62 comments

[ 3.6 ms ] story [ 120 ms ] thread
The detached start option is very nice. Also the vm group things, as tiny as it is, really helps when you setup multi-VMs netlabs.
"HiDPI!support"

Does that mean Virtual Box will no longer look ridiculous on my 3840 x 2160 laptop screen?!

Downloading....

Yes, that is absolutely what it means. Works beautifully on my Retina Macbook Pro.
Been waiting for this since I bought a MacBook.
Windows 10 support? Anyone tied it yet?
Yes, did you read the linked Data Sheet PDF from the article?

> Improved Huge Range of Guest Platforms – including the very latest Windows 10, Windows Server 2012 R2 and leading edge Linux platforms too.

http://www.oracle.com/us/technologies/virtualization/oraclev...

Well, in theory, given several other comments / complaints about things not quite working right (USB3, and the like), there is a difference between:

"is it supported?" (yes)

and

"does it work well?"

I've been using W10 on v4 fine for the past few weeks. Planning to update right now though.. anything you are concerned about specifically?
have you been able to use virtio for network/storage?
(comment deleted)
Just finished installing on my test PC. Ran Ubuntu (Since it was all I had on hand), and it keeps throwing errors when I try to power on the new machine. Going to install latest release and try again.
Why do I get this from https://www.virtualbox.org when I access from Chrome? The site works from curl and Firefox.

Traceback (most recent call last):

File "/usr/lib/python2.4/site-packages/trac/web/api.py", line 436, in send_error data, 'text/html')

File "/usr/lib/python2.4/site-packages/trac/web/chrome.py", line 803, in render_template message = req.session.pop('chrome.%s.%d' % (type_, i))

File "/usr/lib/python2.4/site-packages/trac/web/api.py", line 212, in __getattr__ value = self.callbacks[name](self)

File "/usr/lib/python2.4/site-packages/trac/web/main.py", line 298, in _get_session return Session(self.env, req)

File "/usr/lib/python2.4/site-packages/trac/web/session.py", line 162, in __init__ self.get_session(sid)

File "/usr/lib/python2.4/site-packages/trac/web/session.py", line 189, in get_session self.bake_cookie()

File "/usr/lib/python2.4/site-packages/trac/web/session.py", line 170, in bake_cookie assert self.sid, 'Session ID not set'

AssertionError: Session ID not set

UPDATE: same error from www

UPDATE: curl --header "Cookie: trac_session=" https://www.virtualbox.org

The issue seems to be that some plugin in Chrome erases my trac_session cookie, and the website cannot handle this. I am running Disrupt, uBlock, and PrivacyBadger.

(1) Why does vbox need to track my session?

(2) What python framework are they using? App should not fail in this case.

I was seeing this a while back until I disabled a plugin I had installed. I can't remember which one it was. Try disabling your Chrome plugins and try again.
As a fun experiment, I do my day-to-day computing entirely in a Virtualbox Windows VM guest that I've given 2 cores, 150GB of storage and 4GB of RAM. I'm about a year and a half into the experiment and still chugging along.

It's a surprisingly performant day-to-day system, which I can snapshot to try out things, move to other machines if I need to, make backups etc. About the only thing it doesn't do well is really CPU intensive or GPU intensive operations.

But it works fine for 2 monitors, web browsing, watching videos, etc.

It's kind of surprising actually.

From time to time I'll also spin up some Linux VMs and do various dev activities in a real Linux, which I usually just background and ssh into from my Windows VM. It's kind of nice having a virtual rack of machines to monkey around on.

Less impressive has been trying to get Ubuntu to not feel terrible, but Centos works fine.

If I need better performance, I'll dive back to the host OS and do those things, but it's mostly just for gaming or music production.

Bonus, my host OS has stayed relatively free of junk and stays really snappy, even all this time later.

My only recent problem is that the Windows 10 updater won't qualify the VM guest for the upgrade. So I'll probably have to grab the ISO and try it that way.

