This chapter covers the following subjects:
• Control plane attack types: Reviews the different types of attacks that are focused on the control plane.
• Control plane security technologies: Goes over the specific technologies that can mitigate the various attack types and describes how to configure them.
The control plane of a device is the heart of each device and must be kept protected to keep everything else that relies on it protected. This chapter looks at the specific threats that are focused on the control plane and the technologies that can be used to mitigate these types of attacks.
The “Do I Know This Already?” quiz helps you decide whether you really need to read the entire chapter. If you already intend to read the entire chapter, you do not necessarily need to answer these questions now.
The ten-question quiz, derived from the major sections in “Foundation Topics” section of this chapter, helps you determine how to spend your limited study time.
Table 9-1 outlines the major topics discussed in this chapter and the “Do I Know This Already?” quiz questions that correspond to those topics.
Table 9-1. “Do I Know This Already?” Foundation Topics Section-to-Question Mapping

The goal of self-assessment is to gauge your mastery of the topics in this chapter. If you do not know the answer to a question or are only partially sure of the answer, you should mark this question wrong for purposes of the self-assessment. Giving yourself credit for an answer that you correctly guess skews your self-assessment results and might provide you with a false sense of security.
1. The central processing unit of each device is tasked to not do which of these?
a. Process fast-path data plane traffic
b. Process control plane traffic
c. Process management plane traffic
d. Process slow-path data plane traffic
2. The route processor is divided into which of the following parts?
a. Distributed switch engine
b. Management plane
d. Control plane
3. The purpose of a slow-path denial of service attack is to force packets to be what?
a. Distributed switched
b. Process switched
c. Routed
d. Switched
4. When using Control Plane Policing, the two types of policing types include which of the following?
a. Distributed control plane services
b. Summarized control plane services
c. Processed control plane services
d. Aggregate control plane services
5. Which of the following traffic is classified as always destined for the control plane?
a. Data traffic packets
b. Routing protocol control packets
c. Management protocol packets
d. Marked QoS packets
6. When using Control Plane Protection, which of the following subinterfaces are not used to further refine control plane security?
a. Control plane host subinterface
b. Control plane transmit subinterface
c. Control plane CEF-exception subinterface
d. Control plane transit subinterface
7. Which of the following features were added with Control Plane Protection?
a. Port filtering
b. Queue thresholding
c. Protocol filtering
d. Port thresholding
8. Which of the CPPr features provide the ability to early-drop specific packets before they get to the process level?
a. Queue thresholding
b. Packet filtering
c. Protocol filtering
d. Port filtering
9. What security mechanism works by creating a hash that is then transmitted to verify authenticity?
a. MD4
b. CPPr
c. MD5
d. AES
10. Which of the following steps is not used by the MQC to create and deploy a traffic policy?
a. Creation of a class map
b. Application of a class map
c. Creation of a policy map
d. Application of a policy map
The answers to the “Do I Know This Already?” quiz are found in Appendix A. The suggested choices for your next step are as follows:
• 8 or less overall score: Read the entire chapter. This includes the “Foundation Topics” and the “Exam Preparation” sections.
• 9 or 10 overall score: If you want more review on these topics, skip to the “Exam Preparation” section.
Although the data plane of a device is very important to protect, it is also vital to protect the control plane of the device. If the control plane of any device is exploited, the security of the device itself and all other actions to mitigate other threats become vulnerable to easy circumvention. The following sections review the major types of attack that are focused at the control plane.
As stated in the previous chapter, slow-path denial of service attacks, when focused at the data plane, attempt to deny a service or services. This is done by sending a large number of packets through the device with the intention of having them be process switched. When the target is the control plane instead of the data plane, the concept is the same; in this case, the control plane is targeted with a large number of packets. As stated in the previous chapter, the CPU of each device is tasked to perform three functions:

