Wednesday, January 19, 2011 - 7:21 AM

I asked Jay Holcomb what he made of the big article about Stuxnet that ran in Sunday's New York Times. Here is his response.
By Jay Holcomb
Best Defense infowar article criticI enjoyed reading the New York Times article, "Israeli Test on Worm Called Crucial in Iran Nuclear Delay" published Jan. 16. Everyone seems to agree that this was by far the most complex cyber event ever seen in the wild. By complex I'm referring to the number of technical features, such as zero-day exploits, industrial control system expertise, intelligence on target configurations, number of cyber exploits used on the target, such as root kits, botnet-type command and control, user view manipulation, etc.
I believe that the more media exposure we can generate from complex cyber events like this one, the better. However, I still believe we are missing the bigger picture with regard to these types of complex events. While I realize many folks really want to know where the Stuxnet package originated, I propose that we should be spending as much (or more) time looking around at what these events mean today, and in the near future, with regard to our cyber exposure -- federal/state/local government resources, critical infrastructure, civilian industries, and even our own personal exposure.
I agree with Mr. Langner's quote in the article, referring to the Stuxnet package, that, "It's like a playbook.... Anyone who looks at it carefully can build something like it." Langner makes an important statement that I have not seen many people outside the industrial control system and cybersecurity industries mention or highlight. We can assume it is not only nation-states that are looking at events like these; terrorists and common criminals are most likely very busy right now looking at this too!
Many of the items highlighted in the article potentially read like a fortuneteller's glass ball: "The vulnerability of the controller to cyberattack was an open secret. In July 2008, the Idaho lab and Siemens teamed up on a PowerPoint presentation on the controller's vulnerabilities that was made to a conference in Chicago at Navy Pier, a top tourist attraction." This is not unusual, as significant vulnerabilities in software will often be publicly known. The vulnerabilities often are not addressed until (what seems like) enough public pressure is applied for a fix/patch to be produced and/or applied. While I have no specific information on "Smart Meters," recent articles which point out potential security concerns related to the deployment of "Smart Meters" make me wonder whether we're not looking into a fortuneteller's glass ball. I'll include some reference links about this at the bottom of this note.
One final thought: While the Stuxnet event and associated reports have generated some public media exposure on complex cyber events, I find myself looking back on a report released by the U.S.-China Economic and Security Review Commission, dated Oct. 9, 2009, which does a great job explaining a very complex cyber intrusion -- I wonder if that was a cyber building block to our current Stuxnet discussion?
Smart Meters:"Money trumps security in smart-meter rollouts, experts say"
"Security Pros Question Deployment of Smart Meters"
"More Researchers Point to Smart Meter Security Holes"
"UK business electricity supplier reaches 12,000 smart meter installations"
"PSO says Owasso customers will be converting to smart grids as part of pilot program"
"The three stages of SmartMeterTM technology"
The U.S.-China Economic and Security Review Commission, "Capability of the People's Republic of China to Conduct Cyber Warfare and Computer Network Exploitation" (p. 59, "Operational Profile of An Advanced Cyber Intrusion"), Oct. 9, 2009
First things first: If we get hit by a Stuxnet, we go to the PLC vendor and ask for a reflash of all infected processors. Iran cannot do this with Siemens today, because they bought the equipment under false pretenses, or whatever.
The way I see it, from the perspective of an industrial control engineer, Stuxnet is not possible without a) cooperation of the vendor (Siemens and probably others) and b) extensive testing such as that in Dimona described in the article.
With respect to greater exposure (civilian industry, etc.), the systems' first line of defense is that they are generally well-secured physically (locked in control cabinets, in areas that are themselves access-controlled) and almost never connected directly to the internet.
When we talk about vulnerabilities to things like smart-meters, the discussion should also be extended to scenarios like a terrorist-mechanic maliciously reflashing your car's engine control module while it's in the shop, to accelerate at an inopportune time. It's possible!
The steady hoofbeats of a manufactured crisis
More & more puff pieces about threats of cyberwar and not much substance. Change the cyber to any other common sector of the economy or even general subjects and it reads the same- Currencywar, Biowar, PRwar, Crimewar the list could go on and on. Let's stand up a whole bunch of other commands to protect against the threat. North Korea into drug trafficking, counterfeiting, and other criminal enterprises. That's good for at least 2 or 3 new commands.
Cyberwar is the perfect make work project, a marriage of the two biggest boondoggles with poor track records: military contracting + IT contracting. Anyone questions the rationale, and it's "Security, security, security". Where's the national security threat in parking meters? People getting tickets for time they paid leading to the Great Parking Uprising of 2013 which allowed Canada to swoop in? If I leave my house unlocked when I go out that doesn't mean that the Fulda Gap is unguarded and the soviets can roll right through.
In any case, I'll sleep better knowing the fine men of CyberCommand ("Cybermen*"?) stand ready behind the CRT to do violence against any and all enemies.
* perhaps it's not a coincidence that gold prices at an all time high
Many are connected to the internet for sending out e-mail alerts, remote access of data, etc.
But the greater exposure is that they are programmed in many cases by developing the control sequences on a laptop and then the PLC itself is programmed by that laptop. Which is usually connected to the internet. So, the laptop gets infected from the internet and then when it is connected to the PLC the virus gets access.
And that's just one degree of separation. Five more and the PLC is connected to Kevin Bacon.
What's made the AK-47 so popular?
A bit OT, and I gather that nuclear** enrichment requires computerized operation, but Stuxnet illustrates why we shouldn't use complex technology unless it's absolutely necessary.
It’s why I don’t trust computerized voting. I’d much prefer paper ballots and clear plastic boxes.
(** PS: Did the technician ever stay in the room with you when you got an X-ray? That’s all you need to know about nuclear energy.)
X-ray vs particulate contamination
The conflation of whole-body radiation (say high altitude flight) with the dusting of radionucleotide particles is one of the Big Lies of the 'friendly atom' age. We didn't double the average dose of whole-body radiation because the x, gamma and alpha keep bouncing around. It's the fallout, local emitters counting half-lives, that now rival the natural radiation emissions.
Both X and alpha are well tolerated, in the proper medical context. But an alpha emitter that mimics calcium and creates highly localized damage in my bone marrow is 'inside the wire', so to speak.
The experienced remedy for any plutonium contamination (say a machinist suffering a metal sliver) in 1945 was high amputation, above the injury. We don't want to know how that experience was gained. Lung amputation isn't an option: Something to think about, since tobacco leaves concentrate radioactive lead from natural radon decay, and can double a smokers annual radiation exposure in a really bad way
Are we really crowing about US joint ventures with that N-weapons production facility? Talk about crossing a line.
(6)
HIDE COMMENTS LOGIN OR REGISTER REPORT ABUSE