM disks via RBD cloning
• Replication and high availability from Ceph

Step 1 - Prepare Ceph for CloudStack

Create a dedicated pool for CloudStack:

kubectl exec -it deploy/rook-ceph-tools -n rook-ceph -- \
  ceph osd pool create cloudstack 128 replicated

kubectl exec -it deploy/rook-ceph-tools -n rook-ceph -- \
  rbd pool init cloudstack

Create a dedicated Ceph user for CloudStack:

kubectl exec -it deploy/rook-ceph-tools -n rook-ceph -- \
  ceph auth get-or-create client.cloudstack \
  mon 'profile rbd' \
  osd 'profile rbd pool=cloudstack' \
  -o /etc/ceph/ceph.client.cloudstack.keyring

Step 2 - Extract Credentials for CloudStack

Get the key for the CloudStack user:

kubectl exec -it deploy/rook-ceph-tools -n rook-ceph -- \
  ceph auth get-key client.cloudstack

Get the monitor addresses:

kubectl exec -it deploy/rook-ceph-tools -n rook-ceph -- \
  ceph mon dump | grep "mon\."

Step 3 - Configure KVM Hosts for Ceph

On each KVM host managed by CloudStack, install Ceph client packages and configure authentication:

apt-get install -y ceph-common

# Create /etc/ceph/ceph.conf with monitor addresses
cat > /etc/ceph/ceph.conf << EOF
[global]
mon_host = 192.168.1.10:6789,192.168.1.11:6789,192.168.1.12:6789
auth_cluster_required = cephx
auth_service_required = cephx
auth_client_required = cephx
EOF

# Store the keyring
echo "[client.cloudstack]
    key = <ceph-key>" > /etc/ceph/ceph.client.cloudstack.keyring
chmod 600 /etc/ceph/ceph.client.cloudstack.keyring

Step 4 - Add Ceph as Primary Storage in CloudStack

In the CloudStack Admin UI, navigate to Infrastructure > Primary Storage > Add Primary Storage:

Protocol: RBD
Name: ceph-primary
Server: 192.168.1.10
Port: 6789
Pool: cloudstack
Username: cloudstack
AuthSecret: <ceph-key>

Or use CloudMonkey (cmk), the CloudStack CLI:

cmk create storagepool \
  name=ceph-primary \
  zoneid=<zone-id> \
  podid=<pod-id> \
  clusterid=<cluster-id> \
  url="rbd://192.168.1.10/cloudstack" \
  provider=DefaultPrimary

Step 5 - Verify VM Disk Creation

After adding primary storage, deploy a VM and verify its disk is created in the Ceph pool:

kubectl exec -it deploy/rook-ceph-tools -n rook-ceph -- \
  rbd ls cloudstack

CloudStack names RBD images using the volume UUID.

Step 6 - Enable Fast VM Cloning

CloudStack uses RBD clone operations for fast template-based VM provisioning. Ensure the image feature layering is enabled on all template images:

kubectl exec -it deploy/rook-ceph-tools -n rook-ceph -- \
  rbd feature enable cloudstack/<template-image> layering

Summary

Integrating Apache CloudStack with Ceph RBD provides a distributed, high-availability primary storage backend for virtual machines. Create a dedicated Ceph pool and user for CloudStack, configure KVM hosts with the Ceph keyring, and add the pool as primary storage in the CloudStack UI. VM disks are provisioned as RBD images, enabling live migration, snapshot-based templates, and thin provisioning.