• Process control plane traffic
• Process management plane traffic
• Process slow-path data plane traffic
A successful attack on any one of these three has the capability to affect the other two. The access control list (ACL) techniques shown in Chapter 8, “Implementing and Configuring Cisco IOS Routed Data Plane Security,” can be used to secure the control plane as they do the data plane. This chapter focuses on the methods to mitigate attacks targeted specifically at the control plane.
Because a routing protocol is responsible for making the decisions as to how data is routed across a given network, the security of the routing protocol is paramount. A spoofing attack on a routing protocol involves an attacker attempting to inject false routing information into the protocol with the intention of rerouting the traffic. The route that this traffic takes, after it is exploited, could be focused on denying the specific traffic a path, or it could involve routing the data so that it can be captured and reviewed.
The following sections describe the technologies available to mitigate the attacks covered in the previous sections and discusses how to configure them.
One of the methods that can be used to mitigate attacks that are focused on the control plane is Control Plane Policing (CoPP). CoPP works the same as normal traffic policing; this includes the capability to control which and how much traffic is able to gain access to the control plane. The first thing that you need to understand is what the control plane is and how the architecture of these devices is laid out with respect to the control plane.

The route processor (RP) of a device is divided into two main parts:
• Control plane: Includes the group of processes that are run at the process level and that control most high-level control IOS functions.
• Central switch engine: Responsible for the high-speed routing of packets that typically come from nondistributed line card interfaces. The high-speed routing of packets on distributed interfaces is typically handled at the processor on the line card.
CoPP takes advantage of this processing capability by offering a separate level of policing at the line card as well as on a level that is performed by the central switch engine. The two available control plane policing types include

• Distributed control plane services
• Aggregate control plane services
It is important to be aware that aggregate control plane services are configured to be used on all control plane traffic, including that coming from line cards that support distributed processing. Distributed control plane services are considered first, and then the conditioned traffic is passed through to aggregate control plane services.
It is also important to know that traffic is not subject to either of the control plane services until the packet has been determined to have the control plane as its destination. Input control plane services are executed only after a routing decision has been made. Output control plane services are applied after the packet exits the control plane and are available only with aggregate control plane services.
Traffic that is classified as destined for the control plane includes

• Routing protocol control packets
• Packets destined for a local IP address of the device
• Packets from management protocols
The second thing that needs to be explained is how CoPP works. When using CoPP, the control plane is configured to act as a separate entity with its own input and output interface. The modular quality of service command-line interface (MQC) is then used to classify traffic into classes and to create policies that dictate the traffic action. The MQC is easy to learn and will be reviewed next.
The first thing that is configured through MQC is the creation of a traffic class.
The configuration of the traffic class involves the creation of the class. The class is then configured to match specific traffic based on a number of available commands. Numerous commands are available to use within the MQC for classifying traffic. Some of the most commonly used commands with CoPP are covered in Table 9-2.
Table 9-2. Configuring an MQC Traffic Class

Example 9-1 shows the configuration of a traffic class based on an access list. The access list is configured to allow Secure Shell (SSH) traffic from the 10.10.10.0/24 network to enter the control plane without restriction and polices all other SSH traffic from other sources. At first glance, this looks reversed; however, the deny statement in this example exempts the 10.10.10.0/24 traffic from CoPP because this traffic is not matched.
Example 9-1. Configuring a Traffic Class
Router# configure terminal
Router(config)# access-list 110 deny tcp 10.10.10.0 0.0.0.255 any eq 22
Router(config)# access-list 110 permit tcp any any eq 22
Router(config)# class-map ssh-class
Router(config-cmap)# match access-group 110
The second thing that must be configured through MQC is the creation of a traffic policy.
The configuration of the traffic policy involves the creation of the policy. The policy is then configured to use a specific traffic class to identify packets that will then be subject to the traffic policy. The next part involves what to do with the traffic if it is matched with the traffic class. With CoPP, the available options include drop and police. Table 9-3 outlines the commands to configure an MQC traffic policy.
Table 9-3. Configuring an MQC Traffic Policy

