Have you deleted a VM; have a corrupted virtual disk; a broken snapshot chain or an unmountable datastore? We recover virtual machines on every type of virtualisation platform including VMware, Hyper-V, Proxmox etc. We restore the virtual disk to be usable again; rebuild the snapshot chain so the VM can be recovered from a point where the data was valid, all from copies, never your original files!
If a virtual machine will not boot (this could be due to a number of reasons); if a snapshot chain has been broken or if a datastore is failing, stop writing to it. Merging snapshots; attempting to merge snapshots again or allowing the host to continue operating may overwrite some of the data we require for recovery purposes. Powering on a VM whilst using a corrupted datastore may further complicate matters. Record details of the VM and the disk files involved; record details of the affected datastore and do not write to any of the above areas and contact us. The first attempt at recovering data is always the safest option.
Deletions (e.g. accidental deletion of a VM) and failures such as snapshot and datastore failures are amongst the types of failures that we recover virtual machines from most frequently: namely deletions, corruptions, snapshot and datastore.
In the case of any virtual machine (VMware, Hyper-V, Proxmox, Citrix, VirtualBox and others) and any virtual disk format, on any datastore, we will recover each one. The solution will depend on the specific platform used.
VMware ESXi, vSphere, Workstation and Fusion · Microsoft Hyper-V · Proxmox VE, Citrix XenServer, XCP-ng, Oracle VirtualBox, KVM and Nutanix · VMDK, VHDX, VHD, VDI and QCOW2 · VMFS, NFS and CSV datastores · snapshots, checkpoints and thin-provisioned disks.
Virtual machine recovery occurs at two levels: the virtual disk and the data contained within. First, we will retrieve all necessary virtual disk files from the datastore without loss; next, we will restore the integrity of those disks and their respective snapshot chains; lastly, we will mount them so we may recover the guest OS, databases and files from the copied images.
Beginning with the setup (hypervisor used; disk format type; which VM(s) were impacted), we will review and provide a written quote to begin service, typically within 48 hours.
We create a read-only clone of all your virtual disks (VMDK, VHDX, VDI, QCOW2 etc.) and first rebuild the storage for the datastore (or other locations) if they have failed. This way nothing will be written to your original data.
If there was an issue with the RAID, NAS, SAN or one of the drives required physical maintenance, we can repair these at the drive level and image them. If repairs require drive level work, this portion of the charge requires payment of 50% prior to beginning.
At this time, we make repairs to your virtual disks (descriptors, headers, grain table) and also construct your broken snapshot or checkpoint chain into a single consistent virtual disk.
Once repaired, we will then mount your disk and restore your guest file system and contents (Windows, Linux, files, SQL & Exchange databases, application data). Repairing content as necessary.
Prior to returning any virtual machines, files or database(s) from our services, we verify each VM and database/file is fully operational and complete.
Finally, we will deliver your restored virtual machines and/or data on new storage to be imported and turned back on.
All virtual machine types (VMware, Hyper-V, Proxmox etc.) will be restored by us through restoring the damaged virtual disk(s); constructing the snapshot chain; and restoring the data contained in them regardless of damage to the host or storage.
Tell us what happened and we will get back to you, usually within one working day.
We will be in touch shortly. For anything urgent, call 0117 332 1137.
Clear, tiered pricing (free to write a quote) for recovering your virtual machine, with a free diagnostic and written quote before you pay a penny for our services.
Recent examples of how we’ve recovered various types of failed virtual machines, including lost VMs, broken snapshots, corrupted disks, and datastore failures. All names have been removed for privacy protection and all have been verified.
The .vmdk file was no longer listed in the inventory for the VM, however the actual blocks that made up the .vmdk were still present on the datastore. We built the .vmdk back and recovered the entire VM.
The AVHDX checkpoint chain was broken. We rebuilt the checkpoint chain into a single .vhdx and then successfully recovered a consistent, bootable VM.
A power failure resulted in corruption to the datastore which contained 6 VMs. We rebuilt the VMFS and recovered all six VMs and all associated data.
The VHDX header was damaged. We repaired the damaged .VHDX and then successfully recovered the SQL databases within the restored .VHDX.
The QCOW2 image was left inconsistent. We rebuilt the QCOW2 image and successfully recovered the Linux guest OS along with its application data.
