Showing posts with label security vulnerability. Show all posts
Showing posts with label security vulnerability. Show all posts

Friday, September 18, 2015

August news roundup: Debunking myths and unveiling threats to critical infrastructure

As the summer enjoyed its last hurrah, industrial cybersecurity became a hot topic in the news. In August, we saw the U.S. Department of Homeland Security working to increase security measures with a new committee, and the National Institute of Standards and Technology’s put out a new proposal to improve international standards in cyberspace. Underlying these developments are the ever-present myths and real threats against our critical infrastructure. For details on these stories and more, check out our monthly snapshot of important industrial cybersecurity news below.

Despite widespread concern about cyberattacks on industrial control systems, IT security models continue to use outdated methods of cybersecurity based on several unfortunate myths. The belief that firewalls provide adequate protection of our critical infrastructure is one myth that perpetuates despite proof to the contrary. The fact is, firewalls offer only a small amount of protection, but not enough to protect our irreplaceable power grids or the many lives that could be affected by an industrial cyberattack.

The U.S. power grid is long overdue for a cyberattack, according to Forbes contributor and CEO of think tank Lexington Institute, Loren Thompson. The security community has noticed an increase in the number of attacks on industrial control systems used to operate the power grid, a majority of which were cyber-related. Thompson noted several reasons for this increase including the grid’s numerous vulnerabilities, consistent oversights in its regulatory structure and a lack of financial incentives to encourage security investments in the industry.

The U.S. Department of Homeland Security is creating a new committee to boost digital defenses for the industrial sector. The reason behind this decision is the increasing risk of cyberattacks to critical infrastructure sites, especially as electric grids are getting smarter. Homeland Security Secretary Jeh Johnson called for the panel to identify how well the department’s “lifeline sectors” are prepared to deal with threats and recover from a cyberattack. The committee is also tasked with providing recommendations for a more unified approach to state and local cybersecurity.

A Critical Time for Critical Infrastructure (Light Reading, Aug. 13, 2015)
According to a recent Intel Security and Aspen Homeland Security Program report, operators of critical infrastructure are over-confident in their ability to defend against attacks and misunderstand the scale of the current threat environment. In North America, this reality is one of the forces behind the North American Electric Reliability Corporation's (NERC) Critical Infrastructure Protection (CIP) requirements. NERC CIP Version 5 calls for utilities of all sizes to meet new cyber security protection requirements and has a compliance deadline of April 2016.

The National Institute of Standards and Technology recently drafting a new proposal, which includes four broad objectives for the government’s pursuit of international standards in cyberspace: improve national and economic security, ensure standards are technically sound, support standards that promote international trade, and develop standards in tandem with industry to boost innovation. If fully implemented, NIST declares the guidance will “enable a comprehensive United States cybersecurity standardization strategy.”


Interested in reading more cybersecurity news? Check out last month’s news roundup.

Friday, November 14, 2014

Patching critical infrastructure: What Bash means for ICS security


On September 25, a bug deemed “Shellshock” was discovered in Bash, a command shell on Unix, Linux and Mac OS X operating systems that is used heavily in scripting and for communication between one program and an operating system for certain kinds of services. Much of the media attention has centered on how Shellshock is a threat to cybersecurity in general, lumping all practice areas under one umbrella. Since critical infrastructure networks are much more difficult to patch and update than corporate networks, many control system security practitioners are wondering what, specifically, are the implications of Shellshock on control system networks, and what can we do to protect against these vulnerabilities. Surprisingly, a recent search of the Internet yielded no clear summary of the impacts of Shellshock on control networks specifically, hence this posting.

In order to be affected by Shellshock in the first place, a device must have Bash installed. Since Bash is not a standard Windows component, it’s unlikely that Windows systems will be vulnerable unless the program was installed for some reason. Mac OS X and Linux both use Bash heavily, and if any non-Linux Unix is running on a network, then Bash is also very likely deployed somewhere within that system, if not everywhere.

Here are some examples of how Bash might affect particular systems:
·         Web servers that use CGI scripts, like Apache, transfer information like the “user agent” string directly to Bash. That string can be set in some browsers, and is easily set in many popular command-line Internet tools. The exploitation of these vulnerable web servers is trivial. This compromise can be accomplished from any IP address that has access to send a web request into the vulnerable server.
·         Most Mac/Linux/Unix gear that uses DHCP on an industrial network is ripe for an attack from the local network. While most critical control systems have been drilled into using static IP addresses rather than DHCP for exactly this reason, some sites still have equipment using DHCP. If a hacker can get his hands on a laptop or other computer connected to a control network and can turn on a DHCP server on the machine, all bets are off.
·         Every device that runs Linux or some other Unix derivative with Bash installed, and has a Web user interface, is vulnerable. This includes a lot of networking gear, firewalls and even some RTUs, PLCs and other equipment. Figuring out which of these firmware-based systems have Bash installed is problematic in itself. Vulnerable equiment can generally be hacked by any machine, which can send a message to the Web server.

Software and firmware updates should of course only be applied to equipment on control system networks after thoroughly researching a patch’s reliability. In principle, while patches are being tested, or in some cases still being developed, all vulnerable DHCP, web and other functionalities should be disabled. This is easier said than done since it is not even clear which devices with embedded Unix-based operating systems have Bash installed at all, not to mention that some of the affected functions may be essential to the current design and operation of the control system.

This is just another example of why many control system vendors deploy Unidirectional Security Gateways. The gateways replicate servers to external networks to provide seamless, safe integration of control system networks with corporate and other networks. IT teams can then feel free to install the latest, up-to-the-second updates to all equipment on corporate networks, including the replica servers, without putting critical operations at risk.

The takeaway here is nothing new, and yet, is underlined with each new serious vulnerability. And all software has bugs, some of which are security vulnerabilities, meaning all software can be hacked. Industrial users should deploy hardware-enforced, stronger-than-firewalls perimeter protections to ensure that the next Shellshock, or dozen Shellshocks, do not expose critical infrastructures to attacks from corporate networks, and from the Internet beyond those networks.

To find out more about ICS security solutions, check out our products page here.