Example 9-2 shows the configuration of a traffic policy that uses the traffic class ssh-class defined in Example 9-1 to identify packets for the policy. The traffic policy then polices all other traffic by limiting the traffic rate to 100 kbps; if this rate is exceeded, those excessive packets are dropped.
Example 9-2. Configuring a Traffic Policy
Router# configure terminal
Router(config)# policy-map ssh-policy
Router(config-pmap)# class ssh-class
Router(config-pmap-c)# police 100000 conform-action transmit exceed drop
The application of the traffic policy on a specific interface is the final thing that must be configured through MQC. In this part of the process, the commands specific to CoPP are used. As stated previously, CoPP allows the control plane to be considered like a separate entity with its own input and output interface. When using the MQC to apply policies to a normal interface, the commands would be entered in interface configuration mode. When using CoPP, the commands are entered in control-plane configuration mode. Table 9-4 displays a list of the commands used to apply a traffic policy.
Table 9-4. Applying an MQC Traffic Policy

When an output-aggregated control plane policy is used, those packets dropped by the policy do not create a system message; this is called silent mode operation.
Example 9-3 shows the application of the traffic policy ssh-policy on traffic entering the control plane; this is an example of aggregated control plane services.
Example 9-3. Applying a Traffic Policy to the Control Plane
Router# configure terminal
Router(config)# control-plane
Router(config-cp)# service-policy input ssh-policy
If there is a line card available that supports distributed control plane services, the syntax in Example 9-3 is slightly modified. Example 9-4 shows the application of the traffic policy ssh-policy entering the control plane on the line card in slot 1; this is an example of distributed control plane services.
Example 9-4. Applying a Traffic Policy to the Control Plane (Distributed)
Router# configure terminal
Router(config)# control-plane slot 1
Router(config-cp)# service-policy input ssh-policy
Although aggregated CoPP supports both input and output policies, distributed CoPP only supports input policies.
The second method available to mitigate attacks that are focused on the control plane is Control Plane Protection (CPPr). The first piece of CPPr is essentially an extension of CoPP. With CPPr, the control plane interface is split into four pieces, an aggregate interface and three subinterfaces. The aggregate interface receives all traffic, which is then output into one of three different subinterfaces; this allows a finer level of control on traffic. At press time, the use of CPPr with distributed processing is not supported. The three subinterfaces are as follows:

• Control plane host subinterface: The host subinterface receives all control plane IP traffic that is directed at one of the device’s interfaces. All the traffic that is processed on the host subinterface is terminated on and processed by the device.
• Control plane transit subinterface: The transit subinterface receives all control plane IP traffic that is software switched on the route processor. All of this traffic is not destined for the device directly but is transiting through the device.
• Control plane CEF-exception subinterface: The CEF-exception subinterface receives all traffic that is either redirected as a result of a configured input feature in the Cisco Express Forwarding (CEF) packet forwarding path for process switching or entered directly in the input queue by the interface driver (ATP, L2 keepalives, and all non-IP host traffic).
The second piece of CPPr includes the addition of two different features that extend the capability to block specific traffic and to limit traffic based on a specific protocol. These two CPPr features are as follows:

• Port filtering: Enhances control plane protection by providing the ability to drop packets early that are directed at closed or nonlistened-to ports.
• Queue thresholding: Enhances CPPr by providing a mechanism for limiting the number of unprocessed packets that a specific protocol can have at the process level. If configured, it is processed before port filtering.
Both the port-filtering and queue-thresholding features are supported only on the control plane host subinterface.
The configuration of CPPr using the aggregate interface is the same as with CoPP. The standard configuration of CPPr using the three additional subinterfaces is similar to that of CoPP; the main difference is in the application of the policy. The configuration of port filtering and queue thresholding still utilizes the MQC concepts but requires a modification of the commands that are specific to each; these will be discussed in their own sections. Table 9-5 displays the commands that are required to apply a traffic policy.
Table 9-5. Applying an MQC CPPr Traffic Policy

