Next Generation Network Security Guide

Transcription

1 Next Generation Network Security Guide

2 This expert e-guide explores the latest challenges in network security. First, learn how to deploy network security devices and how to avoid deployment failures. Next, get tips for evaluating network. Finally, explore the pros and of softwaredefined networking. John Burke, SearchSecurity.com Contributor The enterprise is exploding! Bits of it are winding up on mobile phones and tablets and dangling from Internet connections -- used as WAN links -- in the corner Starbucks and on Amtrak's Acela Express trains. This concept of the extended enterprise -- in which sensitive and valuable data often resides outside the traditional network perimeter -- poses an increasingly vexing problem for IT security organizations. To secure an organization against multi-modal and low-and-slow adaptive persistent threats, IT organizations are deploying a dizzying array of new network security devices: next-generation firewalls; IDS and IPS boxes; security information event management, or SIEM, systems; and advanced threatdetection systems. Ideally, these systems are managed in concert, in accordance with a central security policy, as part of a pervasive protection strategy. Common mistakes that organizations make when deploying these devices can severely hinder their ability to offer pervasive protection, however. This tip examines what to watch out for when planning the deployment of new network security devices and how to avoid the problems that can result in failed defense-in-depth. Page 1 of 12

3 Unlearn assumptions about security appliances The single biggest mistake is to assume that security appliances are themselves secure. This seems too obvious for words, but it's critical as a starting point. How secure is that "hardened" OS? How current is it? What about that "ultra-hardened" Web server it runs? Before you do anything else, create a test plan to validate that your network security devices are actually secure. Start with the basics: Are you delivering timely upgrades, patches and bug fixes to individual devices and to their supporting network, server and storage infrastructure? Check with clearinghouses such as the National Vulnerability Database that maintain current catalogues of known vulnerabilities, and make sure you regularly upgrade and patch your devices. Then move on to the harder stuff: regularly assess multi-device configurations for potential weaknesses. Putting an encryption system and application delivery optimization (ADO) device in the wrong sequence can leave data exposed, even when each device is working perfectly. This process can work in concert with regularly-scheduled penetration (pen) tests. Evaluate how you work with network security devices For any security device, the management/control channel is the greatest vulnerability. So pay attention to how you expect to configure and modify the security device -- and to who is allowed to do the configuration. If you expect to work with a security system via a Web browser, the security appliance is running a Web server and allowing Web traffic. Is that traffic encrypted? Is it on a standard port? Is it on the same port on every device (and, therefore, guessable by an intruder)? Is it accessible via a regular network connection (in-band) or a separate management network connection (out-of-band)? If it is in-band, then any machine that can send traffic across that interface can attack that appliance. If it is on a management network, at least, you only have to worry about the other things on that network. (And if it's configured via a serial port connection and a KVM solution, so much the better.) Your optimal scenario: Ensure that all configuration changes require encryption and multifactor authentication, if not direct physical access to the Page 2 of 12

4 device. And tightly track and control credentials for device administration so that only authorized users can gain admin rights. Contents Apply standard pen testing tools If you've taken the first two steps, you're off to a good start -- but you're not home free. Hacks, assaults and threat vectors tantly grow and evolve, and you need to regularly test your systems to ensure they're protected against recognized attacks, not just vulnerabilities. What's the difference between an attack and a vulnerability? An attack is an organized effort to exploit vulnerabilities. System vulnerabilities make an attack possible, but the existence of the attack raises the stakes -- the exploit has moved from the hypothetical to the real. Pen testing tools and services will tell you if your network security devices are vulnerable to attacks. Open source tools and frameworks -- Network Mapper, or Nmap, Nikto, Open Vulnerability Assessment System (OpenVAS) and Metasploit, for example -- have been around for many years. And, of course, there are multitudes of commercial tools from the likes of McAfee (an appliance to scan your appliance!) and Qualys. These tools are widely used to map out the ports on which a network device will respond to network traffic; record its responses to standard test packets; and with OpenVAS and Metasploit, test its vulnerability to common attacks (more kinds with the commercial versions). Other pen testing tools specifically focus on Web servers and applications, such as OWASP Zed Attack Proxy, or ZAP, and Arachni. By applying standard tools and techniques, and identifying vulnerabilities in security appliances -- SQL injection attacks via a management Web interface, for example -- you can build a clear picture of how the network security devices themselves need to be protected. Mitigate the risks when deploying network security devices Nothing is perfect and no system is invulnerable. Failing to take the proper precautions when deploying and configuring new network security devices Page 3 of 12

