When we talk about virtual machine backups without commercial tools We're entering a territory where very serious business continuity needs are mixed with tight budgets, especially in SMEs and advanced home environments. The good news is that today there are quite a few free or open-source alternatives that allow you to protect environments like these. Proxmox, Hyper-V, VMware, XCP-ng, oVirt, Nutanix or Citrix without having to pay (at least at the beginning).
However, before rushing to install the first thing you find, it's important to understand the different types of backups available in virtual environments, what agent-based and agentless backups entail, the limitations of free versions, and how to fit all of this into a realistic strategy using NAS storage, public clouds, or external drives. The following sections break down the key concepts and main free solutions in detail so you can set up a backup plan. robust virtual machine backup without paying for licenses.
Basic concepts: VM backups versus “classic” backups
Virtual machine backups are not exactly the same as traditional file backups from a physical server; in a virtualized environment, you work with full VM images, virtual disks, and configuration filesThis significantly changes the technical approach and the necessary tools. A VM is a set of files (VHD/VHDX disks, VMDK, qcow2, configuration XML, snapshots, etc.) that the hypervisor manages as if it were an independent machine, so it makes sense to protect it as a logical unit.
Most modern solutions perform VM backups based on hypervisor-level imagesThis method leverages platform-specific APIs (VSS in Hyper-V, VMware APIs, Proxmox native mechanisms, etc.) to freeze the machine's state, take a consistent snapshot, and transfer only the blocks that have changed since the last backup. This contrasts with "classic" guest operating system-level backups, where an agent is installed within each VM and files and databases are copied as if it were another physical server.
Agent-based backup vs. agentless backup on virtual machines
A key concept that appears in almost all current solutions is the distinction between Agent-based backups and agentless backupsEach one has its advantages and disadvantages depending on the environment we want to protect.
In the agent-based approach, we install software within each virtual machine that will handle read the data from the guest operating system itselfMake them consistent with the applications (databases, Exchange, etc.) and send them to the backup server. This method has been used for years and remains useful in scenarios where there are only a few VMs or where we need a very fine level of control over what is backed up.
In the agentless approach, however, it is the hypervisor or virtualization platform that exposes APIs This allows backup software to access virtual disks and configuration files from outside the virtual machine. The tool connects to the host or cluster, creates snapshots of the VMs, uses Changed Block Tracking (CBT) mechanisms, and transfers the information directly from the hypervisor's storage to the backup repositories.
Practical advantages of agentless backup
Agentless backups in modern virtualization environments are becoming the preferred option because they allow for a centralized management and lower resource consumption that the traditional agent-based approach within each VM. Data protection software only needs to deploy a lightweight component on the host (or even communicate directly with the APIs) and from there discovers all the virtual machines and protects them without touching their internals.
This model significantly reduces the CPU and RAM load on virtual machines during backup execution, as most of the work is done in the hypervisor and the backup repository, taking advantage of native mechanisms for tracking modified blocksThis way, incremental backups can be completed in very tight time windows and using significantly less disk space, which is critical when storing dozens or hundreds of recovery points.
Another clear advantage is the operational efficiency in administrationThere's no need to go VM by VM installing agents, updating them, or investigating why a particular one stopped backing up. Simply add the host or cluster to the backup solution's inventory to make all hosted machines available for protection, typically after installing a transport component on the hypervisor node itself.
From an economic point of view, the agentless approach usually translates into a more favorable licensingEven in commercial versions, licensing is per host, per socket, or per number of protected VMs, but the per-agent licensing model is avoided. Although we're focusing on free tools here, this difference becomes very noticeable if you decide to upgrade to paid editions in the future as you scale the project.
Finally, well-implemented agentless backups make both the Complete recovery of VM such as granular restorationIt is possible to restart a complete virtual machine in a few steps from an image or extract individual files from a specific recovery point without needing to boot the original VM, provided that the backup solution includes these advanced features.
Proxmox VE and the importance of combining native and third-party backups
Proxmox Virtual Environment is an open-source virtualization platform based on Debian with Integrated KVM hypervisor and web-based managementWidely used in SMEs, labs, and small data centers due to its flexibility and zero license cost, Proxmox Backup Server is one of its key advantages. It's a highly efficient native solution for backing up VMs, containers, and volumes.