Example 9-5 displays the application of the ssh policy created in the previous section using the control plane host subinterface.
Example 9-5. Applying a Traffic Policy to the Control Plane Host Subinterface
Router# configure terminal
Router(config)# control-plane host
Router(config-cp)# service-policy input ssh-policy
The port-filtering feature provides the capability to configure CPPr to drop specific packets before they even get to the process level (control plane IP input queue). Typically, these filters are configured to automatically drop all traffic that is destined for closed or nonlistened-to ports. However, port filtering can be configured to drop packets from any specific TCP or UDP port that is configured.
The configuration of the port-filtering feature is similar to that used by CoPP and still uses the same MQC class map, policy map, and service policy structure. The creation of a traffic class utilizes the same overall syntax but with the addition of the type port-filter parameter. Table 9-6 displays the commands required to configure a traffic class.
Table 9-6. Configuring a Port Filter Traffic Class

Example 9-6 shows the configuration of a port filter class map that matches all closed ports on the router.
Example 9-6. Configuring a Port Filter Traffic Class
Router# configure terminal
Router(config)# class-map type port-filter portf-class
Router(config-cmap)# match closed-ports
The creation of a traffic policy utilizes the same overall syntax but with the addition of the type port-filter parameter. Table 9-7 displays the commands required to configure a traffic policy.
Table 9-7. Configuring a Port Filter Traffic Policy

Example 9-7 shows the configuration of a port filter policy map utilizing the portf-class class map defined in Example 9-6 for classification. It is important to note that when configuring a port filter policy map, the only available action is drop.
Example 9-7. Configuring a Port Filter Traffic Policy
Router# configure terminal
Router(config)# policy-map type port-filter portf-policy
Router(config-pmap)# class portf-class
Router(config-pmap-c)# drop
Like the traffic class and policy, the application of a port filter requires the addition of the port-filter parameter. The port filter feature is only able to be used on the control plane host subinterface. Table 9-8 displays the commands required to apply a traffic policy.
Table 9-8. Applying a Port Filter Traffic Policy

Example 9-8 displays the application of the portf policy created previously using the control plane host subinterface.
Example 9-8. Applying a Port Filter Traffic Policy
Router# configure terminal
Router(config)# control-plane host
Router(config-cp)# service-policy type port-filter input portf-policy
The queue-thresholding feature is used to limit the amount of traffic that can come from a given protocol into the control plane. The purpose of this is to limit the ability of a single protocol from overwhelming the process level (control-plane IP input queue) should it malfunction.
The queue-thresholding feature, like the port-filtering feature, continues to use the MQC structure, except with minor modification. The creation of a traffic class utilizes the same overall syntax but with the addition of the type queue-threshold parameter. Table 9-9 displays the commands required to configure a traffic class.
Table 9-9. Configuring a Queue Threshold Traffic Class

Example 9-8 shows the configuration of a queue threshold class map that matches all HTTP traffic.
Example 9-8. Configuring a Queue Threshold Traffic Class
Router# configure terminal
Router(config)# class-map type queue-threshold queue-class
Router(config-cmap)# match protocol http
The creation of a traffic policy utilizes the same overall syntax as the port-filtering feature, as shown previously with the class map, but with the addition of the type queue-threshold parameter. Table 9-10 displays the commands required to configure a traffic policy.
Table 9-10. Configuring a Queue Threshold Traffic Policy

Example 9-9 shows the configuration of a queue threshold policy map utilizing the queue-class class map defined in Example 9-8 for classification and limiting the number of packets to 50. It is important to note that when configuring a queue policy map, the only available action is queue limit. The queue limit can be set from 0 through 255 packets, which is the number of matched protocol packets allowed at the process level (control-plane IP input queue).
Example 9-9. Configuring a Queue Threshold Traffic Policy
Router# configure terminal
Router(config)# policy-map type port-filter queue-policy
Router(config-pmap)# class queue-class
Router(config-pmap-c)# queue-limit 50
Like the traffic class and policy, the application of a queue threshold requires the addition of the queue-threshold parameter. As stated previously, the queue threshold feature is able to be used only on the control plane host subinterface. Table 9-11 displays the commands required to apply a traffic policy.
Table 9-11. Applying a Queue Threshold Traffic Policy

