Creating a virtual copy of your computer. Hyper-V: Cloning a virtual machine

Consistency of backup data is the sum of the validity, accuracy and integrity of the data in relation to files, application data and the operating system of the computer or virtual machine. Data consistency is critical to any backup system, regardless of its size or company security policies. In this article, I will discuss how to ensure that your backup is consistent and can be restored at any point in time without any data corruption.

Backup file consistency levels

There are several levels of data consistency in backup files. Levels are determined according to the following criteria:

  • system type(whether a backup was made of a “live” or shutdown system)
  • file accounting(can interdependent files be clearly defined relative to each other)
  • application accounting(whether protection was used to prevent data loss at different levels, including the transaction level)

Let's look at all types of consistency in order.

Inconsistent backup

The first and oldest type of backup is using inconsistent copies. It can be thought of as the operation of copying all the computer data while it is running to a USB flash drive or network storage. If even one file changes during the backup, the backup will become inconsistent. This means that the amount of backup data will differ from the current state of the system, and data recovery will not be accurate or complete. This type also does not involve storing a RAM dump and may have difficulty with read-only files.

Stateless backup

The next type of backup is the so-called application-stateless backup. The main difference from inconsistent copies is that this backup is created over a specific period of time. Think of this copy as a snapshot of the entire contents of your hard drive—as if you had taken a photograph.

However, restoring a system image from such a copy is approximately equivalent to turning on the computer after a power failure. Data from applications open at the time of the crash and active I/O operations (at the time of the crash) cannot be recovered. In some cases, when launching applications after recovery, you need to perform special operations to restore their functionality. This is especially dangerous for database applications. For example, a Microsoft Exchange server may require the installation of a special recovery group and separate log integration. Microsoft SQL or Oracle applications will require database state recovery down to a specific transaction because a failure in their operation has disrupted the sequence of operations, and some transactions must be re-executed before they are committed by the system.

File-level consistent backup

File-level consistency means that the contents of all open files at the time the backup begins must be committed to the hard drive. However, even in this case, the system still has no idea of ​​the ongoing transactions with the application databases.

Transactionally Consistent Backup

Such a copy does not have the disadvantages described above, and all VM data at a precise point in time is saved in a backup copy. The problem with creating such a copy is that file and application reconciliation operations must be carried out when the desired VM is in operation and access to it must be continuous for clients.

For VMs running Windows, Veeam helps make backups fully consistent using Shadow Copy Service (VSS). This service was developed by Microsoft and is included in all editions of Windows operating systems starting with Windows Server 2003. The Shadow Copy Service coordinates the processes that prepare the system for freezing. It allows you to temporarily suspend I/O operations on your hard drive and thus helps your backup software.

“Application-aware image processing” is a Veeam technology that guarantees successful full recovery of VMs and VM application data without data loss.

Veeam Backup & Replication does not install software agents on VMs. Instead, the guest machine runs a service application to coordinate the guest OS backup. Once copying is complete, the application is automatically deleted. The mechanism is designed to prevent potential difficulties with preinstalling, troubleshooting and updating software agents.

Veeam Backup & Replication initiates the shadow copy service and sends a request to create consistent VM disk contents before taking a snapshot of the VM. The shadow copy service provider then coordinates the preparation of applications for backup. This ensures data consistency within a precisely defined period of time. This ensures that there are no pending or uncommitted database operations or incomplete application files when the VM contents are copied.

How can I make sure my backup is consistent?

At the stage of configuring a backup or replication task, you need to make sure that on the “guest processing” tab the “enable application-aware processing” option is activated for the VMs that are selected for this task. You can then run the task and check its execution status.

Using the “application stateful backup” mechanism is especially critical in the case of backups of SQL or Exchange servers, as it allows you to control the growth of transaction logs and truncate them if necessary. Otherwise, control over log growth occurs manually.

The shadow copy service works great for modern Windows machines, but in the case of a Linux environment, you need to use a different method.

Freezing feature (quiescence) from VMware

For example, for virtual machines running on VMware ESXi servers, VMware provides a “quiescence” option. This feature allows guest OS I/O to be frozen in advance when the server is preparing to take down a copy of the virtual machine. Detailed instructions for working with this option can be found in.

Veeam backup jobs can use both VMware's Application Stateful Backup and Freeze options. In this case, Veeam Backup & Replication will use the first option by default, and if it fails or is impossible to use, it will switch to the other.