However, as environments become more complex or hybrid (e.g., combining Proxmox with VMware, Hyper-V, or cloud workloadsMany organizations prefer to unify data protection on a single platform that manages all hypervisors from a central console. In these cases, solutions like NAKIVO Backup & Replication come into play, which, in addition to Proxmox, cover VMware, Hyper-V, or EC2 instances on Amazon with the same backup engine and retention policies.
NAKIVO's approach to Proxmox is based on backups without agent at host levelThis allows virtual machines to be protected without installing anything internally. To do this, the application connects to the Proxmox server or cluster using administrative credentials, automatically deploys a transporter component, and begins leveraging native block change tracking technology to accelerate incremental upgrades.
Its capabilities for Proxmox environments include image-based copies with consistency at the application level (integrating with VSS-type mechanisms within VMs where appropriate), granular recovery of specific files and directories, generation of permanent incremental copies or combined with periodic full backups, and support for various storage destinations both local and in the cloud.
Storage destinations: local, NAS, tape, and public cloud
When designing a virtual machine backup scheme without paid commercial tools, one of the key decisions is where the data will be stored: local storage, NAS, network shares, tapes, or public clouds These are options that many free tools support, although with different nuances and limitations.
Some solutions allow sending backups to local disks on the backup server itself, to external USB drives, or shared resources SMB/NFS hosted on NAS devices like QNAP or Synology. This approach is usually sufficient for small and medium-sized businesses where a NAS is already available for storing VM backups, provided that security is carefully monitored and access to the data is strictly controlled.
Other applications add support for S3-compatible cloud storage (Amazon S3, Wasabi, Backblaze B2, Azure Blob, or other providers with compatible APIs), which facilitates the implementation of 3-2-1 strategies (three copies of the data, on two types of media, one off-site) without having to invest in physical tapes or libraries. In these scenarios, deduplication, compression, and the use of incremental backups are essential to reduce bandwidth and storage costs.
In environments with higher long-term conservation requirements, some free or freemium tools also allow copy backups to tape or synchronize them with other external backup repositories, either through FTPS, scheduled replication, or proprietary mechanisms optimized for large volumes of data.
Immutability, encryption, and automated backups
Beyond the simple ability to create and restore backups, modern solutions for virtual environments are beginning to incorporate functionalities of cyber resilience and advanced automation which are very interesting even in free editions. The immutability of backups is one of them, as it blocks the possibility of modifying or deleting recovery points for a predefined period, even if ransomware compromises the administrative credentials.
By enabling immutability in the backup repository, the data is protected against malicious encryption or accidental deletion for the specified number of days, providing an additional layer of security in environments where there is no budget for dedicated WORM storage solutions. It should be combined with a good retention scheme such as GFS (daily, weekly, monthly, and annual backups) to balance disk space and historical recovery capabilities.
Another key aspect is the encryption of backupsMany tools allow you to define a password or encryption key that is used to protect data before sending it to the repository, whether local or remote. This way, even if someone gains access to the backup files, they will not be able to read their contents without that key, which is especially important when working with cloud storage or locations external to the data center.
Automating copies involves flexible scheduling that allows execution daily jobs, schedules, or specific windowsIn addition to conditioning retention according to custom rules, it's common to be able to define several rules for each backup job (for example, a daily incremental backup plus a monthly full backup) and adjust job priority to prevent an excess of simultaneous tasks from overloading the virtualization host.
NAKIVO Backup & Replication in detail for Proxmox environments
In the field of solutions that support Proxmox without the need for agents within VMs, NAKIVO Backup & Replication offers a fairly deep integration with Proxmox VE and other established virtualization platforms such as VMware vSphere, Microsoft Hyper-V, or even Amazon EC2. It also includes support for agent-based backups on physical Windows and Linux machines, enabling unified protection for virtual and physical servers.
To start working with Proxmox, the first step is Add the host or cluster to the NAKIVO inventory through the Director component's web interface. From the configuration menu, access the inventory section, choose the option to add a new virtual element, and select "Proxmox Host or Cluster" as the platform, after which the basic connection details are entered.
It is necessary to specify a descriptive name to identify the server in the inventory, the type of item (standalone Proxmox host or cluster), the IP or DNS name, a user with administrator privileges (usually root), and a password. web interface port (default 8006)The SSH port (usually 22) and the path to the temporary directory, which is typically /tmp in standard installations, are required. Once these fields are completed, the wizard automatically installs a transporter component on the host or on each node in the cluster.