5 will introduce more risk into the environment. Take appropriate measures to protect the appliances that will defend the rest of your infrastructure, including commonsense precautions that often get overlooked: Change default passwords and account names. Disable unneeded services and accounts. Make sure underlying OSes and systems software are patched and up to date with manufacturer specs. Restrict access to the administrative interfaces of management networks; if that's not possible, use ACLs on upstream devices (switches and routers) to restrict where management sessions can originate. Revisit pen testing regularly, as attacks evolve. Tools such as OpenVAS and Metasploit have advanced to keep up, and the library of exploits they can use grows steadily. The bottom line? Having a pervasive protection strategy is just the beginning. To protect devices and data in today's increasingly perimeter-less world, you need three things: a pervasive protection strategy, the tools and technologies to implement the strategy -- and the policies and processes for ensuring those tools and technologies work in concert to maximize protection. All policies and processes need to take into account both the vulnerability of the network security devices themselves (individually and in concert), as well as the ever-changing landscape of attacks and threat vectors that exploit those vulnerabilities. About the author: John Burke is a principal research analyst at Nemertes Research, where he advises key enterprise and vendor clients, conducts and analyzes primary research, and writes thought-leadership pieces across a wide variety of topics. John leads research on virtual enterprise, focusing primarily on the virtual and mobile desktop, application delivery optimization, and management and orchestration tools for the virtualized data center and the cloud. Page 4 of 12

6 Contents Dave Shackleford, SearchSecurity.com Contributor Along with business units' and IT operations' steady push to virtualize data center servers and components comes a new conundrum for security professionals: how best to maintain adequate controls inside the virtual environment. Fortunately, there is now a new breed of mature network security options that encompass virtualization, with enhanced features that rival those of their physical counterparts. In this tip, we'll review key factors to ider when evaluating network. The first step (and arguably the most important one) in the evaluation process is to determine which would be a good fit for you and your organization. The following specific points can help to determine this: Cost. Cost is primarily a factor when weighing whether to replace existing network security technology that likely has limited or no virtualization security capabilities or augment or replace it with new virtual technology. Many vendors have pricing models for virtual platforms that license per hypervisor, per a certain number of virtual machines or per CPU. This may not only result in applying a totally different formula for evaluating the cost of the product, but also the incurrence of additional costs as virtualization use increases over time. Vendor viability. As with any vendor, make sure you do your homework. Some suppliers are more viable than others, and you should talk to their existing customers to see what they think of both the product and their relationship with the vendor. It's wise to scan the recent headlines for any news pertaining to vendors' executive leadership changes, funding announcements or acquisition rumors. Page 5 of 12

7 Native integration with hypervisor platforms. In looking at more technical iderations, most virtual security vendors focus on VMware as the market leader, but more technology companies support Microsoft Hyper-V, Citrix, KVM, and other platforms as well. If your organization has chosen a single virtualization platform vendor, then the security vendor evaluation process becomes easier; if several different virtualization platforms exist, then multiplatform support is a must. Management capabilities. Consider whether the virtual network appliance is easy to manage, whether it integrates into existing security oles, what type of remote access is available (SSH, for example) and whether the system provides granular role-based access. Performance impact and scalability. How much RAM and other resources does the virtual network appliance require? What are the average peak usage scenarios? Vendors should be able to supply some of this information. Architecture flexibility. How many virtual NICs/ports can the virtual firewall support? What kinds of rules are supported and at which protocol stack layers? Virtualization-specific features. What features are available to help control and protect virtual assets, ranging from the hypervisors to VMs? Speaking of features, there are a number that are good to look for, depending on the type of virtual firewall, switch or gateway you are interested in. One of the most important is API extensibility, allowing integration with orchestration platforms, automation environments and other vendors'. Many virtual firewalls today offer stateful inspection, intrusion detection capabilities, anti-malware features, and configuration and patch assessment and monitoring for the virtual infrastructure. Ensure the platform can perform both intra-vm (internal flows on the hypervisor) and inter-vm (between virtual machines and external networks) monitoring and filtering. Deep integration with the hypervisor environment, preferably at the kernel level, will improve performance and reduce overhead, as well. The ability to identify, monitor and control virtualization-specific traffic and dynamic VM Page 6 of 12

8 migration operations like vmotion should also be a priority when choosing one of these solutions. Contents Many options exist today, from both well-known vendors and startups. Juniper Networks offers its vgw (vgateway) series of virtual appliances, Cisco Systems has the Nexus 1000v virtual switch and ASA 1000v virtual firewall, and 5Nine Security Manager for Hyper-V offers anti-malware and traffic access controls for Microsoft environments. Most IDS/IPS vendors have virtual models, as well, including Sourcefire, McAfee, TippingPoint and others. SDN security pros and Matthew Pascucci In the technology industry, hype is a tant. This is no different with the upand-coming technology of software-defined networks. In this case, however, the hype is justified: SDN could change the network security landscape as we know it. Over the past couple of years, software-defined networking (SDN) has developed from merely an idea to a paradigm that large networking vendors are not only embracing, but also talking up as their model for future enterprise network management. This technology adds greater granularity, dynamics and manageability to networking, but brings up other concerns that should be seen from a security perspective. In this tip, we'll explain what SDN is and explore the network security pros and that enterprise networking and security pros need to know. A definition of software-defined networking To understand a few of the security benefits and downfalls of softwaredefined networking, let's take a quick tour of the technology. Softwaredefined networking is the ability to split the data plane from the control plane Page 7 of 12

