Illustrated TCP/IP by Matthew G. Naugle Wiley Computer Publishing, John Wiley & Sons, Inc. ISBN: 0471196568   Pub Date: 11/01/98

Chapter 226
Domain Name Service (DNS)

Read RFCs 1034 and 1035. These contain the bulk of the DNS information and are supplemented by RFCs 1535–1537. DNS has many uses, but its main function continues to be the mapping of IP address to human-usable names.

There are millions of hosts on the Internet today representing even more millions of users. Most users have no idea what the underlying protocols are doing, nor do they care. But most of them would if they had to memorize IP addresses and determine other functions such as mail. Actually, most would be frustrated by the numbering system and the Internet would not be as popular as it is. When the Internet was young, an early method of mapping the 32-bit address to a hostname required downloading a file maintained by (at the time) the Network Information Center (NIC). It was a single file (hosts.txt) that contained a simple mapping of Internet addresses to hostnames. This file was usually contained in the /etc subdirectory on a workstation and various TCP/IP applications could access the information in this file. Not having this file meant that a user had to type in the 32-bit address for connectivity to a remote host. Secondly, population of the Internet was becoming very diverse and more autonomous. In the 1980s the Internet was known as the ARPAnet (now shut down) and the hosts were primarily time shared. More and more connections to the Internet were sites that had LANs installed and connected to these LANs were mainframe and minicomputers or even personal computers. These sites were administering their own names and addresses in the hosts.txt file, but had to wait for the NIC to change hosts.txt to make changes visible to the Internet at large. Lastly, with the additions of more sites to the Internet, the applications on the Internet were getting more sophisticated and creating a need for a general-purpose name service.

Domain Name Service (DNS)

After many experimental RFCs, the global name system for the Internet became known as the Domain Name System (DNS). DNS is comprised of three components: a name server, a database, and a name resolver. Name servers make information available to the resolvers. The information the name servers contain are IP addresses, aliases, mail information, and so forth. The resolvers usually reside on users’ workstations and are embedded in the applications of TCP such as TELNET and FTP. They are not separate programs. The name server is a separate program and resides anywhere on a network answering queries from the resolvers. The domain servers each maintain a portion of the hierarchical database under separate administrative authority and control. Redundancy is obtained by transferring data between cooperating servers (primary masters and secondary masters).

Your site may not require a DNS. You may have just a few hosts and can depend on another DNS to supply the information you need. For the Internet itself, it must have the DNS system. A great example on the dependency of DNS was when a corrupted database (containing “directions” to other hosts) file was posted on the nine root servers (explained in a moment). Millions of on-liners were without the capability of attaching or communicating with other hosts on the network for hours. Without information (the IP address) of a remote system, two nodes cannot communicate. We could look up the information in the InterNIC database, but without prior knowledge on how to query their database manually, one is literally lost on the Internet. DNS provides information about hosts, not users, on the Internet.