The infosec world is still buzzing about the recent hack of
Sony and how completely their information systems were owned by the attacker.
Researchers will be pouring over this attack for a while trying to figure out
how it all happened and what could have been done to prevent it.
The Hack
The Sony hackers took, among other things, valuable
intellectual property (high quality digital copies of yet to be released
films), thousands of internal documents (one source said 1 terabytes worth),
and system access information to include log-in credentials and passwords. The
attack was so thorough that Sony was forced to shut down all computer systems
for days while they went through and inspected/cleaned every machine.
The TWITERVERSE is all agog, of course about #SonyHack. I’ve
made my own small contribution (sans hashtag, of course). The one tweet of mine
that I would like to look at in more detail came on Saturday:
“I guess we should be glad that Sony didn't make dangerous chemicals or
supply something really important....”
XYZ Chemical Company
So what would have been different if it had been XYZ
Chemical Company that had attracted the ire of the GOP attackers? Other than
differences in what IP was stolen, probably not much in the immediate results
of the attack. The longer term results could have been much more serious to
communities around XYZ plants and possibly to the nation.
Intellectual property theft? Well, formulations come to mind
first. The wholesale release of formulations would give every competitor here
and around the world a leg up in understanding how XYZ makes their products.
Even for fully mature product lines, the little tweaks to the formulation a
company makes can make a major difference in pricing and performance. This
could make for a serious change in competitive status.
The public publishing of the formulations could have other
consequences as well. There are a number of chemicals being used in the
chemical industry that have raised concerns among environmental activists. If a
formulation list included such chemicals the company could come under
additional scrutiny from these activists, potentially leading to image issues.
Worse yet, if any of the chemicals being used were on the lists of chemicals
offensive to various environmental wacko fringe groups, the facility could
become a target for varying degrees of physical attacks.
Control System
Insecurity
From a security perspective the IP that is of more concern,
however, would be process flow diagrams and control system working documents.
Those pieces of information would make it much easier for an attacker that has
gained access to the control system to figure out how to manipulate the
controls in that system in a manner that could disrupt the facility in the
worst possible way.
Since most corporate types use a single set of login
credentials across the entire spectrum of company computer systems, it is very
likely that even if the control systems were not directly breached in the
attack, there would be useable log-in credentials for the control systems to be
culled from the corporate log-ins. Once an attacker can gain legitimate access
to a control system, the various ‘insecure-by-design’ problems become very
large vulnerabilities.
If the control system had been directly breached as well the
problems become much worse. The shut down and start-up of a chemical control
system, particularly with continuous chemical processes like refineries, is not
something that is done lightly or quickly. And searching control systems for
logic bombs, back-doors, and forms of re-programing is a much more time
consuming task than for an IT system of comparable size. The larger the number
of PLCs involved the more difficult the task becomes.
Physical Insecurity
One set of computer data files that no one wants to see
released are physical security plan files. Password and log-in credentials can
be changed fairly easily, but making significant changes to a site security
plan are very expensive and time consuming. The public availability of site
security plans would make the facility much more susceptible to physical attack
or theft of chemicals.
While there are some people that claim that the Sony
attackers had at least some physical access to the corporate computer system
that would not necessarily mean that that access included detailed information
about the site security plan. Most folks entering a facility are aware of only
a small part of the over-all security plan for the facility.
For many chemical manufacturing facilities a major source of
security concerns can be found in the computer systems of the order handling
folks. The timing and sourcing of inbound and outbound chemical shipments makes
them that much more vulnerable to off-site attack. Those attacks may allow for
inbound shipments to be intercepted and converted to a method of attack (VBIED)
or simply diverted to some other nefarious or illegal end.
The Big Picture
The major take away from this attack is that this was no
longer a hack to deface a web site, steal money or credentials. This was a full
scale attack on a company with the intent of seriously harming the company for
some as of yet unknown reason. The scope and scale of the attack is unlike
anything we have seen in the corporate world. Unfortunately, as with all things
cyber, it will undoubtedly set the new gold standard for serious attacks as it
give other groups and hackers ideas about what can be accomplished with a
successful cyber attack.
1 comment:
So, I like the theme and intent of this post, but I'd like to disagree with the IP/data comparison. Modern hackers exploit ICT's (cyber systems) to take advantage of business value chains to alter value or create other value for their own use. In the case of Sony, the value produced by the value chain *is* IP and data-centric. In the case of Chemical Plants, the value produced is physical, by way of data. An equivalent hack to Sony is one where the hackers would have gone after physical (or potentially physical) results, where the IT and data would have been lumped in with ICTs as a means to an end.
This might seem like splitting hairs, but not making this distinction hides/obscures the actual risk being faced by us from thoughtful, adaptable, persistent threats.
Post a Comment