General Design Considerations for Secure Networks

Routing Considerations

As we continue to slowly work our way up the OSI model with these best practices, it is now useful to develop some design considerations in the realm of routing. The most important is the security of the routing protocol.

Routing Protocol Security

Routing security has received varying levels of attention over the past several years and has recently begun to attract more attention specifically around BGP on the public Internet. Despite this new attention, however, the area most open to attack is often not the Internet's BGP tables but the routing systems within your own enterprise network. Because of some of the sniffing-based attacks discussed in Chapter 3 and earlier in this chapter, an enterprise routing infrastructure can easily be attacked with MITM and other attacks designed to corrupt or change the routing tables with the following results:

• Traffic redirection—In this attack, the adversary is able to redirect traffic, enabling the attacker to modify traffic in transit or simply sniff packets.

• Traffic sent to a routing black hole—Here the attacker is able to send specific routes to null0, effectively kicking IP addresses off of the network.

• Router DoS—Attacking the routing process can result in a crash of the router or a severe degradation of service.

• Routing protocol DoS—Similar to the attack previously described against a whole router, a routing protocol attack could be launched to stop the routing process from functioning properly.

• Unauthorized route prefix origination—This attack aims to introduce a new prefix into the route table that shouldn't be there. The attacker might do this to get a covert attack network to be routable throughout the victim network.

There are four primary attack methods for these attacks:

• Configuration modification of existing routers

• Introduction of a rogue router that participates in routing with legitimate routers

• Spoofing a valid routing protocol message or modifying a valid message in transit

• Sending of malformed or excess packets to a routing protocol process

These four attack methods can be mitigated in the following ways:

• To counter configuration modification of existing routers, you must secure the routers. This includes not only the configuration of the router but also the supporting systems it makes use of, such as TFTP servers. See Chapter 5, "Device Hardening," for more information.

• Anyone can attempt to introduce a rogue router, but to cause damage, the attacker needs the other routing devices to believe the information that is sent. This can most easily be blocked by adding message authentication to your routing protocol. More on this subject can be found in the next section. Additionally, the routing protocol message types can be blocked by ACLs from networks with no need to originate them.

• Message authentication can also help prevent the spoofing or modification of a valid routing protocol message. In addition, the transport layer protocol (such as TCP for BGP) can further complicate message spoofing because of the difficulty in guessing pseudorandom initial sequence numbers (assuming a remote attacker).

• Excess packets can be stopped through the use of traditional DoS mitigation techniques, which are discussed later in the chapter. Malformed packets, however, are nearly impossible to stop without the participation of the router vendor. Only through exhaustive testing and years of field use do routing protocol implementations correctly deal with most malformed messages. This is an area of computer security that needs increased attention, not just in routing protocols but in all network applications.

As you can see, stopping all these attacks is not a matter of flipping on the secure option in your routing protocols. As stated in Chapter 2, you must decide for your own network what threats need to be stopped. In addition to the specific threats mentioned here, it is also very useful to follow the network design best practices of not running routing protocols on interfaces with no reason to route and of using distribution lists to limit the routing prefixes that are sent or received by a specific routing instance. Details on distribution lists can be found in your favorite Internet routing book.

Routing Protocol Message Authentication

Although they vary in the strength of the authentication they offer, nearly all routing protocols support some form of message authentication. There are two principal types of authentication used in routing protocols today: plaintext password and MD5 digest.

Plaintext Password Authentication

Plaintext password authentication is just what it sounds like. A password is attached to the routing update and is sent in the clear along with the routing update. The passwords have specific length requirements as defined by the routing protocol in use. Plaintext password authentication should be considered specious security because anyone who sees a single routing update on the wire sees the authentication information if it is in use. From this point on, the attacker can appear to be a member of the trusted routing domain. The plaintext password does offer some benefit in that it prevents routing protocol changes when an invalid router is accidentally introduced into a production routing environment.

MD5 Digest Authentication

MD5 digest works by creating a 16-byte hash of the routing message combined with a secret key. The 16-byte value is, therefore, message-specific, and modification of the message by an attacker invalidates the 16-byte digest appended to the message. Without the secret key, which is never sent over the wire by the routing protocol, the attacker is unable to reconstruct a valid message. It is worth noting that the MD5 option provides authentication and packet integrity, not confidentiality. Figure 6-16 shows how the hash function operates.

Figure 6-16 MD5 Digest for Routing Authentication

Specific Routing Protocol Security Options

This section details the security options available in the most widely used routing protocols.

Routing Information Protocol

Routing Information Protocol (RIP) version 1 (RFC 1058) has no mechanism whatsoever to authenticate routing messages. As such, it should never be used in security-sensitive environments.

