Domain Name System (DNS)
LESSON 13

(October 18, 1995)
Daniel Z. Tabor Jr.
New Jersey Institute of Technology

Providing Readable Names:

Goal: Assign meaningful high-level names to a large set of machines and handle the mapping of those names to a machine’s IP address(es).
To do this, we need to use a large number of geographically distributed servers to map names to addresses.
Mapping is needed in both directions:
- IP Address to Name.
- Readable Name to IP Address.
Also, another level of mapping is still needed:
- IP Address to low-level name (usually hardware address).
The form of high-level names determines:
- How those names are translated to lower-level names (IP addresses) or bound to objects.
- How name assignments are authorized.

Flat Namespace:

The original set of machines on the Internet used flat namespaces.
These namespaces consisted of sequence of characters with no further structure.
Advantage:
- Names were convenient and short.
Disadvantage:
- Flat namespaces cannot generalize to large sets of machines because:
  - Single set of identifiers => conflict potential increases.
  - Single central name authority was overloaded.
  - Frequent name-address binding changes were costly and cumbersome.

Hierarchical Names:

The partitioning of a namespace must be defined in such a way that it:
- Supports efficient name mapping.
- Guarantees autonomous control of name assignment.
Hierarchical namespaces provides a simple yet flexible naming structure.
The namespace is partitioned at the top level.
Authority for names in each partition are passed to each designated agent.
LOCAL.SITE
- SITE => Central naming authority.
- LOCAL => Local namespace agent.

Subset Authority:

Further division of authority is available in the namespace hierarchy:
LOCAL.GROUP.SITE
- GROUP => provides flexible naming control for the local agent (grouping).
- Hertz.njit.edu
U.S. telephone numbers are also hierarchical:
- 3 (area code) - 3 (local exchange) - 4 (subscriber number)
In a TCP/IP internet, hierarchical machine names are assigned according to the structure of organizations that obtain authority for parts of the namespace, not according to the structure of the physical network interconnections.

TCP/IP Internet Domain Names:

DNS (domain name system) - mechanisms that implement a machine name hierarchy for TCP/IP internets (uses hierarchical naming - Domain Names).
- It specifies the name syntax and rules for delegating authority over names.
- Specifies the implementation of a distributed computing system that efficiently maps names to addresses.
3 labels: (Hertz.njit.edu)
* Any suffix of a label in a domain name is a domain. *
- Lowest level => hertz.njit.edu
  (Host at NJIT)
- Second level => njit.edu
  (Domain name of NJIT)
- Top level => edu
  (Domain name for educational sites)

(Un)Official Internet Domain Names:

Most users follow the labeling conventions used by official DNS for two reasons:
- The scheme is comprehensive and flexible. It can accommodate many hierarchical names for organizations (geographical or organizational naming)
- Easier naming method when attached to the Internet (no name changing is needed).
Top-level names provide geographical naming schemes.
Geographical - country codes:
- ugwww.ucs.ed.ac.uk (United Kingdom)
Standard two letter identifiers are used as top-level domain names.

Geographical Domain Names:

AU	Australia	IT	Italy
BR	Brazil	JP	Japan
CA	Canada	MX	Mexico
DE	Germany	NL	Netherlands
ES	Spain	NO	Norway
FI	Finland	NZ	New Zealand
FR	France	SE	Sweden
GR	Greece	TR	Turkey
HU	Hungary	UK	United Kingdom
IN	India	US	United States

The U.S. further divides the geographical naming scheme into two levels:
(State.Country)
- va.us (Virginia, US)
- nj.us (New Jersey, US)
- ca.us (California, US)
Most universities and companies choose organizational names because:
- Easier to type (shorter names)
- Geographical names are more difficult to ‘guess’
Companies can be recognized as such by having .com as a part of their domain name.

Organizational Domain Names:

COM	US Commercial	NET	Network
EDU	US Educational	ORG	Non-Profit Organization
GOV	US Government	ARPA	Old style Arpanet
INT	International	NATO	Nato field
MIL	US Military

Department names can be extended to name machines:
- cs.purdue.edu (subdomain name) =>
  xinu.cs.purdue.edu
Machine names are added (xinu) to further subdivide the domain names used.

Named Items:

To allow clients to distinguish among multiple kinds of entries, each named item (stored in the system) is assigned a type which specifies if it is an:
- Address of a machine
- Mailbox
- User
Name resolving must also include the type of answer desired (specifying the protocol family is optional).
The DNS partitions the entire set of names by class (for mapping to multiple protocol suites).
Naming items is required since one cannot distinguish the names of subdomains from the names of individual objects or their types.

Mapping Domain Names to Addresses:

The DNS also includes an efficient, reliable, general purpose, distributed system for mapping names to addresses using an independent cooperative system called name servers.
Names Servers - are server programs that translate names-to-addresses (maps DN => IP addresses) and usually executes on a dedicated processor.
Name Resolvers - client software that uses one or more name servers in getting a mapped name.
Domain name servers are arranged in a conceptual tree structure that corresponds to the naming hierarchy

Each node may actually contain several hierarchical levels for mapping.
Example:
- xinu.cs.purdue.edu
  (uses two name servers for mapping)

Domain Name Resolution:

Conceptually, name resolution proceeds in a top-down fashion.
Name resolution can occur in one of two different ways:
- Recursive Resolution:
  - A client requests complete translation.
  - The name server itself contacts further name servers for resolution (if need be).
- Iterative Resolution:
  - Only a single resolution is made and returned (not recursive).
  - Clients must now explicitly contact different name servers if further resolution is needed.
Name servers use name caching to optimize search costs (efficient, but may not always be accurate).
Time to Live (TTL) is used to determine a guaranteed name binding during it’s time interval. When time expires, the cache name binding is no longer valid, so the client must make a direct name resolution request once again.

