Hyper Converged Infrastructure (HCI/aSV)

Sangfor HCI and aSV provide a unified infrastructure combining compute, storage, networking, and built-in security to simplify deployment, operations, and services.
{{ $t('productDocDetail.guideClickSwitch') }}
{{ $t('productDocDetail.know') }}
{{ $t('productDocDetail.dontRemind') }}
6.11.1R1
{{sendMatomoQuery("Hyper Converged Infrastructure (HCI/aSV)","Virtual Machine Snapshot")}}

Virtual Machine Snapshot

{{ $t('productDocDetail.updateTime') }}: 2026-01-05

Description

You can take snapshots for VMs when they are powered on or shut down, so that the snapshots can be used to restore the original data after service pack installation or software update.

On HCI 6.9.0 or later, snapshots can be taken for quiesced Windows VMs. That is, a snapshot is taken after data in the file system cache or business system cache is stored in disks, ensuring data consistency. This feature is applicable for the following types of snapshots:

Manual snapshots, VM (not in a consistency group) snapshots, and snapshots created for recovery

VM disk-based and storage-based snapshots

In SCP 6.10.0 and higher, Disk Based Snapshot of Snapshot Type is changed to External Disk Based Snapshot with support for configuring Merging Speed for External Snapshot Deletion. As compared to internal disk based snapshots, external disk based snapshots release space after being deleted, which eliminates the impact of snapshots on the VM performance.

Precautions

Taking VM snapshots is not used as a measure for data backup but only for data protection before high-risk VM operations such as service pack installation and software updates.

If a cluster with at least three nodes uses aSAN virtual storage, the storage-based snapshot (storage-based ROW) technology is used by default. If a snapshot has been taken for a VM in an earlier version of HCI, you can either continue to use the virtual disk-based snapshot technology or switch to the storage-based snapshot technology for the VM. To switch the technology, you need to clear existing disk-based snapshots.

If a cluster uses external storage, the virtual disk-based snapshot (QEMU-based ROW) technology is used for the VMs.

If a two-node cluster uses aSAN virtual storage, the virtual disk-based snapshot technology is used by default. The storage-based snapshot technology is available only after the cluster is expanded to three nodes and you need to clear existing virtual disk-based snapshots before switching the technology.

You can take up to 48 storage-based snapshots.

Taking a VM snapshot will interrupt the business for no more than 1 second.

Snapshots can be taken for quiesced VMs only when the VMs are running the following Windows operating systems.

If the quiesced mode is enabled but a VM does not support this feature or fails to be quiesced, a snapshot will be taken in non-quiesced mode.

When snapshots are taken in quiesced mode, consistency can be ensured for the NTFS and the applications that support this mode.

Quiescing VMs while taking snapshots will affect the business processing. Exercise caution when using this feature.

A VM cannot have an external snapshot and an internal snapshot at the same time, nor a storage based snapshot and a disk based snapshot at the same time.

Snapshots created for VMs with internal snapshots on HCI 6.10.0 upgraded from a lower version are internal snapshots by default.

Disk based snapshots created for newly created VMs or VMs without snapshots are external snapshots by default.

It is not supported to delete snapshots outside fork points. Each snapshot contains incremental changes to the basic disk, and if snapshots outside the fork points are deleted, multiple snapshot chains and a large amount of data will be merged, which has a negative impact on system performance.

External snapshot deletion takes more time than internal snapshot deletion. External snapshots contain incremental changes to basic disks as well as related index information, and when deleting an external snapshot, these incremental changes are merged into the basic disks to recover the VM to the state when the snapshot was deleted, which can be very time-consuming.

To ensure successful snapshot taking for quiesced VMs, you need to enable the following dependent Windows services:

COM+ System Application service: Enable it in manual mode.

COM+ Event System: Enable it in automatic mode.

Volume Shadow Copy: Enable it in manual mode.

Shared virtual disks will be excluded during snapshot.

Prerequisites

You have installed vmTools for the target VM.

Steps

Go to Compute > List, click the drop-down arrow next to the target VM, and select Take Snapshot.

Enter the snapshot name and description, select a snapshot type, and enable Quiesce Guest File System as needed.

Click OK.

Go to the VM details page and view the snapshot information as required.

To delete, clone, or recover the VM from the snapshot, click the camera icon.

If you recover a running VM from the snapshot, the VM will be shut down after recovery by default. To enable the VM to automatically start after recovery, select Power on virtual machine when recovery is complete.