RIP v2

RIP v2 (RFC 1723) supports a 16-byte plaintext password that can be attached to routing updates. RFC 2082 specifies a proposed standard for adding MD5 authentication to RIP v2. Whenever possible, use the MD5 digest instead of the basic password.

RIP v2 plaintext messages have the format shown in Figure 6-17.

Figure 6-17 RIP v2 Plaintext Authentication

RIP v2 MD5 authenticated messages have the format shown in Figure 6-18.

Figure 6-18 RIP v2 MD5 Authentication

The configuration for RIPv2 authentication is as follows:

!Enable RIP authentication
Router(config-if)# ip rip authentication key-chain name-of-chain
!Specify authentication type
Router(config-if)# ip rip authentication mode {text | md5}
!Identify key chain
Router(config)# key chain name-of-chain
!Specify key number
Router(config-keychain)# key number
!Specify actual key
Router(config-keychain-key)# key-string text

Open Shortest Path First

Open Shortest Path First (OSPF) (RFC 2328) is one of the most widely used interior gateway protocols today. It supports nearly every bell and whistle you could ask of your routing protocol. On the security side, it offers both plaintext authentication (with basic message checksum) and the much more secure MD5 digest.

OSPF MD5 authenticated messages have the format shown in Figure 6-19.

Figure 6-19 OSPF Packet Header

Note that there is no special format for OSPF when you use authentication. Authentication is assumed, even though it defaults to null authentication. In Figure 6-19, AuType specifies the authentication type.

The configuration for OSPF MD5 authentication is as follows:

!The MD5 key is always defined per interface but enabling MD5 can be
!done either on the interface as shown in the first command
!or at the area as in the second command. The
!third command is required for both options.
!Specify OSPF authentication type
Router(config-if)# ip ospf authentication message-digest
!Enable MD5 for an area
Router(config-router)# area area-id authentication message-digest
!Specify MD5 key
Router(config-if)# ip ospf message-digest-key key-id md5 key

BGP

BGP is most widely used in routing between two different routed domains, such as between you and your ISP or your ISP and the upstream ISP. BGP supports MD5 authentication. Note that because BGP uses TCP as a transport protocol, the MD5 authentication is done as a TCP option. More details on this can be found in RFC 2385. TCP Option 19 is specified for this authentication and takes the format specified in Figure 6-20.

Figure 6-20 TCP Option 19 for BGP MD5 Authentication

The configuration for BGP MD5 authentication is as follows:

! Enable TCP MD5 authentication for a specific neighbor
Router(config-router)# neighbor neighbor_ip_addr password text

Because BGP is unicast as opposed to broadcast or multicast, IPsec can be used with it to provide even greater security. As of this writing, some networks were in the testing phase of their deployments. I would recommend waiting until IPsec in combination with BGP receives more testing before deploying on your own network. Even then, the complexity of the configuration and troubleshooting difficulty might prevent this from being a viable option.

Interior Gateway Routing Protocol

Interior Gateway Routing Protocol (IGRP) is a proprietary Cisco routing protocol meant to address some of the limitations of RIP. The initial version did not address any of its security limitations, however, because IGRP supports no form of authentication. Like RIP, IGRP should be avoided in security-sensitive environments.

Enhanced Interior Gateway Routing Protocol

Enhanced Interior Gateway Routing Protocol (EIGRP) is an extension to IGRP that is also Cisco proprietary. It supports MD5 message authentication.

The configuration for EIGRP authentication is as follows:

!Specify EIGRP MD5 authentication
Router(config-if)# ip authentication mode eigrp autonomous-system md5
!Specify authentication key
Router(config-if)# ip authentication key-chain eigrp autonomous-system 
 name-of-chain
!Identify key chain
Router(config)# key chain name-of-chain
!Specify key number
Router(config-keychain)# key number
!Specify actual key
Router(config-keychain-key)# key-string text

Asymmetric Routing and State-Aware Security Technology

As networks increase in size, so do the chances that they have asymmetric traffic somewhere within them. Asymmetric traffic is traffic that uses a different path for its return than the original path of the request. The topology in Figure 6-21 shows a representative network with several places where asymmetric traffic can occur.

Figure 6-21 Asymmetric Traffic

Traffic between the user PC and either the finance server or the WWW server can flow in an asymmetric manner at several points along the network. Between the PC and the finance server, switches S1 and S3 are the main location it can occur. Between the PC and the WWW server, traffic could take an asymmetric route at S1 and S2 or at the Internet when returning through ISP A or ISP B.

So far, this is network design 101. Most network designers don't have any problem with asymmetric traffic because IP networks are asymmetric by nature. At each point in the transmission, an IP router makes a forwarding decision based on its view of the network.

This becomes problematic when security devices are introduced that rely on state information to make forwarding decisions. Consider the revised diagram in Figure 6-22, where two stateful firewalls are introduced between campus A and the two Internet connections.

