Showing posts with label Intrusion-Prevention Systems. Show all posts
Showing posts with label Intrusion-Prevention Systems. Show all posts

March 13, 2013

Balancing Cybersecurity And Citizen Freedom


There is a very interesting discussion of the protection of Federal Networks and the Fourth Amendment in “Cybersecurity, Selected Legal Issues,” Congressional Research Service (CRS) Report for Congress (3 May 2012). 

The Department of Homeland Security (DHS) in conjunction with the National Security Agency (NSA) rolled out EINSTEIN, an intrusion detection system (IDS) in early iterations, and later an intrusion prevention system (IPS) at all Internet points of presence (POPs) for the government.

The system works through copying, storage, and deep packet inspection of not only the metadata for addressing information, but also the actual contents of the flow. This handling is necessary in order to identify suspicious malware signatures and behavior and alert the United States Computer Emergency Response Team (US-CERT) in order to block, quarantine, clean, and respond to the attacks and share information about these.

However, the civil liberties and privacy issue with EINSTEIN is that according to the Fourth Amendment, we are protected from unreasonable search and seizures. Thus, there are concerns about the violation of the Fourth Amendment, when DHS monitors and inspects addressing and content of all email and Internet communications to and from federal agency employees and the public–including not only from government email accounts and systems, but also from private email accounts such as Yahoo and Gmail and social media sites like Facebook and Twitter.

 The justification for the use of EINSTEIN includes:

1.    The government cannot reasonably get warrants in real time in order to safeguard the federal network and systems at the speed that the attacks are occurring.

2.    The government places banners and user agreements on all Federal networks notifying users of monitoring, so there is no expectation of privacy in the communications.

3.    The monitoring is conducted only for malicious computer activity and not for other unlawful activities—so “clean” traffic is promptly removed the system.

4.    Privacy protections are ensured though review mechanisms, including Attorney General and Director of National Intelligence (DNI) reporting to Congress every six months and a sunset provision requiring monitoring reauthorization every four years.

This tension between monitoring of Federal networks and traffic and civil liberties and privacy is a re-occurring issue when it comes to cybersecurity. On one hand, we want cybersecurity, but on the other hand, we are anxious about this security infringing on our freedoms---whether freedom of expression, from search and seizure, from surveillance, or from potentially costly regulation, stifling innovation, and so forth.  It is this tension that has stalled many cybersecurity bills such as the Stop Online Privacy Act (SOPA), Cyber Intelligence Sharing and Protection Act (CISPA), The Computer Security Act of 2012 and more.

In the absence of a clear way forward with legislation to regulate and enforce, or incentivize, standards and best practices for cybersecurity, particularly for critical infrastructure protection, as well as information sharing, the White House released Presidential Policy Directive/PDD-21 on Critical Infrastructure Security and Resilience to establish DHS and other federal agency roles in cybersecurity and to manage these on a risk-based model, so that critical infrastructure is identified, prioritized, assessed, and secured accordingly.

While PDD-21 is a step in the right direction, it is an ongoing challenge to mediate a balance between maintaining our values and constitutional freedoms, while at the same time securing cyberspace.

One thought is that perhaps we can model cybersecurity after the Posse Comitatus Act of 1878 that separated federal military from domestic national guard and law enforcement powers. Using this model, we can create in cyberspace a separation of cybersecurity from our borders outward by the federal government, and within the domestic private networks by our national guard and law enforcement.

Thus, we can create stronger security radiating out at the national periphery, while maintaining our important freedoms within, but always working together to identify and neutralize any and all threats to cyberspace. ;-)

(Source Photo: Andy Blumenthal)

Share/Save/Bookmark

January 29, 2008

Intrusion-Prevention Systems and Enterprise Architecture

Firewalls have traditionally been used to “wall off” the enterprise from computer attack, but now intrusion-prevention systems are augmenting the organization’s defenses.

The Wall Street Journal, 28 January 2008 reports that “intrusion prevention systems promise an even smarter defense” than firewalls.

Firewalls are intended to keep intruders out. However, because certain traffic, such as email, needs to get through, holes or open ports allow in traffic that can carry viruses or malware into the network.

Intrusion-prevention systems work differently—they don’t wall off the enterprise networks like firewalls, but rather like a metal detector, they filter or scan every piece of traffic entering the organization for suspicious activity, and reject any item that is identified as a threat.

According to Wikipedia, Intrusion prevention systems (IPS)... [are] a considerable improvement upon firewall technologies, IPS make access control decisions based on application content, rather than IP address or ports as traditional firewalls had done.

Intrusion-prevention systems can be hardware that is physically attached to the network or software that is loaded onto individual computers.

Are intrusion-prevention systems really necessary?

Yes. “According to the Computer Security Institute 2007 Computer Crime and Security Survey, the average annual loss suffered by U.S. companies from computer crime more than doubled last year to $350,424 from $168,000 in 2006. And these reported losses tend to underestimate the number of attacks.”

Gartner analyst recommends antivirus on PCs and an intrusion –prevention system on the network.

Are there any problems with intrusion-prevention systems?

One of the biggest issues is false positives, which if not adjusted for will block desired incoming traffic. One way to handle this is to use the intrusion-prevention system to “detect threats and flag them,” rather than simply block them altogether. Additionally, the organization can adjust the filters that they may not need. This is the tuning required to ensure performance in terms of network speed and an appropriate level of filtering.

If your organization is not using an intrusion-prevention system, this is something your enterprise architecture needs to plan for and implement ASAP.


Share/Save/Bookmark