Linux server migration to Windows workstation

I’m on a remote site where an incident took down most networking equipment. I have a basic background as a programmer, not in network engineering/virtualisation.

Situation
About half of the network is up again, but the DHCP server is highly unreliable and needs replacing. The best our supplier could source is a Dell Workstation – Servers at the moment can’t be found and timing is short. The workstation has two issues making migration impossible at the moment.

  1. the old server uses both ethernet ports. The new one only has one. One ethernet port is used to connect to a modem, the other connects the intranet. Everything from the intranet is going through the server then the other port to the modem and the outside world.
  2. Storage; we’re limited to a 256GB SSD, and we have a 4×4 RAID10 old server to migrate. RAID10 means striping and mirroring, so that’s a backup and speed improvement. Physically we can fit three 3.5″ drives in the new machine – if we find them. We need at the very least 4TB of storage and that’d hardly leave any room. My preference is 6TB or even 8TB if we manage.
    The old machine was set up without virtualization.

Schematically the ‘network’ looks like this:
Outside world <–> Broken Server <–> Switch <–> Intranet (printers, cameras, toughbooks/notebooks/…)

Goal
replace the old server with the new Dell workstation.

My plan

  1. get a second network controller card PCI2.1 and mount it in the new workstation in the PCI3.1 slot. We now have a second ethernet port.
  2. Order 2 6-8TB HDDS the fastest we can get our hands on and have it in a RAID1 configuration.
  3. Have a VM on the SSD and mount those RAID1-drives to it. Then install an LINUX-environment on it – or even better try to clone the old system to that virtual RAID.

Questions

  1. Internet here is slow, so top notch ethernetcontrollers will just be as slow as a 20$ controller. However, is there anything special to look out for?
  2. Can a VM be on the SSD and mount HDD raids? How do I even approach it? I hadn’t got the time to properly analyse this aspect, but any input is greatly appreciated. Or do I have to do it the other way around and configure the RAID1 in Windows and just store the disk image in that RAID1 volume?
  3. The old system is running Ubuntu 14. Can it just be cloned into something a virtual environment can read?
  4. Is the guest-OS (so the Linux VM) capable of reaching the machines connected to the switch? I mucked about with a VMBOX a few times and I always had trouble with connecting to Network Attached Equipment.
  5. Windows 10 introduced a strong update policy (good thing); however, what can be done to configure the machine in such a way that upon reboot the VM automatically resumes and that no planned reboot happens when the VM is running.
  6. My experience with VM-system is limited to VirtualBox; is this a solid choice for the described setup?

Many thanks in advance!