bind(2) - f38

1bind(2)                       System Calls Manual                      bind(2)
2
3
4

NAME

6       bind - bind a name to a socket
7

LIBRARY

9       Standard C library (libc, -lc)
10

SYNOPSIS

12       #include <sys/socket.h>
13
14       int bind(int sockfd, const struct sockaddr *addr,
15                socklen_t addrlen);
16

DESCRIPTION

18       When a socket is created with socket(2), it exists in a name space (ad‐
19       dress family) but has no address assigned to it.   bind()  assigns  the
20       address  specified  by  addr  to the socket referred to by the file de‐
21       scriptor sockfd.  addrlen specifies the size, in bytes, of the  address
22       structure  pointed to by addr.  Traditionally, this operation is called
23       “assigning a name to a socket”.
24
25       It is normally necessary to assign a local address using bind()  before
26       a SOCK_STREAM socket may receive connections (see accept(2)).
27
28       The  rules used in name binding vary between address families.  Consult
29       the manual entries in Section 7 for detailed information.  For AF_INET,
30       see  ip(7);  for  AF_INET6,  see ipv6(7); for AF_UNIX, see unix(7); for
31       AF_APPLETALK, see ddp(7); for AF_PACKET, see packet(7); for AF_X25, see
32       x25(7); and for AF_NETLINK, see netlink(7).
33
34       The  actual  structure  passed for the addr argument will depend on the
35       address family.  The sockaddr structure is defined as something like:
36
37           struct sockaddr {
38               sa_family_t sa_family;
39               char        sa_data[14];
40           }
41
42       The only purpose of this structure is to  cast  the  structure  pointer
43       passed  in  addr in order to avoid compiler warnings.  See EXAMPLES be‐
44       low.
45

RETURN VALUE

47       On success, zero is returned.  On error, -1 is returned, and  errno  is
48       set to indicate the error.
49

ERRORS

51       EACCES The address is protected, and the user is not the superuser.
52
53       EADDRINUSE
54              The given address is already in use.
55
56       EADDRINUSE
57              (Internet  domain sockets) The port number was specified as zero
58              in the socket address structure, but, upon attempting to bind to
59              an  ephemeral  port,  it was determined that all port numbers in
60              the ephemeral port range are currently in use.  See the  discus‐
61              sion of /proc/sys/net/ipv4/ip_local_port_range ip(7).
62
63       EBADF  sockfd is not a valid file descriptor.
64
65       EINVAL The socket is already bound to an address.
66
67       EINVAL addrlen  is  wrong,  or  addr  is  not  a valid address for this
68              socket's domain.
69
70       ENOTSOCK
71              The file descriptor sockfd does not refer to a socket.
72
73       The following errors are specific to UNIX domain (AF_UNIX) sockets:
74
75       EACCES Search permission is denied on a component of the  path  prefix.
76              (See also path_resolution(7).)
77
78       EADDRNOTAVAIL
79              A  nonexistent  interface was requested or the requested address
80              was not local.
81
82       EFAULT addr points outside the user's accessible address space.
83
84       ELOOP  Too many symbolic links were encountered in resolving addr.
85
86       ENAMETOOLONG
87              addr is too long.
88
89       ENOENT A component in the directory prefix of the socket pathname  does
90              not exist.
91
92       ENOMEM Insufficient kernel memory was available.
93
94       ENOTDIR
95              A component of the path prefix is not a directory.
96
97       EROFS  The socket inode would reside on a read-only filesystem.
98

STANDARDS

100       POSIX.1-2008.
101

HISTORY

103       POSIX.1-2001, SVr4, 4.4BSD (bind() first appeared in 4.2BSD).
104

BUGS

106       The transparent proxy options are not described.
107

EXAMPLES

109       An  example  of  the  use of bind() with Internet domain sockets can be
110       found in getaddrinfo(3).
111
112       The following example shows how to bind a stream  socket  in  the  UNIX
113       (AF_UNIX) domain, and accept connections:
114
115       #include <stdio.h>
116       #include <stdlib.h>
117       #include <string.h>
118       #include <sys/socket.h>
119       #include <sys/un.h>
120       #include <unistd.h>
121
122       #define MY_SOCK_PATH "/somepath"
123       #define LISTEN_BACKLOG 50
124
125       #define handle_error(msg) \
126           do { perror(msg); exit(EXIT_FAILURE); } while (0)
127
128       int
129       main(void)
130       {
131           int                 sfd, cfd;
132           socklen_t           peer_addr_size;
133           struct sockaddr_un  my_addr, peer_addr;
134
135           sfd = socket(AF_UNIX, SOCK_STREAM, 0);
136           if (sfd == -1)
137               handle_error("socket");
138
139           memset(&my_addr, 0, sizeof(my_addr));
140           my_addr.sun_family = AF_UNIX;
141           strncpy(my_addr.sun_path, MY_SOCK_PATH,
142                   sizeof(my_addr.sun_path) - 1);
143
144           if (bind(sfd, (struct sockaddr *) &my_addr,
145                    sizeof(my_addr)) == -1)
146               handle_error("bind");
147
148           if (listen(sfd, LISTEN_BACKLOG) == -1)
149               handle_error("listen");
150
151           /* Now we can accept incoming connections one
152              at a time using accept(2). */
153
154           peer_addr_size = sizeof(peer_addr);
155           cfd = accept(sfd, (struct sockaddr *) &peer_addr,
156                        &peer_addr_size);
157           if (cfd == -1)
158               handle_error("accept");
159
160           /* Code to deal with incoming connection(s)... */
161
162           if (close(sfd) == -1)
163               handle_error("close");
164
165           if (unlink(MY_SOCK_PATH) == -1)
166               handle_error("unlink");
167       }
168