You can check whether data can be restored from a backup using .

Useful Resources

  • Veeam Blog: (English)
  • Help Guide: " " (English)

GD Star Rating
a WordPress rating system

How to ensure backup data consistency, 3.2 out of 5 based on 5 reviews

Windows Server 2008 Hyper-V does not have the option to create a clone of an existing virtual machine. However, this does not mean that it is impossible or that it is difficult to do. In fact, there is such a possibility and it is quite simple to use.

There are two ways to clone a virtual machine in Hyper-V.

1. Use the export/import function of virtual machines
2. Copy the virtual hard disk and create a new virtual machine with this disk.

1. Use the export/import function in Hyper-V

This method is the easiest and most affordable way to clone a virtual machine.

Exporting a virtual machine

While in Hyper-V Manager, right-click on the desired virtual machine and select Export. Specify where the virtual machine will be imported. It is important to know that when you import the virtual machine back, the location of the virtual machine and its virtual disk will point to this location.

During export, you can choose what to export: the entire virtual machine or just the virtual machine configuration.

Importing a virtual machine

To import a virtual machine, click “Import Virtual Machine” from the Action menu. Specify the directory with the exported virtual machine and click Import.

It is important to indicate exactly the directory of the exported machine, and not one of its subdirectories.

Once the virtual machine is imported, you can change its settings, such as IP address, hostname, etc.

2. Copy the VHD and create a new virtual machine

Simply copy the VHD file of the original virtual machine and then create a new virtual machine, but specify the copied file as the hard drive. In the menu, when you reach the “Connect Virtual Hard Disk” window, select the “Use an existing virtual Hard Disk” option and specify its location.

Guard

Chinese doesn't always mean bad. Chinese front loaders are sold here at factory prices.

Do you need a fence for your dacha, cottage, or home? Contact "Profzabor" - production of custom-made fences from any materials.

Many users who start using virtualization tools from VMware sooner or later wonder how to create a copy of a virtual machine (clone it) in the free version of VMware ESXi. If you have a commercial edition of VMware vSphere and a vCenter management server, then you can clone a VM simply from the machine’s context menu:

However, things are a little more complicated if you have the free version of VMware ESXi Free (aka vSphere Hypervisor). Here's how you can go:

This tool allows you to create a virtual machine on the target host (namely our free ESXi), install the Converter itself inside the machine, and create a clone of it as a physical system. In this case, during the cloning process ("migration"), both systems are preserved, and various settings, such as the size of the virtual disk, the system name, etc., can be customized.

2. Use software for backup and recovery.

Unfortunately, you won't be able to use the excellent Veeam Backup and Replication product for free ESXi, but there are some products that can do this. For example, Trilead VM Explorer and Unitrends.

First, we create a backup copy of the machine, and then restore it in parallel with an existing VM.

3. You can simply copy the virtual machine and its disk.

The first option of this method is simple - copy the folder from the VM (you can use WinSCP or FastSCP to access the ESXi file system). Next, add the VM to the ESXi environment with a *.vmx file through the context menu and the "Add to inventory" item:

The second option is to use the vmkfstools utility. It allows you to clone virtual machine disks by specifying various parameters of the target disk. For example, this command creates a clone of a virtual disk, but the target disk will be in a thin format (that is, growing as it is filled with data):

vmkfstools -i /vmfs/volumes/storage/server1/server1.vmdk /vmfs/volumes/storage/server1_clone/server1_clone.vmdk -d thin

You can do a lot of things using this utility; more details about it are written in KB 1028042. Next, we create a new VM and attach the resulting virtual disk to it. Don't forget to change the machine name and network identification!

There is also a way to clone a VM using (vMA) and vSphere CLI (vCLI) as written in KB 1027872, but it requires deploying vMA and is not worth the hassle of cloning a single VM. But for regular machine cloning, be sure to study this KB.

Please enable JavaScript to view the

Using VM Groups

VM groups allow the user to create special groups of virtual machines for collective management.

Create a group using the GUI

1) Drag one virtual machine on top of another virtual machine.

2) Select multiple virtual machines and select " Group» in the right-click menu as shown below:

You can create nested groups.

Group operations allow you to perform common actions for virtual machines for all members of the group at once.

Pictures

With snapshots, you can save a specific state of a virtual machine for later use. At any subsequent time you can return to this state, even though you may have changed the VM significantly since then. So a snapshot of a virtual machine is similar to a machine in a "saved" state as described above, but there can be many of them, and these saved states are stored permanently.

