A Stuxnet Apocalypse?

Tuesday, September 13, 2011

Infosec Island Admin

7fef78c47060974e0b8392e305f0daf0

STUXNET-APOCALYPSE! Says the Israeli Who Doesn’t Have Nuclear Silo’s and Bombs…

My first reaction to this posting online came from the original source that I saw it on, namely the Greg Evans website where he had scraped the story from another source (never mind why I was there).

Since then, this story ended up on the headlines section of Infosec Island and once again my reaction is HOLY WTF? How does this get into the news cycle at all without people calling it into question rather vociferously?

image

From Infosec Island:

Tomer Teller, a security evangelist for Check Point, warned of the likelihood that the Stuxnet virus could be adapted to undermine systems that control nuclear missile arsenals.

Teller made the prediction at a conference in Sydney, Australia last week hosted by Check Point.

“Nuclear warheads are controlled by computers so if someone managed to slip a worm inside a facility that will reach the warhead component, they could launch it and than aim it back at the country’s facility… Stuxnet is the first cyber weapon that could cause major disruption” Teller explained.

Teller indicated he has conducted a detailed analysis of the Stuxnet code, and given the size and complexity of the file, Teller believes it is likely that a successful attack would require utilization of an insider.

“This is a huge file, it’s 1 megabyte [MB] of code and I respect the skill required to engineer that code as it is very complex,” Teller said.

The most likely avenue of for the attack, Teller postulated, would be through the use of a tainted USB drive.

“In order to get something trusted by Microsoft, you need to get those exploits signed… What we think happened is that an insider broke into JMicron, a chip manufacturing company based in Taiwan, as there is a computer at that office which is dedicated to signing these Microsoft drivers,” Teller said.

So I decided to talk to a source of mine who is in the know about most things nuclear. I asked him if indeed the supposition I had that Stuxnet would be pretty much useless in a Silo because of the way the systems were designed to be ultra redundant as well as segregated within that redundancy.

What I got back was the following:

OK, at a high-level:

The ~concept~ of a Stuxnet and Nuclear Silos really only applies to operational readiness. With the exception of Pakistan no ~known~ nuclear player doesn’t already implement a variant of the PAL and CMS systems in their launch controls.

- The levels of redundancy involved for both the ability to launch, preventing launch, arming, and self-destruction at exist in nuclear silos is quite possible the most perfectly decision/failure tree designed/redesigned systems known to man. This wasn’t due to a one-time effort, its been decades in the making.

- The availability of the hardware, software, network access, peripherals access, etc. is SOOOO ridiculously limited that ~development~ of such a tool would be purely speculative. This isn’t SIPREnet or CRONOS or NAUTILUS we’re talking about here. The levels of control and network isolation, and again redundancy, within these installations at major players is appropriately absurd.

- In terms of ancillary systems and operational readiness there are two ways to look at it. Could something cause a scare a a general shutdown due to FUD? Yes. Could something be used for ~press~ purposes to indicate a lack of operational readiness. Yes. The reality is that those who would be deterred in a MAD scenario know better.

- We’re talking submarines too, a WHOLE different level of player and communication and control systems. The likes of which are even further unknown and more specific.

- W/ submarines there is a different sort of risk because some major players (namely the UK) don’t have the same level of controls on subs so a rogue commander w/ a key could start a launch. If talking StuxNet like situations, you’re not reducing their readiness so easily. It is know in these cases from various leaks that the control pathways are also isolated by design and ~MECHANICAL~ to what might even be considering a fault.

- If you’re a Nation-State that has been tutored in the ‘art’ by the US or USSR you’re also likely to have bizarre levels of controls like fan speed detectors, temperature detection, computational state metrics, etc. to show the slightest change in behavior. Some of this was designed around the idea of more traditional things like a Y2K bug maybe affecting ~some~ system. This that are, to any person who designs hardware, known to be fairly absurd and unnecessary but indeed they would trigger an alarm to ~look~ at something.

- All that said.. could a Stuxnet like system affect the ~production~ and ~development~ of a nuclear weapon? Yes and no. Yes as in it certainly could screw up enrichment (obviously) but wouldn’t screw up the end-result (hence why Stuxnet was found in the first place).

So, once again, I call shenanigans on Tomer and this little story. At worst, if there were a Stuxnet variant that were worked out AND carried into a silo it would cause (maybe) a failure to launch, it is much more likely that Stuxnet and variants would instead be more used (as it was in Iran) to manipulate the production of fissile material so as to have a weapon that would not actually work once launched in the payload vehicle. This story though, and the way it has been put out there by Tomer Teller, makes it sound as though imminent failure could happen to nuclear bombs and this is just not right.

Even more ludicrous is the idea that a Stuxnet variant could infect a system and cause the payload to come back from where it was launched, in effect changing the target coordinates.

THIS would be more along the lines of just some malware, not Stuxnet that would infect specific systems in guidance on board the payload vehicle, and that is a totally different animal from Stuxnet. Indeed, this would be a completely different effort altogether and would require something else completely.

You see, the point of Stuxnet was that it was manipulating PLC code to specific PLC’s Tomer, what you are talking about would be something completely different.

Go back to firewall evangelizing and leave the nuclear weapons alone.

K.

Cross-posted from Krypt3ia

Possibly Related Articles:
7190
Network->General
Military
Insider Threats malware Defense Attacks Stuxnet National Security Nuclear
Post Rating I Like this!
C4363f41d25c216c53c8d71a1ac44a90
Matthijs R. Koot Thank you for correcting/debunking! Useful contribution.
1316028105
The views expressed in this post are the opinions of the Infosec Island member that posted this content. Infosec Island is not responsible for the content or messaging of this post.

Unauthorized reproduction of this article (in part or in whole) is prohibited without the express written permission of Infosec Island and the Infosec Island member that posted this content--this includes using our RSS feed for any purpose other than personal use.