Domain Server Message Format:

Messages are sent between domain clients and domain servers with a specific format.
All messages of this format are used for name resolution and naming queries.
Question sent by the client and answers provided by the server are included within different fields of the same message.
If a server cannot answer a request, it will include information of which server can.

0	16	31

IDENTIFICATION	PARAMETER
NUMBER OF QUESTIONS	NUMBER OF ANSWER
NUMBER OF AUTHORITY	NUMBER OF ADDITIONAL
QUESTION SECTION ...
ANSWER SECTION ...
AUTHORITY SECTION ...
ADDITIONAL INFORMATION SECTION ...

IDENTIFICATION (16-bits):
- Unique value used by the client to match responses to queries.
PARAMETER (16-bits):
- Specifies the operation requested and a response code.
NUMBER OF (16-bits each):
- Each give a count of entries in the corresponding sections that occur later in the message.
Example:
- NUMBER OF QUESTIONS - gives the count of entries that appear in the QUESTION SECTION of the message.

0	16	31

IDENTIFICATION	PARAMETER
NUMBER OF QUESTIONS	NUMBER OF ANSWER
NUMBER OF AUTHORITY	NUMBER OF ADDITIONAL
QUESTION SECTION ...
ANSWER SECTION ...
AUTHORITY SECTION ...
ADDITIONAL INFORMATION SECTION ...

QUERY SECTION (variable length):
- Contains queries for which answers are desired.
The client only fills in the question section while the server returns those questions and the answers in its response
Each question consists of the subfields:
- QUERY DOMAIN NAME - variable length subfield containing the domain name.
- QUERY TYPE - 16-bits which encode the type of question (machine name, e-mail addr., etc.)
- QUERY CLASS - 16-bits allows domain names to be used for arbitrary objects because official Internet names are only one possible class.
ANSWER, AUTHORITY, and ADDITIONAL INFORMATION SECTION (all variable length):
- Consist of a set of resource records that describe domain names and mappings for each specific area.
Each resource record describes one domain name.

Resource Record:

0	16	31

RESOURCE DOMAIN NAME ...
TYPE	CLASS
TIME TO LIVE	RESOURCE DATA LENGTH
RESOURCE DATA ...

Resource Record Format:

RESOURCE DOMAIN NAME (variable length):
- Contains the domain name to which this resource record refers.
TYPE (16-bits):
- Specifies the type of the data included in the resource record.
CLASS (16-bits):
- Specifies the data’s class.
TIME TO LIVE (16-bits):
- Contains an integer that specifies the number of seconds information in this resource record can be cached.
RESOURCE DATA LENGTH (16-bits):
- Length in octets of the name binding.
RESOURCE DATA (variable length):
- Contains the result of the binding.

Shortened Message Formats:

Compressed Name Format:
- To conserve space in the name binding reply packet, the name servers compress names by storing only one copy of each domain name in the message.
- This is done by pointing to similar name domains within the same compressed message.
Abbreviation:
- Provides a method of shortening names when the resolving process can supply part of the name automatically.
- Local network administrators configure a list of possible suffixes to be appended to these names during lookup.
Ex: dzt8474@hertz => hertz.njit.edu

Inverse Mappings:

Inverse Queries - allow the client to ask a server to perform a domain name mapping ‘backwards’.
The answer is provided to the server and if an appropriate question can be generated, it is returned.
They are not generally used since there is no way to resolve the question without having a server search all the other servers (inefficient).

Pointer Query:

Pointer Queries are the inverse mapping technique that is used instead of the former method.
Pointer Query - is a special question which requests the name server to return the correct domain name for the machine with the specified IP address.
The (in-addr.arpa) extension is used within DNS to request reverse name mappings.
The initial IP address is represented in dotted-decimal notation:
- aaa.bbb.ccc.ddd
The client forms a query by rearranging the notation into a string of the form:
- ddd.ccc.bbb.aaa.in-addr.arpa
DNS servers then contact the appropriate servers which can efficiently resolve this mapping.

Object Types:

The domain naming system is quite general in that it can be used for arbitrary hierarchical names.
DNS can be used for translating a domain name to a mail exchanger address.
Type: MX - mail exchanger is used with DNS to denote the correct type of name resolution. (user@domain-part)
Mail systems use the DNS to resolve the domain-part of an e-mail address using MX.
Most data queried in DNS is type A, which consists of a host name and it’s IP address.
Other types include:
- HINFO: Name of CPU and Operating Sys.
- MINFO: Mailbox or mail list information.
- NS: Name of authoritative server for domain
- PTR: Domain name of symbolic link.
- TXT: Uninterpreted string of ASCII text.

Domain Name System (DNS) LESSON 13

(October 18, 1995) Daniel Z. Tabor Jr. New Jersey Institute of Technology

Providing Readable Names:

Flat Namespace:

Hierarchical Names:

Subset Authority:

TCP/IP Internet Domain Names:

(Un)Official Internet Domain Names:

Geographical Domain Names:

Organizational Domain Names:

Named Items:

Mapping Domain Names to Addresses:

Domain Name Resolution:

Domain Server Message Format:

Resource Record:

Resource Record Format:

Shortened Message Formats:

Inverse Mappings:

Pointer Query:

Object Types:

Domain Name System (DNS)
LESSON 13

(October 18, 1995)
Daniel Z. Tabor Jr.
New Jersey Institute of Technology