Create UPD socket (dynamically allocated port number)

http://www.cs.rutgers.edu/~pxk/417/notes/sockets/udp.html

Step 1. Create a socket

A socket, s, is created with the socket system call:

int s = socket(domain, type, protocol)

All the parameters as well as the return value are integers:

domain, or address family —

communication domain in which the socket should be created. Some of address families are AF_INET (IP), AF_INET6 (IPv6), AF_UNIX (local channel, similar to pipes), AF_ISO (ISO protocols), and AF_NS (Xerox Network Systems protocols).

type —

type of service. This is selected according to the properties required by the application: SOCK_STREAM (virtual circuit service), SOCK_DGRAM (datagram service), SOCK_RAW (direct IP service). Check with your address family to see whether a particular service is available.

protocol —

indicate a specific protocol to use in supporting thesockets operation. This is useful in cases where some families may have more than one protocol to support a given type of service. The return value is a file descriptor (a small integer). The analogy of creating a socket is that of requesting a telephone line from the phone company.

For UDP/IP sockets, we want to specify the IP address family (AF_INET) and datagram service (SOCK_DGRAM). Since there's only one form of datagram service, there are no variations of the protocol, so the last argument, protocol, is zero. 

Step 2. Indentify (name) a socket

The system call for bind is:

#include <sys/socket.h>intbind(int socket, const struct sockaddr *address, socklen_t address_len);

The first parameter, socket, is the socket that was created with the socket system call.

For the second parameter, the structure sockaddr is a generic container that just allows the OS to be able to read the first couple of bytes that identify the address family. The address family determines what variant of the sockaddr struct to use that contains elements that make sense for that specific communication type. For IP networking, we use struct sockaddr_in, which is defined in the header netinet/in.h. This structure defines:

struct sockaddr_in { __uint8_t sin_len; sa_family_t sin_family; in_port_t sin_port; struct in_addr sin_addr; char sin_zero[8];};

Before calling bind, we need to fill out this structure. The three key parts we need to set are:

sin_family

The address family we used when we set up the socket. In our case, it's AF_INET.

sin_port

The port number (the transport address). You can explicitly assign a transport address (port) or allow the operating system to assign one. If you're a client and don't need a well-known port that others can use to locate you (since they will only respond to your messages), you can just let the operating system pick any available port number by specifying port 0. If you're a server, you'll generally pick a specific number since clients will need to know a port number to which to address messages.

sin_addr

The address for this socket. This is just your machine's IP address. With IP, your machine will have one IP address for each network interface. For example, if your machine has bothWi-Fi and ethernet connections, that machine will have two addresses, one for each interface. Most of the time, we don't care to specify a specific interface and can let the operating system use whatever it wants. The special address for this is 0.0.0.0, defined by the symbolic constantINADDR_ANY.

Since the address structure may differ based on the type of transport used, the third parameter specifies the length of that structure. This is simply the size of the internet address structure, sizeof(struct sockaddr_in).