The Federal CISO Dilemma. You have to do FISMA. You must defend against cyber threats.

Transcription

1 The Federal CISO Dilemma You have to do FISMA. You must defend against cyber threats. October 2012

2 Executive Summary Federal CISOs face a unique cyber security challenge copious amounts of regulatory compliance paperwork. In fact, agencies spend more than one in four cyber dollars on FISMA paperwork* well more than $1 billion each year. At the same time, the Federal government responded to more than 106,000 cyber attacks in 2011 including cyber exploits that injected viruses, stole information, and disrupted Federal network operations. While agencies must comply with FISMA, the regulations focus on complying with a narrow set of security and privacy controls. What these regulations don t do is actually protect government networks from Advanced Persistent Threats (APTs) or other nefarious activity. The intent of these regulations is good there should be security and privacy control regulations, and of course we need minimum levels of protection across the board for all Federal agencies. But the reality is that every dollar spent on compliance leaves CISOs with fewer remaining resources to combat APTs and advanced malware, so every dollar counts. Fortunately, there is a solution to this dilemma. Many Federal CISOs are deploying game-changing technologies that offer robust protection without breaking their limited budgets. This white paper is divided into three sections for your convenience: Section I, page 1, focuses on the anatomy of an APT, featuring an in-depth look at how the enemy gains access to a network, including the end goal of an attack. Section II, page 4, focuses on the key characteristics of a strong defense system that agencies can put in place to combat sophisticated APTs. Section III, page 5, focuses on a broadly adopted game-changing security technology that can help Federal CISOs solve their unique dilemma deploying robust cyber protection with limited resources. *MeriTalk The FISMA Secret II

3 Section I: Anatomy of an APT APTs: Taking Advantage of Gaping Holes in Agency Network Defenses Spear Phishing Entices You work at Agency X. You receive an . A quick glance tells you it s from Dave Richards, your colleague down the hall. The flags a new expense approval process and includes an attachment. You know Dave, you care about your expenses, you click on the attachment. Spear Phishing Penetrates This isn t obvious spam from a prince asking you for your credit card number it s spear phishing, targeted to affect specific individuals. By mining available online data, the personalization and impersonation used in spear phishing s can be extremely accurate and compelling. These attacks are generated by cunning, agile, and quick adversaries who literally trick well-meaning employees to invite the attackers in. All employees, even security staff, can fall prey to well-crafted spear phishing attacks. As long as your organization accepts from the Internet, spear phishing will remain a clear and present danger. and Spear Phishing Delivers a Payload Once an employee clicks a link or opens an attachment, the enemy establishes a foothold in the network setting the stage for the next phase of the attack. Now it s time for perpetrate deploying the malicious code, moving laterally within your network, and stealing your critical data. For most APT attacks, the objective is to gain longterm access to your sensitive networks, data, and resources. Using spear phishing to penetrate and advanced malware to perpetrate isn t an anomaly; it represents a clear shift in the current approach of adversaries. Increasingly, criminals are moving from massive phishing attacks to spear phishing on a much smaller, more targeted scale because it has proven very effective. Here are some of the key characteristics of advanced, targeted spear phishing attacks: Blended/multi-vector threat: Spear phishing uses a blend of spoofing, zero-day application exploits, weaponized attachments, dynamic URLs, and drive-by downloads to bypass traditional defenses. Leverages zero-day vulnerabilities: Advanced spear phishing attacks leverage zero-day vulnerabilities in browsers, plug-ins, and desktop applications to compromise systems. Multi-staged attack: The initial system exploit is the first stage of an APT attack that involves further stages of malware outbound communications, binary downloads, and data exfiltration. Lack characteristics of spam: Spear phishing s are targeted, often on an individualized basis, so they don t bear a resemblance to the highvolume, broadcast nature of traditional spam. This means reputation filters are unlikely to flag these messages, minimizing the likelihood of spam filters catching them. Read the FireEye report, Top Words Used in Spear Phishing Attacks, for more information on the latest trends in spear phishing jargon. 1