You can see snapshots of a virtual machine by first selecting the machine in VirtualBox Manager and then clicking the " Pictures" in the upper right corner. Until you take a picture of the car, the list of pictures will be empty, except for the item " Current state", which represents the point " Now"in the life of a virtual machine.

Creating, restoring and deleting snapshots

There are three operations associated with snapshots:

  1. You can to take a photo. This makes a copy of the current state of the machine, which you can later return to at any time.
  • If your virtual machine is currently running, select " Take a snapshot of the state..." in the drop-down menu " Car» VM windows.
  • If your virtual machine is currently in the " saved" or " off" (as shown next to the virtual machine in the main VirtualBox window), click the " Pictures" in the upper right corner of the main window, and then
    • or on the small camera icon (for “ To take a photo"), or
    • right click on the element " Current state" in the list and select " To take a photo».

In any case, a window will appear asking for the name of the image. This name is for reference purposes only to help you remember the snapshot status. For example, a useful name would be " Fresh installation from scratch, without guest additions th" or " Service pack only 3". You can also add longer text in the " Description", if you want to.

Your new snapshot will then appear in the list of snapshots. Below the new photo you will see an item called " Current state", meaning that the current state of your virtual machine is a variant based on the snapshot you used previously. If you later take another photo, you'll see that they will appear sequentially, with each subsequent photo derived from the earlier one:

VirtualBox doesn't place any restrictions on the number of pictures you can take. The only practical limitation is the disk space on your host: each snapshot saves the state of the virtual machine and thus takes up some disk space.

  1. You can restore snapshot by right-clicking on any photo you took in the list of photos. Restoring a snapshot is like going back to the moment it was taken: the current state of the car is lost, and the car is returned to the exact state the car was in when the snapshot was taken.

Note Note: Restoring a snapshot will affect the virtual hard disks attached to your virtual machine. This also means that all files created since the snapshot was taken and all other changes to files will be lost. To prevent data loss when using the Snapshot feature, you can add a second hard drive in " write-through» using the VBoxManage interface and use it to store your data. Because writable hard drives are not included in snapshots, they remain unchanged when the machine is returned.

To avoid losing the current state when you restore a snapshot, you can create a new snapshot before restoring.

By restoring an earlier snapshot and taking more snapshots, you can create a sort of alternate reality and switch between these different histories of the virtual machine. This may result in a tree of virtual machine snapshots as shown in the screenshot above.

  1. You can also delete the snapshot, which will not affect the state of the virtual machine, but only release the disk files that VirtualBox used to store the snapshot data, thereby freeing up disk space. To delete a snapshot, right-click it in the snapshot tree and select " Delete" Since VirtualBox 3.2, snapshots can be deleted even while the machine is running.

Note Note: While creating and restoring snapshots are fairly fast operations, deleting a snapshot can take a significant amount of time because it may require copying a large amount of data between multiple disk image files. Temporary disk files may also require a large amount of disk space while the operation is running.

There are some situations that cannot be processed while the virtual machine is running, and you will receive a corresponding message that you should delete this snapshot while the virtual machine is turned off.

Think of a snapshot as time that you have saved. More formally, a photograph consists of three things:

  1. It contains a complete copy of the virtual machine settings, including hardware configuration, so that when the snapshot is restored, the VM settings are also restored. (For example, if you change the hard drive configuration or system settings of the virtual machine, that change is discarded when the snapshot is restored.)
  2. A copy of the settings is saved in the device configuration, an XML text file and therefore takes up very little space.
  3. The full state of all virtual disks attached to the machine is saved. Returning to a snapshot means that all changes made to the computer's disks - file by file, bit by bit - will also be undone. Files created since creation will disappear, files that have been deleted will be restored, changes to files will be undone.

(Strictly speaking, this is only true for virtual hard disks in "normal" mode. As mentioned above, you can configure the disks to behave differently with snapshots. Even more formally and technically correct, restoring a snapshot does not restore the virtual disk itself Instead, when a snapshot is taken, VirtualBox creates diff images that contain only the changes since the snapshot was taken, and when the snapshot is restored, VirtualBox discards that diff image, thus reverting to the previous state. This is faster and uses less disk space.

Creating a differential image as such initially does not take up much space on the host's disk, since the differential image will initially be empty (and later grow dynamically with each disk write operation). However, the longer you use the machine after taking a picture, the more the size of the different image will increase.

