Dark Reading is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Risk

8/12/2008
11:05 PM
George V. Hulme
George V. Hulme
Commentary
50%
50%

Securing Virtualization, Or Is That Virtualizing Security?

One of the big topics at last week's Black Hat and Defcon security confabs was virtualization security, but few speakers talked about what is really important: how we approach virtualizing security, and how virtualization itself changes the way we approach information security. All of that changed when I was trampled over by The Four Horsemen Of the Virtualization Security Apocalypse.

One of the big topics at last week's Black Hat and Defcon security confabs was virtualization security, but few speakers talked about what is really important: how we approach virtualizing security, and how virtualization itself changes the way we approach information security. All of that changed when I was trampled over by The Four Horsemen Of the Virtualization Security Apocalypse.First, when it came to virtualization security, there were talks covering Layer 2 VLAN attacks (if you've been around awhile, you didn't miss much from this talk, folks), Taking the Hype Out of Hypervisors, Subverting the Xen Hypervisor, Detecting & Preventing the Xen Hypervisor Subversions, and Bluepilling the Xen Hypervisor. Of these talks, I found Tal Garfinkel's "Taking The Hype Out Of The Hypervisor" the most interesting as it spoke to some of the important operational issues involved in an ever-changing virtual environment.

As for the presentations covering hypervisor subversion and so called "bluepilling," I found to be more sizzle than the actual security steak. I've said it previously (and have the flame-mail to prove it) that while attacking the hypervisor itself is important, and no doubt sexy, at the end of the day it's only sizzle. Which is to say that while it smells and sounds good while cooking, you end up pouring it down the drain before you sit down to eat.

Which brings us to the main course, Christofer Hoff's The Four Horsemen of the Virtualization Security Apocalypse. This was the only talk that encapsulated the challenge of how security technologies deployed as virtual appliances or as software agents must be accounted for in capacity planning; how technologies such as VMotion and live migration can break all of those mature switch, port, and NIC groupings we've come to love (or, at least understand and manage) in the physical environment; the collapsing of multiple security technologies (ID/PS, anti-malware, etc.) into a single virtualized God Box. That last point, Hoff illustrated as collapsing security decisions to a few checkboxes clicked by newly formed "virtualization administrative" teams who have little in the way of security chops.

What does all of this mean? If you understand how networks and security are architected today, you understand that virtualizing security will add to the costs (at least temporarily) of your existing security efforts. Hoff said it best during his talk:

You're going to have to deploy the same security software agents to each individual virtual machine. And also add, at a minimum, a virtual appliance to each physical host. And you're most likely going to have to maintain the hardware appliances already protecting the physical networks.

Now, when you consider how many VMs the typical company is trying to cram on a single physical box, you see just how much security software you're asking an individual box to run. And it's in that way that virtualizing security will add complexity and cost organizations more, especially in the short term -- if they're doing it right.

Did I walk away from the Four Horsemen talk thinking that virtualization was too costly, or that hypervisors would usher in the IT Armageddon? Absolutely not. It's clear that, eventually, security vendors will adequately virtualize their gear and solve many of these challenges. But that may be well over a year away.

What I did walk away from the presentation understanding is that it's dangerous to rush your virtualization efforts from the lab and testing environments to core production systems. And that highly agile and dynamic environments are possible today, and will be even easier to attain (as toolsets mature) in the near future.

But to get there, you had better plan and architect your virtualization efforts carefully -- or you just may find yourself overrun by the Four Horseman of the Virtualization Security Apocalypse. In fact, I'm sure there are security managers out there with four sets of hoofprints engraved permanently in their backs already.

If you'd like to see Hoff discuss the topic in his own words, check out this video interview conducted by Dark Reading's Kelly Jackson Higgins.

Comment  | 
Print  | 
More Insights
Comments
Threaded  |  Newest First  |  Oldest First
Why Vulnerable Code Is Shipped Knowingly
Chris Eng, Chief Research Officer, Veracode,  11/30/2020
Register for Dark Reading Newsletters
White Papers
Video
Cartoon Contest
Current Issue
2021 Top Enterprise IT Trends
We've identified the key trends that are poised to impact the IT landscape in 2021. Find out why they're important and how they will affect you today!
Flash Poll
Twitter Feed
Dark Reading - Bug Report
Bug Report
Enterprise Vulnerabilities
From DHS/US-CERT's National Vulnerability Database
CVE-2018-21270
PUBLISHED: 2020-12-03
Versions less than 0.0.6 of the Node.js stringstream module are vulnerable to an out-of-bounds read because of allocation of uninitialized buffers when a number is passed in the input stream (when using Node.js 4.x).
CVE-2020-26248
PUBLISHED: 2020-12-03
In the PrestaShop module "productcomments" before version 4.2.1, an attacker can use a Blind SQL injection to retrieve data or stop the MySQL service. The problem is fixed in 4.2.1 of the module.
CVE-2020-29529
PUBLISHED: 2020-12-03
HashiCorp go-slug before 0.5.0 does not address attempts at directory traversal involving ../ and symlinks.
CVE-2020-29534
PUBLISHED: 2020-12-03
An issue was discovered in the Linux kernel before 5.9.3. io_uring takes a non-refcounted reference to the files_struct of the process that submitted a request, causing execve() to incorrectly optimize unshare_fd(), aka CID-0f2122045b94.
CVE-2020-17527
PUBLISHED: 2020-12-03
While investigating bug 64830 it was discovered that Apache Tomcat 10.0.0-M1 to 10.0.0-M9, 9.0.0-M1 to 9.0.39 and 8.5.0 to 8.5.59 could re-use an HTTP request header value from the previous stream received on an HTTP/2 connection for the request associated with the subsequent stream. While this woul...