TS Supplement (Includes RFC-1006 Support)

Introduction to the Tandem TCP/IP Subsystem

Introduction to RFC-1006

1–12 107751 Tandem Computers Incorporated

Every TCP/IP transport endpoint (socket) is addressable by the combination of an

Internet protocol (IP) address and a port number.

The connection between two applications can be identified by:

The Internet address of the local node

The port number of the application on the local node

The Internet address of the remote node

The port number of the application on the remote node

Servers use ports 1 through 1023. Clients use ports 1024 and up.

In the Internet, port numbers 1 through 255 are preassigned. These are called well-

known ports; they are used by servers to identify widely used applications called

well-known services. TCP port 102 is reserved for OSI transport services

implementing the RFC 1006 standard. (TCP/IP literature commonly refers to port 102

as being reserved for ISO-TSAP.)

The well-known port is used between the two host computers to identify which

application program is to receive incoming traffic. When using RFC-1006, clients

specify well-known port 102 in connection requests. Servers listen on well-known port

102 for any incoming connection requests.

Internet Address

An Internet protocol (IP) address is a four-octet (32-bit) numeric value, composed of

two parts: the network address, and the local address. The network address portion

identifies a particular network, and the local address portion identifies a local host on

that network.

An IP address is usually specified in a dotted decimal notation, which represents 4

octets in ASCII, separated by periods; for example, 133.50.85.43. Each octet value is a

number in the range 0 through 255. In OSI/AS and OSI/TS, an IP address can also be

specified in an eight-digit (32-bits) hexadecimal string (the format used to enter an

NSAP address); for example, 08000124, representing a class A address (network ID 8

and host ID 124 in hexadecimal). Section 3, “RFC-1006 Changes to the SCF, SPI, and

PTrace Subsystems,” explains when to use each IP address format.

An IP address is classified as class A, B, or C, depending on how the 32 bits are

divided between the network address and the local address. Figure 1-9 shows the

formats of the different IP address classes.