I do this because my employer has some silly policies and it was the only way they would let my run Linux on my localhost. Really I forget that I'm in the VM.. RAM is a little scarce, but all in all it works surprisingly well.
Same reason here, our IT department only supports windows and I've been mostly using VBox to work in linux at work for the past 3 years without any incident whatsoever. The clipboard, mouse and keyboard sharing are really fantastic. It really makes you forget this is a guest OS.
i've tried doing the same but the performance hit is unfortunately too big for me to find it acceptable. silly policies.
You should try it in Hyper-V. GPU-accelerated desktops are still no good, but everything else performs really well.
from what i understand it, hyper-v is only supported by a couple of distros and even then, the performance gains are not that noticable. definitely worth looking into none the less, thanks.
I need 32GB RAM, so no chance for virtualization. But Kubuntu handles snapshots (BTRFS) and moving between computers (move SSD) without a glitch.
How has your experience with BTRFS been? I'm intrigued and want to try it out.
I have mostly good experiences with it. I have more examples of surprising good things it does and scenarios it will survive OK. But then the edge cases, if you hit them, can be brutal. So you have to amp up the backup strategy for important data, just because of the unknown factor. Development is really incredible the 4.2 merge window included almost 2000 additions for Btrfs, and almost as many subtractions. It's really hard for a mortal person to keep track of the bugs, bug fixes, optimizations and regressions.

For example right now, a rather major regression in ext4 conversions: The convert goes find, mount is fine, rollback is fine. But if you scrub it, hard panic; even console is lost, let alone all services. But the fs survives. If you balance it, corruption of the fs (and so far not repairable). So right now, which could change at any moment, I'd say avoid ext4 to btrfs conversions. Create a new fs, and keep backups. Usually it's pretty good about at least mounting read only so you can update a backup, even if the fs is beyond repair.

Despite piles of kernel features, it still lacks much needed notification to user space of device failures. For multiple device volumes, you probably want to be informed of either flakey or failed devices and right now you have to be watching for it. Some edge cases in degraded operation exists when replacing devices that you have to be careful of, etc. So this whole area needs work, which means more developers interested in such things are needed.

But anyway, it's going pretty good.

move to other machines if I need to

Last time I tried that meant having to do a rather slow and disk hungry export followed by an import. Was I doing it wrong and is there no way to just copy the disk and machine definition, or are there other ways already?

I never do the export/import process for that exact reason.

If I need to move a VM for some reason, I copy the entire VM directory to the new location then open VirtualBox and use "Machine" -> "Add ..." and select the directory with the VM config and files.

I hope that helps, unless I missed something.

It would be nice to have the whole thing on an external SSD, which I could switch between my desktop and laptop.

Does this work or could the VMs get corrupted?

I have had no issues running a VM on an external drive.
Even if you ran it on one machine with a Linux host and on another with a Windows host?
I've not moved VM's between Windows and Linux, but it works fine going back and forth between OS X and Windows. The only caveat being that you have to re-visit the VM's settings for things like host volume mapping and especially networking config.
Seems like the .vdi files are configured relative, so at least this should go automatically.

Host volume mapping is a folder shared between the host and the VM?

> Host volume mapping is a folder shared between the host and the VM?

Yes.

Bought a 128GB USB 3.0 stick. Works like a charm.

Hosts:

Notebook runs Linux Mint 17 with VirtualBox 5.0

Desktop runs Windows 7 with VirtualBox 4.3 (5.0 ha an issue on Win7 that Oracle won't fix, but prevent my VM from starting)

Guest:

Xubuntu 15.04 with the VirtualBox 5.0 Guest Additions

thanks (also to bane for providing the link which essentially states the same), will try that next time
Same thing here. At work my host OS has nearly nothing installed on it, with an unusual toolbar and background color to remind me I'm not on the "right" machine.

I do all of my work in a VM, and as my org gets new machines, I xcopy the VHD over with minimal downtime. Full screen remote desktop is fine. My work is generous with hardware dollars, so I'm not really in any need of more memory or disk space.

That said, my work doesn't really involve anything involving lots of screen drawing. Video sucks through remote desktop (audio/video de-sync or rdp connection drops), so if I need to watch a video (presentation or youtube), I either suffer through it or do that on the host machine.