In this case, the datastore was sitting on a failed array. We rebuilt the RAID for the failed array. Recovered the datastore & all virtual machines (VMs) running off it.
All reviews posted on this website are based upon real customer responses; none have been created by us.
No invented reviews here — every testimonial we publish is genuine. Ask us to arrange a reference for a job like yours on 0117 332 1137.
Send us your device for a no-obligation free diagnostic and let us know what went wrong. One of our engineers will inspect it and confirm your exact quote in writing prior to doing any work.
To get started with recovering your data, first get your device to us. Wrap it securely, write your contact details on a slip of paper inside the package, include a brief description of what happened and post it to us. After running a complimentary diagnostic we will write to you confirming the exact price prior to starting work.
Do you want to post the device to us? We highly recommend using a trackable postal service with insurance. Would you prefer to deliver it yourself? That’s fine too. Our office hours for accepting deliveries are Monday to Friday, 9am to 5:30pm. Please make sure to wrap the device as described previously.
Need some additional information regarding our data recovery services? Fill out the form and provide further detail about your problem and one of our engineers will review it and offer you a personalised quote.
We’ll be in touch shortly. For anything urgent, call 0117 332 1137.
These are the top questions we get asked most about recovering a virtual machine.
Yes, most times if you act quickly enough. When you delete a VM, normally, until those disk blocks are written over with new data by that host writing to that datastore; they remain there. Stop the host writing to that datastore. Contact us & we’ll get your VM back up and running as soon as possible; however, the sooner, the better.
Almost never. Generally speaking when a VM will not start up properly it is due to a corrupted virtual disk, a broken snapshot chain or lost configuration for the machine while still having all data intact. We can repair/rebuild any corrupted VMDK/VHDX, fix any broken snapshot chains & have your VM & its contents back online.
Yes, failure in an existing (usually) snapshot or checkpoint chain is a very typical type of VM failure that results in a non-booting VM. In cases where the snapshot/checkpoint chain has been either orphaned, broken or fails to merge, we rebuild the base & delta disks back into a single consistent virtual disk and will have your machine working again.
Generally yes. We repair virtual disks at format level (descriptor, headers, grain tables & block maps) for VMDK, VHDX, VHD, VDI & QCOW2. Then mount the disk & recover guest file system and data within.
A single virtual machine (or virtual disk) on healthy storage begins from £800 + VAT. A full datastore or host containing multiple VMs begins from £1,200 + VAT, depending on what storage is involved and how much work needs doing. Each job begins as free diagnostic followed by written quote, and in most cases you only pay if we recover your data.
Yes, on most of our jobs. Where a drive needs mechanical/physical repair, we will charge a 50% deposit prior to starting recovery work and do not collect the second 50% until successful recovery of your data. Therefore, if we cannot recover your data, you are not left with the full bill.
All platforms: VMware ESXi & vSphere, Microsoft Hyper-V, Proxmox, Citrix, VirtualBox, KVM & Nutanix, and all formats of virtual disks: VMDK, VHDX, VHD, VDI & QCOW2. What platform you use dictates method of recovery, not if data can be recovered.
Yes. If the RAID, NAS or SAN which holds all of your virtual machines (VMs) is down, we will rebuild that underneath before we try to get the datastore back online and each of the VMs running on it. We deal with the storage layer and the virtualisation layer as part of the service.
Yes. After we have recovered/mounted your virtual disk we will restore/recover the guest file system and what is contained inside, i.e. files for Windows or Linux, SQL Server and Exchange databases, application data etc., and we will repair them where needed and check they are good and can be opened before restoring back to you.
A single VM or a single virtual disk generally takes around 3 to 4 working days. A full datastore/host or jobs where the underlying storage has failed could potentially take longer. The free diagnostic is typically completed within 48 hours. Urgent business cases are often prioritised over others.
If the storage is healthy you should be able to send the virtual disk files via a drive, post the disks that the datastore lives on fully insured, or drop off at our Bristol location Monday to Friday, 9am to 5:30pm. Please label any RAID disks with bay order and include your name, company, address, phone number and email so we can book it in and quote before any work is carried out.
Free diagnostic with tiered pricing from £800 & in most cases you only pay if we recover your data. Every virtual machine recovered (VMware / Hyper-V / Proxmox) and data / disk. Get your recovery started today.