9 in routers and switches. The control plane, which has historically been proprietary and known only to the vendors that developed them, would be open and controlled centrally with SDN while having commands and logic sent back down to the data planes of the hardware (routers or switches). This provides a view of the entire network and affords the ability to make changes centrally without a device-centric configuration on each router or switch. The ability to manage the control planes through open protocols such as the OpenFlow standard allows for precise changes to networks or devices that will increase the speed and security of the network. SDN security benefits Like everything else, there will be both benefits and concerns when implementing new technology. Let's review some of the benefits of softwaredefined networking: By having the free-moving network of SDN, engineers are able to change the rules by having a quick, high-level view into all areas of the network and being able to modify the network. This freedom and control also allows for better security of your systems. By having the ability to quickly limit and see inside the network from a centralized viewpoint, managers can make changes with efficiency. For example, if there were a malware outbreak within your network, with SDN and OpenFlow you'd be able to quickly limit the outbreak from one centralized control plane that would stop the traffic without having to access multiple routers or switches. Being able to quickly change things in the network enables managers to perform traffic shaping and QoS of packets in a more secure matter. This ability exists now, but the speed and efficiency doesn't exist and will limit the manager's ability when attempting to secure the network. SDN security concerns With innovative new technology come security concerns that could easily go overlooked. Let's take a look at a few security-related issues to be aware of when implementing SDN. The majority of software-defined networking security concerns are going to evolve around the controller itself. Page 8 of 12

10 The controller can be idered the brains of the switching/routing, which allows the control panel from each system to be centrally managed. Contents The largest SDN challenge for security managers is securing the controller at all costs. Now that the brains have been taken out of the routers or switches and replaced with the new controller, this device needs to be hardened and secured through the following steps: Knowing and auditing who has access to the controller and where it resides on the network is a big security concern. It's important to remember that access to the controller could potentially give complete control to an attacker, so it's vital that it is secured. Verify the security between the controller and end nodes (routers or switches) -- specifically that they're communicating over SSL to prevent any malicious intent from accessing the controller. As with anything else, if security isn't baked in from the start, it must be added later on, and it's always more difficult and expensive to do it that way. Make sure the security between the node and controller is configured properly. Verify that there is high availability in the controllers. Creating a business continuity effort for controllers is important because if they are lost, the ability to manage the network is also lost -- and equently, so are all the benefits of SDN and OpenFlow. Verify that everything that comes out of the system is logged. Since managers have control over the network centrally, log every change made and send it to the company's log management solution. When implementing SDN, verify that the organization's SIEM, IPS and any other filtering technology that might block or log changes is updated accordingly. Correlate the logs from the SIEM to alert the manager of changes. Tracking custom events with the SIEM on the control, like login failures and policy changes, will assist with the security of the system. Verify that the IPS isn't identifying any of this traffic as malicious. Configure the appropriate rules in the filtering systems to allow the controller to speak with the nodes when needed. Page 9 of 12

11 In conclusion, software-defined networking is an emerging technology that can allow for granular security by giving an administrator a complete view of the enterprise network. However, by giving the SDN controller centralized management over network nodes to push down changes to these systems, it becomes imperative that the security around this system is locked down. This system is the brains of SDN, and without proper security wrapped around it, the network becomes completely vulnerable to malicious attacks or accidental changes, both of which can take a network down. Now is the time for organizations to ensure that security is a primary ideration in the design, deployment and management of SDNs. About the author Matthew Pascucci is senior information security engineer at a large retail company where he leads the threat and vulnerability management program. He has written for various information security publications and spoken for many industry companies, and is heavily involved with his local InfraGard chapter. You can follow him on or check out his blog at Page 10 of 12

12 Free resources for technology professionals TechTarget publishes targeted technology media that address your need for information and resources for researching, developing strategy and making cost-effective purchase decisions. Our network of technology-specific Web sites gives you access to industry experts, independent content and analysis and the Web s largest library of vendor-provided white papers, webcasts, podcasts, videos, virtual trade shows, research reports and more drawing on the rich R&D resources of technology providers to address market trends, challenges and solutions. Our live events and virtual seminars give you access to vendor neutral, expert commentary and advice on the issues and challenges you face daily. Our social community IT Knowledge Exchange allows you to share real world information in real time with peers and experts. What makes TechTarget unique? TechTarget is squarely focused on the enterprise IT space. Our team of editors and network of industry experts provide the richest, most relevant content to IT professionals and management. We leverage the immediacy of the Web, the networking and face-to-face opportunities of events and virtual events, and the ability to interact with peers all to create compelling and actionable information for enterprise IT professionals across all industries and markets. Related TechTarget Websites Page 11 of 12