Ctrlk
For the complete documentation index, see llms.txt. This page is also available as Markdown.

Nutanix Acropolis Hypervisor (AHV)

Overview

Storware Backup & Recovery protects virtual machines running on Nutanix Acropolis Hypervisor (AHV). It discovers virtual machines, backs up their configuration and disk data, and restores them to a Nutanix environment. Backup policies allow you to schedule full and incremental backups and exclude disks that do not require protection.

Storware Backup & Recovery provides two Nutanix integrations: Nutanix Prism and Nutanix v4. Both are configured as virtualization providers. This page covers both integrations; choose the provider that matches your environment, then follow its configuration section.

Choose a virtualization provider

UI option
Backup method
Management endpoint
Key requirements

Nutanix Prism

Disk attachment to a proxy VM

Prism Central or Prism Element

At least one proxy VM running Storware Backup & Recovery Node in each protected cluster

Nutanix v4

Nutanix API v4.1 or later, with iSCSI data transfer

Prism Central

Prism Central 7.5 or later, compatible AOS, and node access to Nutanix iSCSI interfaces

For new installation we recommend using Nutanix v4.

Supported versions

Nutanix v4
Nutanix Prism

Supported versions

7.5

7.3, 7.5

The last snapshot is kept on the hypervisor for incremental backups

No

Yes (required by CRT)

Access to hypervisor OS required

No

No

Proxy VM required

No

Yes

Full backup

Supported

Supported

Incremental backup

Supported

Supported

Synthetic backup

Supported

Supported

File-level restore

Supported

Supported

VM disk exclusion

Supported

Supported

Snapshot management

Supported

Supported

Quiesced snapshots

Supported

Supported

Pre/post command execution

Supported

Supported

Access to VM disk backups over iSCSI

Supported

Supported

VM name-based policy assignment

Supported

Supported

VM tag-based policy assignment

Supported

Supported with Prism Central

Power-on after restore

Supported

Supported

Access to VM disk backups over iSCSI is a recovery feature. It is separate from the iSCSI data transport used by the API v4 backup method.

Nutanix Prism

Backup strategy: disk attachment

This method uses Nutanix APIs to create snapshots and attach their disks to a proxy VM running Storware Backup & Recovery Node. The node reads data from the attached disks and transfers it to the backup destination. Incremental backups use Nutanix changed-region information to read only changed blocks.

Deploy at least one proxy VM in each protected cluster. The node must run inside that cluster so that it can attach the snapshot disks.

Requirements

  • Install at least one Storware Backup & Recovery Node as a proxy VM in each protected cluster.

  • Use a Prism account with cluster-admin privileges.

  • When connecting through Prism Central, use the same credentials for Prism Central and every managed Prism Element accessed by the node.

  • Run inventory synchronization for each node so that Storware Backup & Recovery can identify its proxy VM and record its UUID.

  • Configure the proxy VM's disk layout and LVM settings as described in Proxy VM disk configuration.

Network access

Source
Destination
Port
Purpose

Storware Backup & Recovery Node

Prism Elements

TCP 9440

Nutanix API communication

Storware Backup & Recovery Node

Prism Central, if used

TCP 9440

Nutanix API communication

Add the provider

  1. Create a proxy VM in each protected Nutanix cluster using a supported operating system.

  2. Install Storware Backup & Recovery Node on each proxy VM.

  3. Configure the proxy VM's disks and LVM settings as described below.

  4. Go to Virtual Environments → Virtualization Providers and click Create.

  5. Select Nutanix Prism. On the General tab, enter the Prism Element or Prism Central URL:

  6. Enter credentials for an account with cluster-admin privileges. When using Prism Central, the credentials must also be valid on every accessed Prism Element.

  7. Save the configuration and run inventory synchronization for each node. Verify that the hosts and virtual machines are discovered and the node's proxy VM is identified.

  8. Assign a backup policy to a small test virtual machine and run a backup.

  9. Check the node logs to confirm that the node reads the attached snapshot disk, rather than one of the proxy VM's own disks.

  10. Perform a test restore before enabling production schedules.

Proxy VM disk configuration

Keep the proxy VM's disk layout simple. Where the backup destination permits staging on the destination storage, use a single local disk for the proxy VM operating system.