Figure 6-22 Asymmetric Traffic with Security Devices

Now asymmetric flows really start to cause problems! Again, consider the PC communicating with server HTTP://WWW. A perfectly reasonable packet flow might have the outgoing connection flow through S4, S1, FW1, Inet_RTR_1, ISP A, and then to server HTTP://WWW. Along the way, FW1 learns that the PC is trying to communicate with server WWW, and so it adds an entry in its state table to enable the return traffic to flow when it comes back from server HTTP://WWW. Unfortunately, the return path for the packet from server WWW to the user PC happens to be ISP B, Inet_RTR_2, FW2, S2, S4, user PC. The packet never reaches the PC, though, because FW2 doesn't have any state information for the communication. As far as it is concerned, server WWW is initiating new communications to the user PC that are blocked based on the configured security policy.

This problem can be further complicated by intrusion detection systems (IDS) deployed within the campus or near the firewalls. If traffic flows by an IDS in an asymmetric manner, it won't see all of the data. Consequently, it might alarm on traffic that is benign (false positive), or it might miss an attack altogether (false negative).

I wish there were an easy answer to this problem, but unfortunately there isn't. This section is included as much to bring the problem to your attention as it is to offer possible solutions. You do have some options, however:

• Make your routing symmetric.

• Load balance per flow rather than per packet.

• Use state-sharing security devices.

• Consider L2 redundancy as a workaround.

• Manipulate flows by using routing or NAT.

• Use stateless security features.

Make Your Routing Symmetric

This might seem easy, but in real network designs it can be a significant challenge. Even still, you would be surprised to see how many large networks use symmetric routing at certain parts of their network to enable state-aware security devices to function or to solve other networking issues. This is particularly common at Internet edges, where it is not unheard of to see an entire connection to an ISP lying dormant while the primary connection handles all of the load.

Load Balance Per Flow Rather Than Per Packet

Most L3 devices can be configured to do one of two things when equal-cost paths exist for a given network destination. In the first option, packets are simply balanced in round-robin format, with each successive packet going to the next available upstream router. This option causes the most heartache with internal security systems such as IDS. The second, more preferred, option is to load balance based on a given flow. This means traffic with a particular source and destination IP address and port (often called a four tuple) is always sent by a specific upstream router. This allows IDS systems and other state-aware devices to at least see half of the communication in a consistent manner. Unfortunately, this does nothing to the return traffic, which still might flow over a different link.

Use State-Sharing Security Devices

As the problem of asymmetric traffic manifests itself more and more in networks, network security vendors are starting to offer options allowing the state information within one security device to be shared with another. In Figure 6-22, FWs 1 and 2 could exchange their state table information to ensure that if the other device sees part of a given flow, it will know to permit the traffic. Often, the amount of information exchanged is significant and requires that dedicated links be configured between the firewalls to exchange the state information.

Consider L2 Redundancy as a Workaround

With the careful introduction of L2 redundancy as opposed to L3, technologies such as Virtual Router Redundancy Protocol (VRRP) or Hot Standby Router Protocol (HSRP) can allow traffic to flow through a single location while still providing redundancy. This option works best on high-speed connections where the use of only one path instead of two or more does not affect network performance.

The result is that normally asymmetric flows can be made symmetric for short distances in the network, such as while traffic passes through a firewall. Again, in Figure 6-22, if FWs 1 and 2 were connected on both sides to the same L2 network, they could use something like VRRP to appear as a single firewall to the upstream and downstream routers. This means that traffic can flow in an asymmetric manner out to the Internet and to the internal network but in a symmetric manner when passing through the firewall. This is generally impossible when the two devices are not in close geographic proximity to one another. For example, if FW 1 is in Brussels, Belgium, and FW 2 is in Hackensack, New Jersey, you are out of luck.

Manipulate Flows by Using Routing or NAT

Because this is a book on security, the ins and outs of BGP path preference have no place within the text. It is worth noting, however, that there are a number of things that can be done with routing protocols to affect the paths that packets take. To some degree, you can also influence which path outside networks take when they must communicate with you. Although not very elegant, some other workarounds involve using different NAT pools based on which security device a packet passes through. Return packets can then be forced to a specific security device based on the unique NAT pool they allocate from.

Use Stateless Security Features

Even though firewalls have been around for many years, a number of companies still use basic ACLs instead of stateful firewalls for, among other things, this asymmetric issue. Some security functionality is clearly lost. Basic ACLs don't track state information, but if your traffic flows are fairly easy to categorize, you can still achieve some security without needing symmetric traffic flows. Remember that if you have properly implemented a true security system as defined in Chapter 1, the access control function of a firewall is only one part of the overall security story.

With IDS, the signatures that work improperly in asymmetric environments can be turned off to prevent false positives. Again, this will reduce the security such systems provide but will still allow a number of signatures to fire properly.