Concepts in IP Security
No topic related to the Internet, with the possible exceptions of the fl ee availability
of pornography and the plague of unwanted spam email, has received more
attention in the mainstream media than “ security. ” For the average user the concerns
are predominantly viruses that may infect their personal computers, causing
inconvenience or damage to their data. Increasingly we also hear about white-collar
e-criminals who steal personal fi nancial details or defraud large institutions
after illegally gaining entry to their computer systems.
We are also now all familiar with catastrophic failures of parts of the Internet.
Although these are sometimes caused by bugs in core components (such as routers)
or by the perennial backhoe cutting a cable or fi ber, they are increasingly the
responsibility of individuals whose sole joy is to pit their wits against those who
maintain the Internet. Sometimes known as hackers, these people attempt to penetrate
network security, or cause disruption through denial of service attacks for
a range of motives.
Corporate espionage is of relatively little concern to most people, but within
every forward-looking company there is a person or a department responsible for
keeping the company’s secrets safe. At the same time, the populist war against terrorism
invokes contradictory requirements—that the government should be able
to keep its information private while at the same time examining the affairs of suspects
without them being able to hide their communications.
Whatever the rights and wrongs of the politics and sociology, Internet security
is a growth industry. This chapter provides an overview of some of the issues
and shows the workings of the key security protocols. It introduces the security
algorithms without going into the details of the sophisticated mathematics behind
encryption algorithms or key generation techniques. For this type of information
the reader is referred to the reference material listed at the end of the chapter.
The fi rst sections of the chapter examine the need for security, where within
the network it can be applied, and the techniques that may be used to protect data
that is stored in or transmitted across the network. There then follows a detailed
examination of two key security protocols: IPsec, which provides security at the
IP packet level, and Transport Layer Security (TLS), which operates at the transport
layer and provides the Secure Sockets Layer (SSL). After a brief discussion of some
of the ways to secure Hypertext Transfer Protocol (HTTP) transactions, which are
fundamental to the operation of web-based commerce, the chapter describes how
hashing and encryption algorithms are used in conjunction with keys to detect
modifi cation of data or to hide it completely—the Message Digest Five (MDS)
hashing algorithm is presented as the simplest example. The chapter concludes
with an examination of how security keys may be securely exchanged across the
network so that they may be used to decrypt or verify transmitted data.
THE NEED FOR SECURITY
It is fair to say that when the Internet was fi rst conceived, security was not
given much consideration. In fact, the whole point of the Internet was to enable
information to be shared and distributed freely. It is only as a greater number of
computers have been connected together, and the sort of information held on
computers and distributed across the Internet has grown in quantity and sensitivity,
that network security has become an issue.
There are two fundamental issues. First, there is a need to keep information
private for access only by authorized parties. Whether it is classifi ed government
material, sensitive commercial information, your credit card number, or just a note
suggesting that you meet your friend in the bar in half an hour, there is strong motivation
to protect any information sent across the Internet from prying eyes. This
desire extends beyond protection of data transmitted over the Internet, and should
also be considered to cover the safeguarding of fi les stored on computers attached
to the Internet, and access to computing resources and programs. Some of the solutions
to this issue can be seen by users on private networks as they are required to
log on to their workstations, password protect key documents, and digitally sign
their emails.
The second security issue concerns protection of the infrastructure of the
Internet. This covers prevention of attacks on the confi guration of devices in the
network, theft of network resources, and the malicious jamming of nodes or links
with spurious data that makes it impossible for legitimate messages to get through.
Somewhere between these two cases comes prevention of unauthorized
access to secure locations on computers. This access may be in order to read privileged
information, or it may be to replace it with something else, or even simply
to delete it. A popular gag among hackers is to replace the content of a web site
with slogans or pictures that are neither relevant nor helpful to the cause that the
site was promoting.
The Internet has been shown repeatedly to be quite fragile. The accidental misconfi
guration of a key router may result in large amounts of data looping or being
sent off into a void. Malicious changes to routing information may have a similar
effect. At the time of writing, the English-language web site of the Arab news serv ice
al-Jazeera is unreachable because someone has stolen its DNS entry on several key
servers, resulting in all attempts to reach http://www.aljazeera.net being redirected
to another site that displays an American patriotic message. Such intervention in the
smooth operation of the Internet, although no doubt a great deal of fun to the perpetrator,
is at best an inconvenience for the normal user of the Internet. For the commercial
organizations that depend on exchanging information across the Internet or
on customers visiting their web sites, these disruptions are a more serious matter.
Various techniques are used to compromise Internet security. The most obvious
technique involves simply impersonating another user to access that user’s
computer. Remote access protocols such as Telnet and FTP make this particularly
easy. Of course, data that is sent on the Internet can be examined quite easily
using a sniffer, provided access to a computer on the network can be gained or a
sniffer can be hooked up to the network at some point.
Even when passwords and authentication or encryption are used, it may be
possible for someone to capture a sequence of commands and messages and
replay them at a later time to gain access. Such replay attacks can at least confuse
the receiving application and waste system resources, but may return information
such as encryption keys, or may provide access to applications on a remote server.
Denial of service attacks result in degradation of service to legitimate network
users. There is no immediately obvious benefi t to the perpetrator, although the
example in the next section describes how denial of service may be used to trick
network operators into giving away their secrets. Denial of service is increasingly
a tool of “ Internet anarchists ” who target organizations with whom they have a
disagreement and block access to or from those organizations ’ private networks.