Finally, if you took a snapshot while the machine was running, the device's memory state is also saved in the snapshot (just as memory can be saved when closing a VM window). When you restore such a snapshot, execution resumes at the exact moment the snapshot was taken.

The memory state file can be as large as the virtual machine's memory and therefore take up quite a lot of disk space.

Removing virtual machines

To delete a virtual machine that you no longer need, right-click it in the VM Manager list, select " Delete».

A confirmation window appears allowing you to choose whether to remove the machine from the list of machines only, or whether to also delete the files associated with it.

Menu item " Delete

Cloning virtual machines

To experiment with a virtual machine's configuration, test different levels of a guest OS, or simply create a backup of a virtual machine, VirtualBox can create a full or linked copy of an existing virtual machine.

The wizard will guide you through the cloning process:

This wizard can be called from the VM Manager list context menu (select " Clone") for the selected virtual machine. First, choose a new name for the clone. You can select the option to generate MAC addresses of all network cards, then each network card in the clone will receive a new MAC address. This is useful when both the source VM and the cloned virtual machine need to run on the same network. If you leave this unchanged, all NICs will have the same MAC address as the one in the source VM. Depending on how you invoke the wizard, you have different options for the cloning operation. First you need to decide whether the clone should be linked to the original VM or a completely independent clone:

Full cloning: In this mode, all disk image dependencies are copied to a new VM folder. The clone can run completely without the original virtual machine.

Linked cloning: This mode creates new distinctive disk images where the parent disk images are images of the source disk. If you selected the current state of the source VM as the cloning point, a new snapshot will be created implicitly.

After selecting the cloning mode, you need to decide what exactly you want to clone. You can always create a clone of just the current state or everything. When you select all, the current state and in addition all snapshots are cloned. If you started with a snapshot with additional children, you can also clone the current state and all children. This creates a clone starting from this snapshot and including all child snapshots.

The cloning operation itself can be lengthy depending on the size and number of disk images attached. Also keep in mind that each snapshot has different disk images that need to be cloned as well.

Menu item " Clone» switches off while the machine is running.

Import and export of virtual machines

VirtualBox can import and export virtual machines in the standard Open Virtualization Format (OVF).

OVF is a cross-platform standard supported by many virtualization products that allows you to create ready-made virtual machines that can then be imported into a virtualizer such as VirtualBox. Importing and exporting into VirtualBox in OVF format is very simple and is done from the Manager window, as well as the command line interface. This allows you to package so-called virtual devices: disk images along with configuration settings that can be easily deployed. This way you can offer complete, ready-to-use software packages (operating systems with applications) that require no configuration or installation other than importing into VirtualBox.

Note: The OVF standard is complex and its support in VirtualBox is an ongoing process. In particular, VirtualBox is not guaranteed to support all devices created by other virtualization software.

Devices in OVF format can be displayed in two options:

  1. They may be supplied in multiple files as one or more disk images, usually in the widely used VMDK format and an XML text file with the extension .ovf. To be able to import them, the files must be in the same directory.
  2. Alternatively, the above files may be packaged together into a single archive file, typically with the extension .ova. (Such archive files use a variant of the TAR archive format and therefore can be unpacked outside of VirtualBox with any utility that can unpack standard TAR files.)

To import device into one of the above formats, simply double-click the OVF/OVA file. Or select " File» → « Importing configurations" in the Manager window. In the file dialog box that appears, navigate to the file with the extension .ovf or .ova.

If VirtualBox can process this file, a dialog box similar to the following will appear:

It represents the virtual machines described in the OVF file and allows you to change the virtual machine's settings by double-clicking the description elements. Once you click " Import", VirtualBox will copy the disk images and create local virtual machines with the settings described in the dialog box. They will then appear in the Manager's list of virtual machines.

Please note: Because disk images tend to be large, and the VMDK images that come with virtual appliances are typically sent in a special compressed format that is not suitable for direct use by virtual machines, the images will first need to be uncompressed and copied, which may take several minutes .

And vice versa, for export virtual machines that you already have in VirtualBox, select " File» → « Export configurations" Another dialog box will appear that allows you to combine multiple virtual machines into an OVF device. Then select the location where the target files should be saved and the conversion process will begin. It can take some time.