4 APTs: How the Enemy Gains Access You ve seen that today s threats can fool even security-savvy professionals into inviting criminals into otherwise secure networks, but the question is how? Understanding just how easy it is for the enemy to convince a user to let his or her guard down and then execute a successful spear phishing attack is critical to devising better ways to secure our government networks. As an industry leader in detecting and stopping sophisticated threats, FireEye keeps close tabs on the ever-changing tactics enemies use to gain access. The following are just a few real-world examples FireEye collected in recent months: Malicious legal agreements: A threat actor compromises a law firm to obtain a recent legal agreement used between a vendor and a Federal agency. The attacker then crafts a fake posing as the vendor and sends it to three key Federal personnel. This includes a weaponized PDF of the revised legal agreement recently discussed between the agency and vendor. Once the Federal personnel open the attachment, the compromised system slowly contacts the attacker s command and control infrastructure, which looks like a benign set of dynamic DNS servers that host cute animal pictures. Malicious newsletters: Five recipients inside a U.S. agency routinely receive PDF-based newsletters around the same time each day. Knowing this, the attacker sends an executable file disguised as a fake PDF newsletter from a spoofed news source around newsletter time. When opened, the fake newsletter appears, while the malware installs a persistent backdoor inside the workstation, which appears to be a fake instance of the Adobe Update service. Then, the workstation beacons out to fake news websites controlled by the attacker, and the traffic returned to the workstation appears to be dynamic Web content required to legitimately render the fake newsletter. Malicious conference forms: The threat actor sees that a Federal employee recently registered for a conference on a topic of interest. As such, the actor s the attendee a fake conference registration form that appears to come from the conference organizer. When the attendee opens this malicious Word document, the malware immediately installs three different hidden services and displays the fake form to the user. Hours later, the attendee s compromised system slowly beacons to a variable command and control infrastructure over SSL, ultimately installing a remote access tool so that the threat actor maintains long-term persistence on the system and can exfiltrate future conference data from the attendee s system after the conference. Malicious anything: You name it, hackers will try it. Criminals step into the middle of conversations and send files that look legitimate and are relevant to the existing chain. Threat actors send weaponized Excel spreadsheets that look like purchase order agreements between two related program managers in a U.S. defense industrial base. A user clicks, and the beacons, call-backs, and login credential harvesting begins. The list goes on, and government organizations must remain vigilant. 2

5 APTs: A Closer Look at How They Deliver Their Payload Talking about APT characteristics is one thing seeing what perpetrators do once they get in is another. That s why FireEye deployed its Malware Protection System into a Federal organization to monitor and block inbound malware infection attempts as well as outbound malware callbacks. FireEye was deployed behind existing layers of security to identify and stop the attacks penetrating the traditional layers of defense already in place. This three-month evaluation detected 183 inbound infection attempts and 501 outbound malware beacon attempts. More specifically, the following illustrates just a few of the advanced malware attacks that FireEye detected and stopped: Data exfiltration to China: The system attempted to send information about all dll and exe files to a server in China. DHS itself stated that decadeold Federal cyber security regulations only require manual testing of major systems every three years. As such, if an agency only adhered to the letter of the regulation, this new type of attack could have gone undetected long enough to leak large volumes of data and seriously compromise an agency. Malware that uses SSL: The attacker used custom encryption over port 80 the access point for all HTTP traffic leveraging call-backs over SSL. While the Trusted Internet Connection initiative intends to reduce the number of access points to agency data, it is unlikely that port 80 will be included in this reduction. Weaponized Firefox: A user tried to install Firefox from a disreputable site. The malware then tried to install a fake anti-virus as well as the subsequent beaconing out to an X-rated site, presumably to perform click fraud. This type of attack is much more targeted, and lulls the user into consciously downloading the malware even if the user had a PIV card and was compliant with HSPD-12, the malware would still get inside the network. APTs in Action A Case Study RSA: How Spear Phishing Introduced an APT in the Heart of Security The 2011 attacks targeting RSA, the security division of EMC Corp., provide a very clear picture of the way spear phishing can set the stage for a devastating and incredibly far-reaching assault on an organization and its customers. The assault began with spear phishing attacks that sent targeted users an with a Microsoft Excel file attachment that leveraged a zero-day flaw in Adobe Flash. It is clear that not only was RSA the focus of the attack, but only four individuals within RSA were the recipients of the malicious s. It took just one user to open the and attachment, which downloaded a Trojan onto the user s PC. This successful spear phishing attack was part of a much more complex advanced targeted attack. With this malware installed on the victim s PC, criminals were able to search the corporate network, harvest administrator credentials, and ultimately gain access to a server housing proprietary information on the SecurID two-factor authentication platform. The attack didn t end there. In fact, all of this was a precursor to the ultimate objective: gaining entry to the networks of RSA s customers and focusing on those in the defense industry base. With the stolen data, the criminals then targeted numerous high-profile SecurID customers, including defense contractors. The takeaway from this example is that even seemingly rudimentary attacks may be just the first in a series of advanced, coordinated, and devastating crimes. Advanced targeted attacks against seemingly low-level resources or employees without particularly sensitive roles or permissions can still open the door to vital information and huge consequences. 3

