NAME
inet6
—
Internet protocol version 6
family
SYNOPSIS
#include
<sys/types.h>
#include <netinet/in.h>
DESCRIPTION
The Internet Protocol version 6 family is an updated version of
the Internet Protocol version 4 family. It comprises Internet Protocol
version 6 (IPv6), Internet Control Message Protocol version 6 (ICMPv6),
Transmission Control Protocol (TCP), and User Datagram Protocol (UDP).
tcp(4) is used to support the SOCK_STREAM
abstraction while udp(4) is used to support the
SOCK_DGRAM
abstraction. A raw interface to IPv6 is
available by creating an Internet socket of type
SOCK_RAW
. The ICMPv6 message protocol is accessible
from a raw socket.
IPv6 addresses are 128-bit quantities, stored in network standard
byteorder. The include file
<netinet/in.h>
defines this
address as a discriminated union.
Sockets bound to the inet6
family utilize
the following addressing structure:
struct sockaddr_in6 { u_int8_t sin6_len; sa_family_t sin6_family; in_port_t sin6_port; u_int32_t sin6_flowinfo; struct in6_addr sin6_addr; u_int32_t sin6_scope_id; };
Sockets may be created with the local address “::”
(which is equal to IPv6 address 0:0:0:0:0:0:0:0
) to
effect “wildcard” matching on incoming messages.
For security reasons, OpenBSD does not
route IPv4 traffic to an AF_INET6
socket, and does
not support IPv4 mapped addresses, where IPv4 traffic is seen as if it comes
from an IPv6 address like “::ffff:10.1.1.1”. Where both IPv4
and IPv6 traffic need to be accepted, bind and listen on two sockets.
Global addresses utilise the first 48 bits of the address for the routing prefix. The next 16 bits designate the subnet, and the final 64 bits are used as a host identifier.
The IPv6 specification also defines link-local addresses, which are scoped. A scoped address is ambiguous to the kernel if it is specified without a scope identifier. To manipulate scoped addresses properly from userland, programs must use the advanced API defined in RFC 3542. A compact description of the advanced API is available in ip6(4). If scoped addresses are specified without explicit scope, the kernel may raise an error.
KAME supports an extended numeric IPv6 address notation for link-local addresses, such as “fe80::1%de0” to specify “fe80::1” on the “de0” interface. This notation is supported by getaddrinfo(3) and getnameinfo(3), as well as userland programs such as telnet(1) and ftp(1).
Scoped addresses are handled specially in the kernel. In kernel
structures like routing tables or interface structures, scoped addresses
have their interface index embedded into the address. Therefore the address
on some kernel structures is not the same as that on the wire. The embedded
index will be visible on PF_ROUTE
sockets, kernel
memory access via kvm(3), and some other occasions. HOWEVER, users should never
use the embedded form.
IPv6 SETUP
Generally speaking, IPv6 connectivity is achieved in a fashion
similar to that for IPv4. For native IPv6 setup, routers attach to the
network either manually or using autoconf
to connect
to an ISP; hosts receive an address prefix from a router advertisement
daemon such as rad(8) and use autoconf
for stateless
address configuration (SLAAC). For setups which tunnel IPv6 over IPv4, see
gif(4).
The INET6 and TUNNEL sections of ifconfig(8) contain information relevant to IPv6 setups; settings can be made permanent using hostname.if(5) files. Routers need to set the net.inet6.ip6.forwarding sysctl(2).
SEE ALSO
socket(2), icmp6(4), ip6(4), tcp(4), udp(4), hostname.if(5), ifconfig(8), rad(8)
STANDARDS
Tatsuya Jinmei and Atsushi Onoe, An Extension of Format for IPv6 Scoped Addresses, internet draft, draft-ietf-ipngwg-scopedaddr-format-02.txt, June 2000, work in progress material.
R. Gilligan, S. Thomson, J. Bound, J. McCann, and W. Stevens, Basic Socket Interface Extensions for Ipv6, RFC 3493, February 2003.
W. Stevens, M. Thomas, E. Nordmark, and T. Jinmei, Advanced Sockets Application Programming Interface (API) for IPv6, RFC 3542, May 2003.
HISTORY
The implementation described herein appeared in WIDE/KAME project.
CAVEATS
It is advisable to explicitly reject all packets to your network not used by any of your interface prefixes. Otherwise packets that have a destination address belonging to your network may be routed back to your provider via the default route. Set a reject route for your assigned prefix:
# route add -net 2001:db8::/48 ::1
-reject