Example 9-10 displays the application of the queue policy created previously using the control plane host subinterface.
Example 9-10. Applying a Queue Threshold Traffic Policy
Router# configure terminal
Router(config)# control-plane host
Router(config-cp)# service-policy type queue-threshold input queue-policy
The security of the routing infrastructure is very important because it is responsible for determining the paths for all traffic using it. If an attacker was able to exploit this, the security of not only the routing infrastructure but also all devices relying on this infrastructure would be affected. One of the easiest ways to ensure the security of these routing protocols is through the use of routing protocol authentication. Typically this involves a configuration change on every routing device to exchange some type of shared secret (typically using message digest algorithm 5 [MD5]) to ensure neighbor authenticity.
MD5 works by creating a one-way hash out of a shared secret and sending this hash between the source and destination. The destination then calculates the hash from its configured shared secret and verifies the source authenticity. Because this is a one-way process, the process cannot be reversed to determine the originating shared secret.
The way that each routing protocol performs this authentication is slightly different. Each of the specific routing protocol configuration requirements will be covered in separate sections.
Because both Enhanced Interior Gateway Routing Protocol (EIGRP) and Routing Information Protocol version 2 (RIPv2) take advantage of the key chain functionality within IOS, it makes sense to review the use and configuration of a key chain. A key chain is essentially an electronic repository of keys and their respective shared secret and validity schedules. The key chain function supports over 2 billion configured keys, so the likelihood of running into a key storage issue is rather slim. Multiple keys can then be set up and configured to be used based on a specific time schedule. This makes it easy for keys to be routinely changed automatically within the configuration. Table 9-12 displays the commands required to configure a key chain.
Table 9-12. Key Chain Configuration

Example 9-11 displays the creation of a key chain with a key number of 1 and a shared secret of “sharedsecret” that can be received and sent from October 29, 2008 at noon through January 27, 2009 at noon.
Example 9-11. Key Chain Creation
Router# configure terminal
Router(config)# key chain auth-chain
Router(config-keychain)# key 1
Router(config-keychain-key)# key-string sharedsecret
Router(config-keychain-key)# accept-lifetime 12:00:00 Oct 29 2008 12:00:00 Jan 27
2009
Router(config-keychain-key)# send-lifetime 12:00:00 Oct 29 2008 12:00:00 Jan 27 2009
As previously stated, EIGRP uses the key chain functionality that is built into IOS. After the key chains have been configured, EIGRP must be configured to use this information for authentication. Table 9-13 displays the commands required to configure EIGRP authentication.
Table 9-13. EIGRP Authentication Configuration

Example 9-12 displays the configuration of EIGRP authentication on autonomous system 100 using interface fastethernet0/0 and key chain auth-chain. In this cast, this device and the remote neighbor device on the same network would require the same shared secret to communicate.
Example 9-12. EIGRP Authentication Configuration
Router# configure terminal
Router(config)# interface fastethernet0/0
Router(config-if)# ip authentication key-chain eigrp 100 auth-chain
Router(config-if)# ip authentication mode eigrp 100 md5
Like EIGRP, RIPv2 supports authentication using the key chain functionality and is configured very similarly to EIGRP authentication. RIPv2 also supports the use of clear-text authentication using the key chain functionality, although it is not recommended that this be used because it is so easily exploited. Table 9-14 displays the commands required to configure RIP authentication.
Table 9-14. RIP Authentication Configuration