Note: OVF cannot describe the snapshots that were taken for the virtual machine. As a result, when exporting a virtual machine with snapshots, only the current state of the machine will be exported, and the disk images in the export will have a “flattened” state identical to the current state of the virtual machine.

Global Settings

The global settings dialog can be found in the menu " File", selecting the item " Settings….» It offers a set of settings that apply to all virtual machines of the current user or, in the case of extensions, to the entire system:

  • Are common. Allows the user to specify the default folder/directory for VM files and the VRDP authentication library.
  • Enter. Allows the user to specify a Host key. It is used to switch the cursor out of focus of the virtual machine or host operating system windows, and is also used to trigger certain virtual machine actions.
  • Updates. Allows the user to specify various automatic update options.
  • Language. Allows the user to specify the language of the graphical user interface.
  • Display. Allows the user to specify the screen resolution, as well as its width and height.
  • Net. Allows the user to configure network information for the host only.
  • Plugins. Allows the user to view and manage installed extension packs.
  • Proxy. Allows the user to configure an HTTP proxy server.

Perhaps I’ll start with the fact that if you want backups on VMWare, then get ready to pay. Free VMWare is free as long as we are not talking about migrations, backups and the like. At this point you can start an endless holiwar, but without my participation. My stories will only be about Hyper-V on Windows Server 2012R2. Although some of the article can be applied to VMWare, there are likely to be pitfalls.

We can backup to Hyper-V for free, or rather, using Windows tools for which we have already paid by purchasing Windows Server licenses. For the convenience of working with our backups (besides, we also paid for this), we will use WDS and deduplication (maybe group policies).

1. Backup from inside virtual machines

1.1. Backup of today

As far as we know, any Windows can make backups. Moreover, any Windows backup settings through the interface ultimately come down to the background use of the wbadmin utility. What exactly can wbadmin do? And it can make both a backup of an image with a system partition and a backup of individual folders. In this part of the article, we are only interested in the backup image (of the system partition). The rest is specific data of virtual machines and needs to be backed up separately. Hence the conclusion: Do not store any valuable information, databases, or individual applications on the system partition of virtual machines (and on regular computers too). We install MS SQL Server / MS Exchange / “1C Application Server” and others only on non-system partitions or on separate disks.

So, what is needed for the backup to work? And you only need one command:

Wbadmin.exe start backup -backupTarget:\\Backup Server\FolderForInternal Backup -allCritical -quiet
In fact, this command requires special rights, but more on them later. Now it is important to understand one thing. This command does more than just backup. She makes an incremental backup. Moreover, different backups are generated for server and desktop (client) Windows. And the difference is that for server operating systems we will get snapshots of each backup, but for desktop OSes, only the last snapshot will always remain. You may ask, what kind of incremental backup is this? But it remains “incremental”, because we are not backing up the entire image, but only the part that has changed since the last backup (which means less traffic and faster backup creation).

Those who have encountered a similar situation will notice that the backup will always be “incremental” (full). Since the backup takes place in our case to a network drive. That is, for server Windows, only the latest images remain.

Later, I discovered that there is no difference in the work of wbadmin on the server and client OS. The only difference is in the interface. wbadmin performs incremental backups (except for the first backup) if the hard drive is specified in the -backupTarget key (the command uses the default key -vssCopy). Or it produces a full backup if you add the -vssFull switch.

1.2. Backup with the history of previous images

At the moment, we have made a backup of virtual machine images. But this is just a backup of today’s photos. Tomorrow it will be completely different... But what will happen if you make backups? Yes, and truly incremental. That's what we'll do.

Wbadmin.exe start backup -backupTarget:e: -allCritical -include:d: -quiet #where drive D is backups for today, and drive E is backups with history
But this was not enough for me and I did this:

(echo select vdisk file="\\Backup Server2\FolderFor Backup2\BackupNameForThisServer2.vhdx" echo attach vdisk) | diskpart
The script connects a virtual disk from the network. After the backup, a similar script disconnects the disk. The OS remembers that the drive has the letter E defined. But God forbid you slip in someone else’s drive with the same letter E, the backup will work in full (not incrementally and to someone else’s drive). Keep this in mind and use a letter towards the end of the alphabet (X, Y, Z)…

Let me note right away that if today’s backup is carried out in parallel with the backup with history, we will end up with a backup that cannot be restored.

To get a backup of previous days, you can use the interface (GUI) of the server on which backups with history are made. Moreover, it knows and remembers all runs of the wbadmin command in the Windows console. The recovery service will allow you to select the desired archive in backups with history.