After this process, all virtual machines hosted on the host or in the Proxmox cluster appear in the data protection section and can then be include in backup and replication jobsThis is similar to what happens with ESXi or Hyper-V hosts added to the same inventory. From the node panel, you can verify which transporters are active and on which servers they are deployed to ensure adequate backup performance.
Creating a backup job for Proxmox VE is done from the same data protection tab, by clicking the "New Job" button specifically for Proxmox and following the wizard. In the first step, you choose the virtual machines to protect (for example, a Debian 12 VM), and in the second, you select the... destination repository (local, NAS, cloud storage, deduplication appliance, etc.) and in the third, planning and retention are defined, allowing the immutability of the backup to be activated for a specific number of days if desired.
In the job options section, you configure the job name, its relative priority compared to other concurrent jobs, whether it uses app-aware or non-app-aware mode to ensure database and service consistency, and the type of change tracking between copies. The most efficient approach is usually to use the Proxmox native change trackingHowever, you can also use NAKIVO's proprietary mechanism or disable it completely (always making full copies) if you are looking for maximum simplicity while assuming a higher space consumption.
Additionally, techniques can be enabled for network acceleration through compressionEncryption of traffic between the host and the backup repository, configuration of periodic full backups (active or synthetic) to reinforce the reliability of the incremental chain, and various additional options aimed at fine-tuning the performance and security of the backups according to the specific environment.
Once the job is defined, it can be launched immediately or allowed to run according to the established schedule. Each execution generates recovery points, which can be full or incremental. If, after the first full backup, information is modified within the VM (for example, by creating a new file in /opt/), the subsequent incremental job will only transfer the modified data. new or modified data blocks, significantly reducing times and bandwidth usage.
Proxmox virtual machine recovery: complete and granular
Recovery is the true moment of truth for any backup scheme, and in the case of Proxmox environments protected with an advanced solution, it can be performed full or granular restorations depending on the needs of the incident. From the data protection panel itself, the Proxmox backup job is selected, and the recovery menu is accessed, where the type of operation to be performed is chosen.
To restore individual files, select the corresponding option and choose which virtual machine and recovery point to use; a single backup can contain multiple VMs from different Proxmox hosts, offering considerable flexibility. Then choose whether to download the files. to the browser, send them by email, restore them to a specific server or dump them onto shared network resources, allowing adaptation to different workflows.
Once the method is chosen, you navigate through the directory structure contained in the backup, select the file or folder you want to recover (for example, a file in /opt/ generated after the last full backup), and define the recovery type (simple download, overwrite, etc.). The system executes the task and allows you to track its progress from the activities tab until the data is accessible again.
When the need is boot a full VM from the copy (for example, after a host hardware failure or due to severe file system corruption), the Proxmox VE full recovery wizard guides the administrator through several steps: selecting the necessary backup and recovery points, choosing the target host or cluster, datastore or storage where the virtual disks will reside, the virtual network to which the restored machine will connect, and optionally, the resource pool if used in the environment.
In the final phase, the task name, network encryption or compression options, and other advanced parameters are configured, after which the restoration is launched with the click of a button. The result is a virtual machine rebuilt from the backup, with its automatically generated disks and configuration files on the selected Proxmox host and ready to boot with the exact state of the chosen recovery point.
VMware backups: features and tools
In established virtualization platforms like VMware vSphere, the conceptual approach to backups is the same, but it's important to note that VMware's APIs allow for different backup methods. very specific integrations with the hypervisor For application-consistent snapshots, tracking of modified blocks, and rapid recovery of entire VMs, VMware backups are as critical as those of physical servers, since a single data loss in a VM can have a serious impact on the organization.
The differences compared to "classic" backups of physical servers lie in the fact that here we always work at the level of virtual machine images, and that the hypervisor acts as an intermediary layer offering specific backup APIs. The best VMware tools, even their free versions, leverage these capabilities to offer Highly efficient incremental backups and file-level restores from the images themselves without needing to restore the entire VM, something highly valued in IT departments with limited resources.
Free backups in Hyper-V: from native tools to dedicated solutions
Microsoft's Hyper-V ecosystem offers several options for performing backups without paying for additional licenses, ranging from the use of Windows Server backups with Hyper-V VSS even third-party solutions with Community or free editions. Each option has its own complexity, limitations, and level of automation, so it's important to choose based on the size of the environment and the level of risk tolerance.