Example 9-13 displays the configuration of RIPv2 authentication using interface fastethernet0/0 and key chain auth-chain. As with EIGRP, the neighbors off of this interface will require a matching shared secret to communicate with this device.
Example 9-13. RIP Authentication Configuration
Router# configure terminal
Router(config)# interface fastethernet0/0
Router(config-if)# ip rip authentication key-chain auth-chain
Router(config-if)# ip rip authentication mode md5
Open Shortest Path First (OSPF) authentication configuration does not use the key chain functionality and requires commands specifically used for OSPF. OSPF provides three different authentication options: null (no authentication), simple (clear text), and MD5. OSPF authentication also offers two different methods of configuration. One method is configured for all devices off of a specific segment (interface), and the other method is configured for an entire OSPF area. If both area- and segment-level OSPF authentication commands are used, the segment-level commands override the area-level commands. The choice of which one to choose depends on the specific network configuration.
When configuring OSPF area authentication, the method selected must be the same across the entire area, whether it be null, simple, or MD5. Table 9-15 displays the commands required to configure OSPF area authentication.
Table 9-15. OSPF Area Authentication Configuration

Example 9-14 displays the configuration of OSPF area 0 authentication with a shared secret of “sharedsecret” and using the MD5 hash method.
Example 9-14. OSPF Area Authentication Configuration
Router# configure terminal
Router(config)# interface fastethernet0/0
Router(config-if)# ip ospf message-digest-key 1 md5 sharedsecret
Router(config)# router ospf 10
Router(config-router)# area 0 authentication message-digest
OSPF also supports authentication on a specific network segment. This way, only those neighbors that share interfaces in the same segment must be configured for authentication. As with area authentication, segment authentication supports both simple and MD5 authentication methods. As stated previously, if configured with area authentication, segment authentication will override area authentication settings. Table 9-16 displays the commands required to configure OSPF segment authentication.
Table 9-16. OSPF Segment (Network) Authentication Configuration

Example 9-15 displays the configuration of OSPF segment authentication with a shared secret of “sharedsecret” and using the MD5 hash method.
Example 9-15. OSPF Segment (Network) Authentication Configuration
Router# configure terminal
Router(config)# interface fastethernet0/0
Router(config-if)# ip ospf message-digest-key 1 md5 sharedsecret
Router(config-if)# ip ospf authentication message-digest
In Border Gateway Protocol (BGP), the configuration of authentication (MD5) is done by configuring a shared secret between each BGP neighbor or peer. This can also be configured to work for an entire BGP peer group, which can simplify the configuration of multiple neighbors (peers). As with all other MD5 implementations, the devices exchanging the MD5 must have identical shared secrets to communicate. Table 9-17 displays the commands required to configure BGP authentication.
Table 9-17. BGP Authentication Configuration

Example 9-16 displays the configuration of BGP neighbor (peer) authentication to neighbor 172.16.1.1 on BGP autonomous system 10000 with a shared secret of “sharedsecret.”
Example 9-16. BGP Authentication Configuration
Router# configure terminal
Router(config)# router bgp 10000
Router(config-router)# neighbor 172.16.1.1 password sharedsecret
As mentioned in the section, “How to Use This Book,” in the Introduction, you have several choices for exam preparation: the exercises here, the memory tables in Appendix D, the final exam preparation chapter, and the exam simulation questions on the CD-ROM. The following questions present a bigger challenge than the exam itself because they use an open-ended question format. By using this more difficult format, you exercise your memory better and prove your conceptual and factual knowledge of this chapter. You can find the answers to these questions in Appendix A, “Answers to the DIKTA Quizzes and Fill in the Blanks Questions.”
Review the most important topics in the chapter, noted with the Key Topics icon in the margin of the page. Table 9-18 lists a reference of these key topics and the page numbers on which each is found.

Print a copy of Appendix C, “Memory Tables” (found on the CD), or at least the section for this chapter, and complete the tables and lists from memory. Appendix D, “Memory Table Answers,” also on the CD, includes completed tables and lists to check your work.
Define the following key terms from this chapter, and check your answers in the Glossary:
Table 9-19 lists the important commands from this chapter. To test your memory, cover the right side of the table with a piece of paper, read the description on the left side, and then see how much of the command you can remember.







