All jobs / IT support / sysadmin
Virtualization interview questions
Virtualization questions a IT support / sysadmin can be asked, from first job to senior. Open one to see the strongest answer and why it works.
What is the difference between a hypervisor like ESXi or Hyper-V on a server and something like a desktop virtualization app on your laptop?
The strongest answer: A bare-metal hypervisor is what runs on the hardware and shares its processor, memory, storage and network among the guests; a desktop one runs as an application on top of your own operating system and competes with it
Why it works: Right. “Who owns the hardware” is the whole distinction, and it is why the server one can schedule guests fairly.
- Ask who owns the hardware.
- Bare metal: the hypervisor does.
- Desktop: your own OS does.
- Resource sharing follows from that.
You take a snapshot of a virtual machine before an upgrade. Is that your backup?
The strongest answer: No — it is a way back from the change I am about to make, and it lives on the same storage as the machine. Lose that storage and you lose both; a backup is a separate copy elsewhere
Why it works: Exactly. “Same storage, same fate” is the sentence to have ready; a backup is defined by being somewhere else.
- Same storage, same fate.
- A snapshot undoes your change.
- A backup survives losing the platform.
- Never call one the other.
You find a virtual machine carrying a snapshot taken eleven months ago, labelled “before patch”.
Why is that a problem, and what do you do about it?
The strongest answer: Every write since then went into a growing difference file, so the machine is slower and the datastore is filling — and merging one that large needs free space and heavy I/O, so I book a window
Why it works: Exactly. The removal is the risky part, not the snapshot — and it needs both space and a quiet hour.
- Snapshots grow with every write.
- Expect a slower machine and a filling datastore.
- Removal needs free space and I/O.
- Schedule it; do not do it at lunchtime.
An alert says a datastore is at 99%. The machines on it were all built with thin-provisioned disks totalling far more than the datastore holds.
What happens next, and what do you do first?
The strongest answer: Thin disks claim space only as it is written, so the overcommitment was invisible until today — and when it fills, machines writing to it stop. Free space first, then migrate one off, then fix the alert
Why it works: Right. Reclaim inside the datastore first because it is fastest, then relieve it properly, then stop it happening silently again.
- Thin disks let allocation exceed capacity.
- Full datastore means machines stop writing.
- Reclaim first, then migrate.
- Fix the alert threshold afterwards.
Users say one virtual machine is slow. Inside the guest, processor use sits around 30% and there is free memory.
Where do you look on the host?
The strongest answer: At contention the guest cannot see: how long it waits for physical processor time, whether the host is reclaiming memory from its guests, and datastore latency — and whether extra virtual processors made it harder to schedule
Why it works: Exactly — the guest cannot see the queue it is waiting in. Wait-for-processor time, memory reclamation and datastore latency are host-side numbers.
- The guest cannot see its own queue.
- Check wait-for-processor time.
- Check memory reclamation and latency.
- More vCPUs can be slower.
Everything the company runs sits on two virtualization hosts, both at about 80% memory. Both domain controllers are on the same host.
What do you argue for?
The strongest answer: Enough spare capacity that one host can fail and everything still runs, and rules that keep paired services on separate hosts — starting with splitting the two domain controllers, which costs nothing and can be done this week
Why it works: Right — and leading with the free half of the fix is what gets the paid half funded.
- Capacity to lose one host.
- Do not count headroom you cannot use.
- Split paired services across hosts.
- Lead with the free fix.