One of the basic possibilities is to register the Hyper-V VSS writer in the Windows Server Backup tool using a series of changes to the Registry (creating the WindowsServerBackup key, the application compatibility subtree, and associating the Hyper-V writer GUID with the corresponding identifier). After these steps, Windows Server Backup can make consistent copies of the virtual machines hosted on the host, provided that all volumes containing VM-related files are included in the copy.
In practice, this means including in the backup job all disks where configuration files, virtual disks, and auxiliary files reside (for example, protecting C:, D:, and E: if the system stores the configuration XML on one volume and the VHDs on another). Virtual machines whose operating system version does not support VSS or that lack Integration Services correctly installed They will be put into a saved state while the snapshot is being taken, which may cause interruptions if it involves production services.
To restore these VMs using Windows Server Backup, the procedure involves launching the recovery wizard, selecting the server and the point in time, choosing the recovery type (applications or files and folders, as applicable), specifying Hyper-V as the data source, indicating the restore location, and running the process. It's important to note that machines with several linked snapshots They can cause problems during restoration and in some cases require removing the VM and restoring the snapshot files first before performing a second Hyper-V application recovery.
Free and open-source solutions for VM backups
In addition to native Windows tools, there is a fairly broad ecosystem of free or open-source solutions that allow backing up virtual machines on different hypervisors with varying levels of functionality depending on the product. Some are limited versions of commercial solutions, while others are completely open-source projects that require some manual work to get up and running.
Veeam Community Edition is probably one of the best known, offering the ability to protect a limited number of Hyper-V and VMware virtual machines (typically in the range of 5 to 10 VMs) with many of the features of the paid edition, including incremental backups, granular file restoration, and support for external disks or NAS as destinations. It is common to deploy Veeam Community on a separate physical server or even a dedicated VM, although it is advisable to carefully evaluate the risks of having the backup server on the same infrastructure that you want to protect.
Vinchin Backup Free Edition is another option that offers support for some specific virtualization platforms with a reduced limit on protected VMs (around 3 virtual machines in the free edition). Although its scope in the free version is more limited, it can be useful in labs or small environments that use only the compatible platforms without needing upgrades.
BDRSuite Free Edition (from Vembu) provides free virtual machine backup for VMware, Hyper-V and KVMThe idea is to allow IT administrators to familiarize themselves with the platform at no cost. One of its key advantages is the use of a proprietary driver to copy only the blocks used in Hyper-V VMs, which, according to the manufacturer, significantly improves performance compared to other solutions. This free edition typically limits the number of protected VMs, but retains many features of the commercial version.
Among open-source projects, names like Bacula, Bareos, and Bacular (in their various versions and forks) stand out, and are recommended by many administrators in technical communities. These solutions are extremely powerful and very flexible, but they also require a higher level of knowledge and configuration than the tools with more guided graphical wizards; in return, they allow orchestrating complex backups in heterogeneous environments without paying for licenses.
Other commercial tools such as Storware Backup & Recovery or Altaro's free solution have at some point offered free editions, although in some cases The free version has been discontinued or now requires a license.Therefore, it is advisable to check the current status on the manufacturer's website before basing a company's copying strategy on those options.
BackupChain: professional approach with limited free options
BackupChain is a suite of backup tools specifically geared towards IT professionals, with a clear emphasis on Hyper-V and Windows serversIt also covers other virtual deployments. Its philosophy involves offering high-granular configuration, open standard data formats, deduplication, and delta compression to minimize space consumption in virtual machine copies.
A BackupChain license can cover a Unlimited number of Hyper-V VMs on a single hostWithout the need to purchase additional licenses per virtual machine or per CPU socket, it becomes attractive when switching to a paid model. It is compatible with Windows Server versions 2008 through 2022 and also with Hyper-V in Windows 8, 10, and 11, supporting Cluster Shared Volumes (CSV), failover clusters, and sequential backups to minimize the impact on host resources.
VM backups in BackupChain can be configured in mode file-based (selecting the folder where the VHD/VHDX files reside) Alternatively, you can perform the backup automatically by selecting virtual machines directly from a list and letting the tool configure the task. All backups are performed with the VMs running, taking consistent snapshots and also preserving checkpoints or snapshots if desired, although Microsoft does not recommend their intensive use in production.
