NAS Backup and Recovery Plugin
About the NAS Backup and Recovery Plugin
The NAS Backup and Recovery Plugin provider simple B&R operations for KVM instances to any shared storage (NAS). It is based on libvirt push backup mode to take full instance backups (qcow2) and requires libvirt-7.2.0 and QEMU-4.2, or high versions on the KVM hosts.
The NAS B&R plugin requires admin to first add backup repositories which are network-attached storage (shared storage). Currently it supports NFS, and may support other shared storage such as CephFS and CIFS/Samba in future.
When initiating B&R operations on KVM instance, the assigned backup offering is used to infer backup repository (NAS) details which are then used to mount the shared storage temporarily on the KVM host to peform instance backup/restore disks operations. This also requires that admin installs NAS-storage specific utilities on the KVM hosts such as nfs-utils/nfs-common (ceph-common, cifs-utils).
Consider the following mount, typically performed on a KVM/Linux host to mount storage:
mount -t <nas/storage type> -o <mount options> <address> <mount point>
Some examples for variety of shared storage can be:
mount -t nfs 10.10.1.10:/export /target -o vers=4.2,defaults
mount -t ceph 10.10.1.10,10.10.1.11,10.10.1.12:/ /target -o name=user,secret=xyz,defaults
The backup repository is designed to accept these parameters (type, address and mount options) as configurations to be used to execute mount operations such as illustrated above.
With ‘nas’ B&R plugin enabled, after a backup repositories are added, root admins can create new backup offerings by selecting the zone and the backup repository. These backup offerings are then assigned and used with KVM instances to perform support B&R actions and operations.
Using the NAS Backup and Recovery Plugin
To use the NAS Backup and Recovery Plugin, the Backup and Recovery framework needs to be enabled first. Then the backup plugin ‘nas’ needs to be enabled on either the global or zone settings.
Configuration |
Value |
|---|---|
backup.framework.enabled |
true |
backup.framework.provider.plugin |
nas |
Once the above two configurations are set, restart the cloudstack-management service. Once the service is restarted we can add the backup repository for the ‘nas’ Backup and Recovery plugin. Navigate to the configuration -> Backup Repository. Click on ‘Add Backup Repository’ and fill the form.
Field |
Value |
|---|---|
Name |
A suitable name to represent the Backup Repository |
Address |
URL, in case of NFS <server IP>:/path |
Type |
NFS ( only NFS type in 4.20) |
label.mountopts |
Any mount point options to be passed while mouting this storage on the hypervisor. |
Zone |
The zone in CloudStack with which this Backup Repository must be associated. |
Once the Backup Repository is created, we need to add a Backup Offering, in this plugin the Backup offering is a placeholder to associate an instance to a Backup Repository. While creating the Backup Offering, select the desired Backup Repository. Associate the Backup Offering on an instance to create an Adhoc or scheduled backup.
Support Information and Limitation
The NAS B&R plugin has been tested with EL8, EL9, Ubuntu 22.04 and 24.04. Older KVM distros such as EL7, Ubuntu 20.04 etc may not work due to libvirt/qemu version requirements. Other supported KVM-distros are not tested but may work such as OpenSUSE 15, Debian 11 and Debian 12.
Instance backups are full disk backups and limited by libvirt’s ability to initiate and handle backup. All such backups are exported and stored in qcow2 format. Due to this, restore operation are supported for volumes of type qcow2 and limited to NFS and local storage based primary storage pools.
For running instances, their disks (of any format/storage type) are backed up by libvirtd’s push based efficient-backup mechanism exported as qcow2 disks on the backup repository.
For stopped instances, qemu-img is used to convert and export full-disk backup in qcow2 format to the backup repository.
For restore operations, the KVM instance must be stopped in CloudStack. Currently, only volume(s) restoration is supported only to NFS and local storage based primary storage pools, and restored volumes are fully baked disks (i.e. not using any backing template file).
Restoring fully expunged and unmanaged instances are not supported. Backup and restore operations are not fully supported for CKS cluster instances and should be avoided.