Developing Network Security Strategies
Developing security strategies that can protect all parts of a complicated network while having a limited effect on ease of use and performance is one of the most important and difficult tasks related to network design. Security design is challenged by the complexity and porous nature of modern networks that include public servers for electronic commerce, extranet connections for business partners, and remote-access services for users reaching the network from home, customer sites, hotel rooms, Internet cafes, and so on. To help you handle the difficulties inherent in designing network security for complex networks, this chapter teaches a systematic, top-down approach that focuses on planning and policy development before the selection of security products.
The goal of this chapter is to help you work with your network design customers in the development of effective security strategies, and to help you select the right techniques to implement the strategies. The chapter describes the steps for developing a security strategy and covers some basic security principles. The chapter presents a modular approach to security design that will let you apply layered solutions that protect a network in many ways. The final sections describe methods for securing the components of a typical enterprise network that are most at risk, including Internet connections, remote-access networks, network and user services, and wireless networks.
Security should be considered during many steps of the top-down network design process. This isn't the only chapter that covers security. Chapter 2, "Analyzing Technical Goals and Tradeoffs," discussed identifying network assets, analyzing security risks, and developing security requirements. Chapter 5, "Designing a Network Topology," covered secure network topologies. This chapter focuses on security strategies and mechanisms.
Network Security Design
Following a structured set of steps when developing and implementing network security will help you address the varied concerns that play a part in security design. Many security strategies have been developed in a haphazard way and have failed to actually secure assets and to meet a customer's primary goals for security. Breaking down the process of security design into the following steps will help you effectively plan and execute a security strategy:
- Identify network assets.
- Analyze security risks.
- Analyze security requirements and tradeoffs.
- Develop a security plan.
- Define a security policy.
- Develop procedures for applying security policies.
- Develop a technical implementation strategy.
- Achieve buy-in from users, managers, and technical staff.
- Train users, managers, and technical staff.
- Implement the technical strategy and security procedures.
- Test the security and update it if any problems are found.
- Maintain security.
Chapter 2 covered steps 1 through 3 in detail. This chapter quickly revisits steps 1 through 3 and also addresses steps 4, 5, 6, and 12. Steps 7 through 10 are outside the scope of this book. Chapter 12, "Testing Your Network Design," addresses Step 11.
Identifying Network Assets
Chapter 2 discussed gathering information on a customer's goals for network security. As discussed in Chapter 2, analyzing goals involves identifying network assets and the risk that those assets could be sabotaged or inappropriately accessed. It also involves analyzing the consequences of risks.
Network assets can include network hosts (including the hosts' operating systems, applications, and data), internetworking devices (such as routers and switches), and network data that traverses the network. Less obvious, but still important, assets include intellectual property, trade secrets, and a company's reputation.
Analyzing Security Risks
Risks can range from hostile intruders to untrained users who download Internet applications that have viruses. Hostile intruders can steal data, change data, and cause service to be denied to legitimate users. Denial-of-service (DoS) attacks have become increasingly common in the past few years. See Chapter 2 for more details on risk analysis.
Analyzing Security Requirements and Tradeoffs
Chapter 2 covers security requirements analysis in more detail. Although many customers have more specific goals, in general, security requirements boil down to the need to protect the following assets:
- The confidentiality of data, so that only authorized users can view sensitive information
- The integrity of data, so that only authorized users can change sensitive information
- System and data availability, so that users have uninterrupted access to important computing resources
According to RFC 2196, "Site Security Handbook:"
- One old truism in security is that the cost of protecting yourself against a threat should be less than the cost of recovering if the threat were to strike you. Cost in this context should be remembered to include losses expressed in real currency, reputation, trustworthiness, and other less obvious measures.
As is the case with most technical design requirements, achieving security goals means making tradeoffs. Tradeoffs must be made between security goals and goals for affordability, usability, performance, and availability. Also, security adds to the amount of management work because user login IDs, passwords, and audit logs must be maintained.
Security also affects network performance. Security features such as packet filters and data encryption consume CPU power and memory on hosts, routers, and servers. Encryption can use upward of 15 percent of available CPU power on a router or server. Encryption can be implemented on dedicated appliances instead of on shared routers or servers, but there is still an effect on network performance because of the delay that packets experience while they are being encrypted or decrypted.