If local staging requires a separate disk, use a volume from Nutanix container storage. Verify device selection during a test backup, particularly when the proxy VM has multiple disks.

Configure LVM filters according to the Storware guide LVM setup on Storware Backup & Recovery Node for disk attachment backup mode. Include the proxy VM's own OS disks, partitions, and any required local staging devices. Reboot the proxy VM after applying the changes as directed by that guide.

Verify the dynamically attached disks slot offset

The Dynamically attached disks slot offset setting determines where the node expects dynamically attached disks. A proxy VM with several local disks may require manual adjustment.

  1. Before starting a backup, connect to the proxy VM and run:

  2. Record the disks that belong to the proxy VM.

  3. Enable debug logging on the node and run a test backup of a small VM.

  4. Inspect the node logs and verify that the exported device is an attached snapshot disk.

For example, if /dev/sda and /dev/sdb belong to the proxy VM, a snapshot disk appearing as /dev/sdc is consistent with that layout. Device names depend on the actual environment and must be checked locally.

If the node selects one of its own disks, correct the offset using KB10037: How to change the Dynamically attached disks slot offset parameter before running further backups.

Multiple nodes

You can deploy several nodes in a cluster and map individual hypervisors to them. Each node handles VMs assigned to its mapped hypervisors, providing static distribution of backup jobs. VMs without an assigned hypervisor are handled by the node configured for the virtualization provider.

Run inventory synchronization for every node to register its proxy VM UUID.

Backup and restore

Backup

Storware Backup & Recovery creates a VM snapshot, exports its metadata, and attaches the snapshot disks to the proxy VM one at a time. The node reads the disk data locally. Incremental backups use Nutanix changed-region information to read only changed blocks.

If at least one schedule assigned to the VM specifies incremental backups, the latest snapshot is retained for the next incremental backup.

Restore

Storware Backup & Recovery creates empty disks on the proxy VM and writes the restored data to them. It then recreates the target VM and attaches the restored disks to it.

Snapshot consistency

Storware Backup & Recovery supports crash-consistent snapshots and optional application preparation through pre/post snapshot commands. The QuiesceBeforeSnapshot setting requests a quiesced snapshot and requires guest tools in the protected VM. The resulting snapshot type depends on the availability of those tools; requesting quiescing does not guarantee application consistency.

Limitations and scope

  • Backup of VMs with vTPM enabled is not supported.

  • Only standard VM disks are protected through the VM backup integration. Attached Nutanix Volume Group disks are not included in VM snapshots, backups, or snapshot reverts.

  • Nutanix Volume Group and Nutanix Files protection must be configured separately in the Storage Provider section.

  • Tag-based policy assignment requires Prism Central.

  • At least one proxy VM is required per protected cluster.

Nutanix v4

Backup strategy: API and iSCSI

This method uses Nutanix API v4.1 or later to manage backup and restore operations through Prism Central. Disk data is transferred through Nutanix-managed iSCSI interfaces, without attaching disks to a proxy VM or accessing the hypervisor operating system.

Full backups skip empty disk regions to reduce the amount of stored data. Incremental backups use Nutanix Recovery Points and changed-region information to read the blocks that have changed since the previous backup. Disk data and metadata can be processed in parallel, with Nutanix distributing data access across Controller VMs (CVMs).

This method requires Prism Central 7.5 or later and compatible managed clusters.

Requirements

  • Prism Central must run version 7.5 or later and provide API v4.1 or later.

  • Managed clusters must run an AOS (Prism Element) version compatible with the connected Prism Central instance.

  • Use an account with administrator-level permissions. Limited or custom roles are not recommended because missing privileges can cause backup or restore failures.

  • The node must be able to reach Prism Central and the Nutanix cluster interfaces used for iSCSI data transfer.

Network access

Source
Destination
Required access

Storware Backup & Recovery Node

Prism Central

TCP 9440 for API communication

Storware Backup & Recovery Node

Nutanix-managed iSCSI interfaces on the protected clusters

iSCSI data transfer for backup and restore

Provide a stable connection with sufficient bandwidth. The node requires both API access and access to the Nutanix iSCSI data interfaces.