2. Backup vhdx files of virtual machines

It is done easily and naturally:

Wbadmin start backup -backupTarget:$BackupPath -hyperv:$VMList -Quiet
But with some peculiarities. This command must be executed in PowerShell and first obtain the list of virtual machines into a variable. For a detailed example, please contact Google.

Backup of virtual machines in Windows Server 2012 R2 is done using Hyper-V snapshots. I also note that virtual machines are suspended if they have a Linux kernel or are missing Hyper-V drivers. I personally refused to backup virtual machines in this way. The reason is that on Windows Server 2012 (not R2) it was necessary to stop virtual machines before backup. And even now on Windows Server 2012 R2, Linux pauses do not suit me, when there is the first good backup method. (there is a remark in the comments to this article). After the next update to Windows Server 2012 R2, backup of any virtual machines occurs without interruption. Linux OS can also be backed up “from within” using Dump (CentOS, Ubuntu), but this is a separate topic with puppets and other software in my case.

3. Restore backup and WDS

And now, in my opinion, the most useful part of this article is about backups.

WDS is Windows Deployment Services and part of the functionality of Windows Server 2012R2. This service used to be called RIS, but I have not encountered it. In general, the essence of WDS is simple. We registered in DHCP (automatically for DHCP Windows Server) in the form of separate parameters and then downloaded to the computer over the network (this is how the computer BIOS is configured to boot over the network) via the TFTP WDS bootloader. Next, the WDS loader allows you to select from the Windows “bootloader” images available on it. There are different bootloaders - these are installer bootloader images, PE, and RE images. The installer loader also requires images of Windows themselves in WDS, but this is in case you need to install Windows over a network. We are interested in RE images that allow you to recover a machine from a backup.

I won’t explain in detail how and what works in WDS. But here are the important notes:

  1. If your RE bootloader loads on a Hyper-V virtual machine over the network, but the keyboard does not work in it. Congratulations, your RE image is for WinXP or older and does not know about the existence of Hyper-V drivers.
  2. If your system starts restoring a backup, but stops. Delete all partitions on the hard drive (on which the backup is restored) and try again. Just don’t forget that the backup may be broken and after deleting all the partitions on your hard drive, you may not have anything left of the old information.
  3. If the backup is loaded with UEFI, and you want to restore it to a computer without UEFI, then don’t waste your time. Most likely you won't be able to deploy the backup.
  4. A backup with UEFI boot and GPT partitions can be restored to machines with a different processor/motherboard, but with MBR format partitions and a regular BIOS boot, it is unlikely to be deployed on another machine. Well, I definitely didn’t succeed.
  5. If you try to deploy a backup to a disk with a smaller capacity, then this will not work. Even if the disk in the backup was almost empty. In this case, restoring to a virtual machine with a dynamic disk helps. Next, reduce this disk and create a new backup. But this is only possible with a UEFI boot loader in a backup (for why, read the previous paragraph).
  6. Before restoring a backup, you should disconnect extra disks so as not to overwrite the information on them.

4. Features of deduplication

You can deduplicate running virtual machines. You can deduplicate today's backups and you can deduplicate historical backups. All this gives a big positive plus to the volume of hard drives (for both HDD and SSD). But don't forget about some things:
  1. If deduplication works on disks larger than 1 TB, the deduplication optimizer will use a lot of memory.
  2. If deduplication works with compressed data, but with more than 10 TB of compressed data, then the duration of the deduplication optimizer will be too long. This can happen if you simply copy data daily to a deduplicated disk in different folders.
  3. It is possible and even necessary to store backups on HDD, but you should not store working virtual machines on HDD in quantities greater than 5-10. This applies to deduplication only in that deduplication of such working virtual machines will reduce HDD performance to zero.

5. Group policies

Here you can implement the installation of a backup script using GPO for a long time and in different ways. But I would like to draw attention to important points:
  1. Backups should only be made on behalf of a separate user account.
  2. Do not store scripts with passwords in group policies.
  3. Run a script with special rights to read all system information.
Well, actually conclusion: Backup using wbadmin tools is possible, quite feasible and, most importantly, viable. But only if you have time and a lot of patience for all sorts of little things. For example, the article does not say anything about how to monitor successfully created backups. I managed to do monitoring on Zabbix, but I could write a couple more articles about this... I hope the article will be useful to you and save a lot of invaluable time.