USB on Virt. Boxes

Dan Egli ddavidegli at
Wed Oct 9 02:38:56 MDT 2013

On October 7, 2013, at 8:33pm, Michael Torrie wrote:

> > bridges or the like?

> That is by definition a bridge.

Sorry, I should have said "without having to define a software bridge or
anything of the like". I get that that's a bridge, but I didn't want to
have to go in and create a bridge for each computer running on the host.

> VMware hosts all have unique MAC addresses from the host. So does

> VirtualBox. and like VMware, you can change it in VirtualBox. It's

> stored in the XML file

Heh, yea, "from the host" is the point I meant. When I setup these
machines, I'd like each virtual machine to have a unique MAC. My experience
of VMware server from about three years ago showed that the GUEST eth0 had
the HOST eth0's MAC address unless specifically overridden. That's what I
was asking about. I was wondering if a new machine automatically generated
a new MAC or if it had to be done (or COULD be done) separately.

> Not directly, but you can write a little script to do it using the

> VBoxManage and VBoxHeadless commands. Note that you'd have to set up a

> script on shutdown that would save the states of the virtual machines

> (or shut them down).

That's something to think about. Under normal circumstances, the machines
WON'T shutdown. They're 24x7x365 workers. The only time I'd expect them to
shut down is in the event of power failure or hardware failure. So, in that
case, how would you manage the state, given any of the hypervisors (KVM,
VMware, VirtualBox, etc...)?

[other Qs and answers deleted]

> Given your list of requirements here, I think that the stock KVM and

> (with virt-manager) stuff that comes with most Linux distributions might

> fit your needs better than VirtualBox (with the exception of USB 2.0

> support).

Well, USB is very important since much of the work that these boxes will be
doing will involve heavy IO between the system and a USB device. I can't be
sure since I haven't actually seen the device yet, but I imagine it's
probably USB 2.0, although 3.0 is a distinct possibility. How does KVM
handle USB 3.0 (or the new 3.1 for that matter)?

> VirtualBox can run headless, but at that point you may as well be using

> KVM which has better init script support, and is designed to run in a

> server fashion.

Yea, sounds like KVM is good, except for the USB issue above. Any
workarounds for USB 2.0 with KVM? If not, what would be your next


--- Dan

More information about the PLUG mailing list