Add the provider

  1. Go to Virtual Environments → Virtualization Providers and click Create.

  2. Select Nutanix v4. On the General tab, enter the Prism Central connection details:

    Field
    Value

    URL

    https://NUTANIX_PC_IP:9440/api

    Username

    Nutanix account with administrator-level permissions

    Password

    Password for that account

  3. Save the configuration and run inventory synchronization.

  4. Verify that the Nutanix hosts appear on the Hypervisors tab and virtual machines appear on the Instances tab.

  5. Assign a backup policy to a test virtual machine and run a backup.

  6. Verify that the backup completes successfully, then perform a test restore to validate the recovery path.

Successful inventory synchronization confirms discovery access. A test backup and restore also validate data access and the permissions required for protection operations.

Backup and restore

Backup

  1. Storware Backup & Recovery retrieves the VM configuration, disk information, and metadata.

  2. It creates the required Nutanix Volume Group and Recovery Point resources.

  3. It prepares iSCSI access on the node and obtains information about empty disk regions and, for incremental backups, changed regions.

  4. The node reads disk data through Nutanix-managed iSCSI interfaces. Disk data and metadata are processed in parallel by default.

  5. Data is transferred to Storware-managed staging space without being staged on the Nutanix host. The backup contains disk data, VM metadata, and disk block maps.

  6. Temporary resources and iSCSI client configuration are cleaned up after the operation. For incremental backup policies, the latest Recovery Point is retained as the reference for the next backup.

Restore

Storware Backup & Recovery creates a temporary Volume Group and restores disk data through iSCSI. Multiple disks can be restored in parallel. When restoring an incremental backup, the required full backup and incremental changes are used to reconstruct the selected disk state.

After data transfer, the iSCSI connection is detached and the VM is created using the restored disks, preserving their layout and order. Temporary Volume Groups, iSCSI client configuration, and other temporary restore resources are then cleaned up.

Advanced settings

Configure optional export settings in /opt/vprotect/node/node.properties on the node:

Property
Default
Description

nutanix.export.strategy

parallel

parallel processes disks and metadata concurrently; single processes them sequentially.

nutanix.export.parallel.disk-threads

3

Number of threads for parallel disk data transfer.

nutanix.export.parallel.metadata-threads

3

Number of threads for parallel metadata processing.

nutanix.export.parallel.timeout

P2D

Maximum duration of parallel export operations, expressed as an ISO 8601 duration. P2D means two days.

nutanix.writer.log.interval

60

Interval between export progress log entries, in seconds.

Parallel processing is the recommended mode. Adjust concurrency to match available node resources, network bandwidth, and Nutanix cluster capacity.

Authentication and error handling

Authentication is managed automatically. Storware Backup & Recovery uses basic authentication to obtain a session token, then reuses the token for subsequent API calls. If the token expires, it authenticates again and retries the request without requiring user action.

Transient API errors are retried automatically. By default, a request can be retried twice at one-second intervals, with a 30-second timeout when the API does not respond. Some endpoints use different settings for longer operations.

API requests are queued and scheduled to control request frequency. Storware Backup & Recovery monitors asynchronous Nutanix tasks before proceeding to dependent steps. If an operation fails, it attempts to remove the temporary resources created during that operation.

Limitations

  • Prism Central versions earlier than 7.5 do not provide the API functionality required by this integration.

  • Prism Element alone cannot be used as the management endpoint for this method.

  • Limited or custom Nutanix roles are not recommended.

  • Backup of VMs with vTPM enabled is not supported.

  • Only standard VM disks are protected through the VM backup integration. Attached Nutanix Volume Group disks are not included in VM snapshots, backups, or snapshot reverts.

  • Nutanix Volume Group and Nutanix Files protection must be configured separately in the Storage Provider section.

  • Tag-based policy assignment requires Prism Central.

Incremental backup reference points

Both providers retain a reference point for subsequent incremental backups: a snapshot for Nutanix Prism or a Recovery Point for Nutanix v4. Do not manually remove the reference point retained by Storware Backup & Recovery.

PreviousCitrix Hypervisor and XCP-ng
NextOpenStack

Last updated