The restoration can be done at the full VM level (with the option to clone to avoid name conflicts) or granular level of files and foldersThis allows you to open virtual machine images directly from the BackupChain console and extract only the necessary data without having to restore the entire virtual disk. This Granular Backup and Granular Restore functionality is available in the Enterprise and Platinum editions of the product, designed for demanding scenarios with databases, Exchange servers, or SQL within the VMs.
Another strength of BackupChain is its ability to work with very varied backup destinationsFrom local disks and USB drives to network shares, remote FTP/FTPS servers, or even self-managed cloud storage infrastructures, the use of deduplication, delta compression, and multi-threading on the CPU allows for maximum utilization of bandwidth and available space, making it possible to replicate large VMs over relatively limited connections.
Practical strategies for SMEs: QNAP NAS, Veeam Community and native tools
In SMEs with few physical servers and a small number of Hyper-V VMs, it is very common to find scenarios where there is a QNAP NAS as centralized storage and several free tools at hand, but without a clear long-term strategy for what to use. In these cases, the dilemma often arises between sticking with the NAS's built-in solution, setting up a dedicated backup server, or running the backup tool in a VM within the existing infrastructure.
The Hyper Data Protector utility, included in many QNAP NAS devices, allows you to perform Backups of Hyper-V and VMware virtual machines to the NAS It's relatively simple, requires basic planning, and has no additional license costs. For small environments where a QNAP system is already in place and a quick and straightforward solution is desired, this option is reasonable, provided the limitations regarding fine-grained control, reporting, and manufacturer support times are accepted.
However, it's understandable that some administrators might feel uncomfortable basing a critical backup strategy on a feature that isn't the NAS provider's main business And it may not receive the same attention in terms of security and updates as dedicated backup software. That's why many consider trying Veeam Community Edition as an alternative, installing it on an older physical server or a virtual machine to compare performance and reliability.
Installing Veeam Community on a dedicated physical server, even if it's older hardware, has the advantage of isolate the backup server from the main infrastructureThis way, if something happens to the Hyper-V hosts, the backups remain available on a different machine. The risk lies in the reliability of the outdated hardware itself, so disks, power supplies, and other components must be closely monitored. Installing Veeam as a VM on one of the existing hosts is more convenient, but it carries some risk: if the host fails completely, the backup server goes down with it.
As a complementary or backup option, you can always resort to Windows Server backups with Hyper-V VSS on the host itself, storing the backups on the QNAP via shared resources. It won't be the most elegant solution, nor will it offer advanced features like granular restoration or immutability, but in very small environments, it can make the difference between having a backup and having none at all.
Volume snapshots and snapshots: usefulness and limitations
Outside the strictly virtual world, some administrators combine VM backups with mechanisms such as Volume snapshots in Windows Server (Shadow Copies) For shared folders, taking advantage of the fact that this functionality has been included by default since older versions such as Windows Server 2003. By activating them on the volumes where the shared data resides, users can quickly recover previous versions of deleted or modified files without administrator intervention.
The idea is to dedicate a local disk (physical or virtual) to store these snapshots, configuring how many are kept and how frequently they are created. You can even install a client for users to restore files themselves By themselves, they reduce the burden on the IT team. However, it's important to keep in mind that this mechanism is intended as a convenience aid and does not replace a comprehensive backup system, as it only stores up to 64 versions and resides on the server being protected.
Volume snapshots are therefore a useful complement to virtual machine backups, especially regarding the quick recovery of user files, but they cannot be considered a substitute for independent backups stored on other media, preferably isolated from the production infrastructure and with well-defined retention and periodic verification policies.
Ultimately, to put together a strategy of virtual machine backups without commercial tools To achieve even a minimally serious solution, it's advisable to combine several of the elements described: leverage free editions of established solutions like Veeam or BDRSuite in small environments, utilize open-source projects like Bacula or Bareos when time and expertise are available to fine-tune them, exploit the support for Proxmox, Hyper-V, and VMware offered by agentless tools that simplify administration, and complement all of this with native mechanisms such as VSS, volume snapshots, or utilities integrated into the NAS. This allows for a reasonable balance between zero licensing costs, manageable complexity, and genuine disaster recovery capabilities.