Another tradeoff is that security can reduce network redundancy. If all traffic must go through an encryption device, for example, the device becomes a single point of failure. This makes it hard to meet availability goals.
Security can also make it harder to offer load balancing. Some security mechanisms require traffic to always take the same path so that security mechanisms can be applied uniformly. For example, a mechanism that randomizes TCP sequence numbers (so that hackers can't guess the numbers) won't work if some TCP segments for a session take a path that bypasses the randomizing function due to load balancing.
Developing a Security Plan
One of the first steps in security design is developing a security plan. A security plan is a high-level document that proposes what an organization is going to do to meet security requirements. The plan specifies the time, people, and other resources that will be required to develop a security policy and achieve technical implementation of the policy. As the network designer, you can help your customer develop a plan that is practical and pertinent. The plan should be based on the customer's goals and the analysis of network assets and risks.
A security plan should reference the network topology and include a list of network services that will be provided (for example, FTP, web, email, and so on). This list should specify who provides the services, who has access to the services, how access is provided, and who administers the services.
As the network designer, you can help the customer evaluate which services are definitely needed, based on the customer's business and technical goals. Sometimes new services are added unnecessarily, simply because they are the latest trend. Adding services might require new packet filters on routers and firewalls to protect the services, or additional user-authentication processes to limit access to the services, adding complexity to the security strategy. Overly complex security strategies should be avoided because they can be self-defeating. Complicated security strategies are hard to implement correctly without introducing unexpected security holes.
One of the most important aspects of the security plan is a specification of the people who must be involved in implementing network security:
- Will specialized security administrators be hired?
- How will end users and their managers get involved?
- How will end users, managers, and technical staff be trained on security policies and procedures?
For a security plan to be useful, it needs to have the support of all levels of employees within the organization. It is especially important that corporate management fully support the security plan. Technical staff at headquarters and remote sites should buy into the plan, as should end users.
Developing a Security Policy
According to RFC 2196, "Site Security Handbook:"
- A security policy is a formal statement of the rules by which people who are given access to an organization's technology and information assets must abide.
A security policy informs users, managers, and technical staff of their obligations for protecting technology and information assets. The policy should specify the mechanisms by which these obligations can be met. As was the case with the security plan, the security policy should have buy-in from employees, managers, executives, and technical personnel.
Developing a security policy is the job of senior management, with help from security and network administrators. The administrators get input from managers, users, network designers and engineers, and possibly legal counsel. As a network designer, you should work closely with the security administrators to understand how policies might affect the network design.
After a security policy has been developed, with the engagement of users, staff, and management, it should be explained to all by top management. Many enterprises require personnel to sign a statement indicating that they have read, understood, and agreed to abide by a policy.
A security policy is a living document. Because organizations constantly change, security policies should be regularly updated to reflect new business directions and technological shifts. Risks change over time also and affect the security policy.
Components of a Security Policy
In general, a policy should include at least the following items:
- An access policy that defines access rights and privileges. The access policy should provide guidelines for connecting external networks, connecting devices to a network, and adding new software to systems. An access policy might also address how data is categorized (for example, confidential, internal, and top secret).
- An accountability policy that defines the responsibilities of users, operations staff, and management. The accountability policy should specify an audit capability and provide incident-handling guidelines that specify what to do and whom to contact if a possible intrusion is detected.
- An authentication policy that establishes trust through an effective password policy and sets up guidelines for remote-location authentication.
- Computer-technology purchasing guidelines that specify the requirements for acquiring, configuring, and auditing computer systems and networks for compliance with the policy.
Developing Security Procedures
Security procedures implement security policies. Procedures define configuration, login, audit, and maintenance processes. Security procedures should be written for end users, network administrators, and security administrators. Security procedures should specify how to handle incidents (that is, what to do and who to contact if an intrusion is detected). Security procedures can be communicated to users and administrators in instructor-led and self-paced training classes.
Security must be maintained by scheduling periodic independent audits, reading audit logs, responding to incidents, reading current literature and agency alerts, performing security testing, training security administrators, and updating the security plan and policy. Network security should be a perpetual process. Risks change over time, and so should security. Cisco security experts use the term security wheel to illustrate that implementing, monitoring, testing, and improving security is a never-ending process. Many overworked security engineers might relate to the wheel concept. Continually updating security mechanisms to keep up with the latest attacks can sometimes make an administrator feel a bit like a hamster on a training wheel.