NAME
inet6
—
Internet protocol version 6
family
SYNOPSIS
#include
<sys/types.h>
#include <netinet/in.h>
DESCRIPTION
The inet6
family is an updated version of
inet(4) family. While
inet(4) implements Internet Protocol version 4,
inet6
implements Internet Protocol version 6.
inet6
is a collection of
protocols layered atop the
Internet Protocol
version 6 (IPv6) transport layer, and using the IPv6 address format.
The inet6
family provides protocol support for the
SOCK_STREAM
, SOCK_DGRAM
, and
SOCK_RAW
socket types; the
SOCK_RAW
interface provides access to the IPv6
protocol.
ADDRESSING
IPv6 addresses are 16 byte 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 use the
following addressing structure:
struct sockaddr_in6 { uint8_t sin6_len; sa_family_t sin6_family; in_port_t sin6_port; uint32_t sin6_flowinfo; struct in6_addr sin6_addr; uint32_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.
The IPv6 specification defines scoped addresses, like link-local or site-local addresses. A scoped address is ambiguous to the kernel, if it is specified without a scope identifier. To manipulate scoped addresses properly from the userland, programs must use the advanced API defined in RFC 2292. A compact description of the advanced API is available in ip6(4). If a scoped address is specified without an explicit scope, the kernel may raise an error. Note that scoped addresses are not for daily use at this moment, both from a specification and an implementation point of view.
The KAME implementation supports an extended numeric IPv6 address
notation for link-local addresses, like
“fe80::1%de0
” to specify
“fe80::1
on de0
interface”. This notation is supported by
getaddrinfo(3) and
getnameinfo(3). Some of normal userland programs, such as
telnet(1) or ftp(1), are able to use this notation. With special programs
like ping6(8), you can specify the outgoing interface by an extra command
line option to disambiguate scoped addresses.
Scoped addresses are handled specially in the kernel. In kernel
structures like routing tables or interface structures, a scoped address
will have its interface index embedded into the address. Therefore, the
address in some kernel structures is not the same as that on the wire. The
embedded index will become visible through a
PF_ROUTE
socket, kernel memory accesses via
kvm(3) and on some other occasions. HOWEVER, users should never use
the embedded form. For details please consult
http://www.kame.net/dev/cvsweb2.cgi/kame/IMPLEMENTATION.
Note that the above URL describes the situation with the latest KAME tree,
not the NetBSD tree.
PROTOCOLS
The inet6
family comprises the IPv6
network protocol, Internet Control Message Protocol version 6 (ICMPv6),
Transmission Control Protocol (TCP), and User Datagram Protocol (UDP). TCP
is used to support the SOCK_STREAM
abstraction while
UDP is used to support the SOCK_DGRAM
abstraction.
Note that TCP and UDP are common to
inet(4) and inet6
. 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.
Interaction between IPv4/v6 sockets
By default, NetBSD does not route IPv4
traffic to AF_INET6
sockets. The default behavior
intentionally violates RFC 2553 for security reasons. Listen to two sockets
if you want to accept both IPv4 and IPv6 traffic. IPv4 traffic may be routed
with certain per-socket/per-node configuration, however, it is not
recommended to do so. Consult
ip6(4) for details.
The behavior of AF_INET6
TCP/UDP socket is
documented in RFC 2553. Basically, it says this:
- A specific bind on an
AF_INET6
socket (bind(2) with an address specified) should accept IPv6 traffic to that address only. - If you perform a wildcard bind on an
AF_INET6
socket (bind(2) to IPv6 address::
), and there is no wildcard bindAF_INET
socket on that TCP/UDP port, IPv6 traffic as well as IPv4 traffic should be routed to thatAF_INET6
socket. IPv4 traffic should be seen as if it came from an IPv6 address like::ffff:10.1.1.1
. This is called an IPv4 mapped address. - If there are both a wildcard bind
AF_INET
socket and a wildcard bindAF_INET6
socket on one TCP/UDP port, they should behave separately. IPv4 traffic should be routed to theAF_INET
socket and IPv6 should be routed to theAF_INET6
socket.
However, RFC 2553 does not define the ordering constraint between
calls to bind(2), nor how IPv4 TCP/UDP port numbers and IPv6 TCP/UDP port
numbers relate to each other (should they be integrated or separated).
Implemented behavior is very different from kernel to kernel. Therefore, it
is unwise to rely too much upon the behavior of
AF_INET6
wildcard bind sockets. It is recommended to
listen to two sockets, one for AF_INET
and another
for AF_INET6
, when you would like to accept both
IPv4 and IPv6 traffic.
It should also be noted that malicious parties can take advantage
of the complexity presented above, and are able to bypass access control, if
the target node routes IPv4 traffic to AF_INET6
socket. Users are advised to take care handling connections from IPv4 mapped
address to AF_INET6
sockets.
SEE ALSO
ioctl(2), socket(2), sysctl(3), icmp6(4), intro(4), ip6(4), tcp(4), udp(4)
Qing Li, Tatuya Jinmei, and Keiichi Shima, IPv6 Core Protocols Implementation, Morgan Kaufmann, 2006.
Qing Li, Tatuya Jinmei, and Keiichi Shima, IPv6 Advanced Protocols Implementation, Morgan Kaufmann, 2007.
STANDARDS
Tatuya 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.
HISTORY
The inet6
protocol interfaces are defined
in RFC 2553 and RFC 2292. The implementation described herein appeared in
the WIDE/KAME project.
BUGS
The IPv6 support is subject to change as the Internet protocols develop. Users should not depend on details of the current implementation, but rather the services exported.
Users are suggested to implement “version
independent” code as much as possible, as you will need to support
both inet(4) and inet6
.