6 Fraudulent Facebook links: This attack came disguised as a Facebook link. The user clicked on the link and launched a data stealer. Think of the possibilities with Facebook-friendly agencies nearly all, even within the DoD downloading dozens of data-stealing bots. Manual testing of systems, as required by some current regulations, can find malware of this type but how long has the malware been accessing agency data? Everythree-year scans, or even yearly scans, just won t cut it anymore. The rise of zero-day attacks means agencies need real-time, dynamic protection beyond what government regulations require. Section II: The Anatomy of Defense Next Generation Threat Protection To be fair, security compliance is necessary but not enough to improve Federal cyber security outcomes. For true security, agencies must go beyond checking boxes on FISMA reports to ensure they are protected from even the most advanced targeted attack. Enterprises need solutions that: Span threat vectors To combat APTs, enterprises must implement integrated protection across both Web and attack vectors. For example, mitigating spear phishing requires capabilities to identify a Web-based attack in real-time, tracing the attack to the initial that spawned the attack, and then analyzing to determine if others within the organization were also targeted. This kind of realtime cyber awareness is the only way to identify and stop advanced targeted attacks in real-time. Combat s varied threat vectors Threat actors are leveraging multiple aspects of assault when attempting to compromise enterprise systems, such as weaponized attachments or URLs that link to a malicious webpage load. Enterprises must demand solutions that address each of these attack vectors when building a next-generation network defense capability. Addresses the mobile computer threat Most organizations have many employees who regularly take their laptops home or on the road. This presents yet another challenge for the enterprise security team identifying infected systems that became compromised while outside of the enterprise protected network. If the traditional anti-virus doesn t detect the advanced malware while on the road, it will not detect the infection once the laptop returns to the office. Enterprises need solutions that conduct thorough analysis of outbound traffic to identify systems that are beaconing to malware command and control servers. Deliver signature-less, dynamic security that thwarts zeroday exploits You need solutions that provide dynamic, real-time exploit analysis of attachments and URLs, rather than just comparing bits of code to signatures or relying on reputations. This signature-less analysis is critical to defending against advanced tactics because it all starts with zero-day exploits. With exploit detection, it is possible to stop advanced malware embedded in attachments as well as malware hosted on dynamic, fast-changing domains. Guard against malicious code installs and block callbacks In addition to exploit detection, you need a solution that identifies whether suspicious attachments and other objects as well as resulting callback communications are malicious or not. This includes monitoring outbound host communications over multiple protocols in realtime to determine if the communications indicate an infected system is on your network. Callbacks can be stopped based on the unique characteristics of the communication protocols employed, rather than just the destination IP or domain name. Once malicious code and its communications are flagged, the ports, IP addresses, and protocols must be blocked in order to halt any transmissions of sensitive data. This prevents the further download of malware binary payloads and the lateral spread inside your organization. 4

7 Yield timely, actionable threat intelligence and malware forensics While is it important for your agency to stop malware attempts, it is important to also learn from them. You need security solutions that fingerprint the malicious code to auto-generate protection data and identify compromised systems preventing the infection from spreading. Also consider having your forensics researchers run files individually through automated offline tests to confirm and dissect malicious code. This can lead to valuable intelligence for future methods to improve your agency s security. Section III: The Federal CISO Dilemma, Solved Today s Federal CISOs need to spend precious budget to comply with regulations such as FISMA, while also protecting some of the most attacked networks in the world. It might be easy for CISOs to spend their entire budgets on compliance or threat protection individually, but the good news is there s room for both. CISOs must not stop at regulatory compliance, said Bob Bigman, former CISO, Central Intelligence Agency. Cyber attacks on the government have increased at a startling rate. Nation states are invested in stealing sensitive, classified data. Agencies must implement robust technology, such as FireEye, to combat these attacks. FireEye s unique, game-changing architecture detects zero-day attacks the favored method of sophisticated cyber criminals. Agency after agency has found that there are solutions, such as those offered by FireEye, which enable them to protect their organizations against sophisticated attacks, despite limited manpower and financial resources. FireEye offers a no-cost threat assessment to all Federal organizations, which uses real-time threat data to help CISOs make the budgetary case to leverage security solutions. This assessment identifies the and Web-based malware that is bypassing the existing security infrastructure, determines the occurrence of data exfiltration from the network, and provides realtime threat intelligence on advanced, zero-day, and targeted APT attacks. Post-assessment, FireEye has a range of products that can help agencies achieve true cyber security. FireEye solutions integrate Web and security, guard against inbound malicious objects and malware callbacks, and leverage signature-less, dynamic code execution to detect zero-day exploits the nextgeneration threat protection necessary to detect and stop APTs in real-time. Contact FireEye today to set up an enterprise threat assessment and take the first step to solving the Federal CISO dilemma. About FireEye, Inc. FireEye is the leader in stopping advanced targeted attacks that use advanced malware, zero-day exploits, and APT tactics. FireEye s solutions supplement traditional and nextgeneration firewalls, IPS, anti-virus and gateways, which cannot stop advanced threats, leaving security holes in networks. FireEye offers the industry s only solution that detects and blocks attacks across both Web and threat vectors as well as latent malware resident on file shares. It addresses all stages of an attack lifecycle with a signature-less engine utilizing stateful attack analysis to detect zero-day threats. Based in Milpitas, California, FireEye is backed by premier financial partners including Sequoia Capital, Norwest Venture Partners, and Juniper Networks. Visit for more information. 5