> the Windows 10 updater won't qualify the VM guest for the upgrade

I was able to get my VM to qualify. Make sure you have the required hotfixes, otherwise you will not qualify -- KB3035583 and KB2976978 for Win8.1, KB3035583 and KB2952664 for Win7 SP1.

Then reboot and run the batch file given here: http://superuser.com/a/922442

I may have to try this again later this year when I buy new desktop parts in anticipation of Fallout 4.

In the present time though, I tried this once on my work laptop, running a Core i7(2011), with 8GB RAM and Windows 7 as the Host OS and VM OS. With the usual platter HDD, Visual Studio got bogged down so much during compilation times(we had a really big project), that it became too unfeasible to continue using.

Installed it for USB3, did not manage to make it work, although I configured it.
It requires installation of the Extension pack - make sure you got that updated.
From the changelog[0]:

Make more instruction set extensions available to the guest when running with hardware-assisted virtualization and nested paging. Among others this includes: SSE 4.1, SSE4.2, AVX, AVX-2, AES-NI, POPCNT, RDRAND and RDSEED

This makes me incredibly happy.

[0] https://www.virtualbox.org/wiki/Changelog

Advise caution if you are a windows 8.1 host. Besides still having to deal with https://www.virtualbox.org/ticket/13187#comment:178 this new version completely disabled networking, even after a reboot.

Oracle has done a very good job of ruining VB for quite a few users.

It also doesn't work with Realtek VLAN.
Meanwhile, performance (on an OS X host) compared to Parallels or VMWare is horrid. I ended up buying a Parallels license for when I do VM/OS install testing, which saves me hours a week.
One of the biggest features here, the virtual USB 3.0 controller, is closed source (as well as USB 2.0!). See:

https://www.virtualbox.org/manual/ch01.html#intro-installing

While it's available for free, the license only allows personal and educational use. Bleh.

I contacted oracle once to buy licenses and they said unless I needed 500 or more not to bother them
I might need to try this before my next upgrade to Parallels. If it works for what I need why spend the 100 bucks for each upgrade.
I agree, I previously ran Parallels for Coherence; but found that some apps just didn't perform well.

This, coupled with not being simultaneously able to run Vagrant/Docker while using Parallels will likely cause me to switch.

Does this still require its own crashy kernel driver on Linux?
Yes. They should find a way to make it work with KVM. Although I don't claim to understand how the new KVM/Hyper-V PV support is implemented, at least on Linux the 5.x RCs loaded the dreaded VBoxDrv.
Sadly, boot2docker 1.7.0 doesn't [yet?] work with VirtualBox 5
boot2docker is not connected to Vagrant, it speaks directly with Virtualbox. Maybe watch out in the boot2docker github repo when they gonna update it https://github.com/boot2docker/boot2docker/issues/979 My advice also: Avoid the official docker installation. It will downgrade your Virtualbox 5.0 pretty sure!
I've used Xen, VMWare workstation 9/10, but primarily Virtualbox since 2008, on linux and windows hosts and linux and windows guests.

Virtualbox has really never let me down, I don't see any reason to use VMware workstation over it.

Personally, I've stuck with VMWare because the licensing costs are low enough that I'm never sufficiently motivated to spend time on a Virtualbox shakedown.
Last time I checked the seamless mode of VMWare was VASTLY superiour to the VirtualBox one. While VirtualBox did not much more than hide your desktop-background and make it transparent, VMware integrated the window manager of the guest OS into the host OS (i.e. guest-windows would appear in the host-OS window manager's window list). Furthermore it overlaid the screen with a button with an applications menu (native to the host-OS) sourced from the guest OS.

Nowadays I am without a WMware licence, and I found that KVM and its frontends are enough for my use cases right now because I do not rely on Windows-VMs for my day-to-day work so I can always use X forwarding for an integrated experience.

Extensively evaluated Vmware player vs virtual box for Ubuntu 14.04 desktop on W8.1 on my Thinkpad. Vmware might be a touch faster but too many things are broken - touchpad scrolling doesn't work - requires hacks to get this running ineffectively. Suspend and resume doesn't work..etc. Went back to Virtualbox since the experience is smoother.
Hmm getting a hash sum mismatch when doing an apt-get update. Anyone else seeing this?