TCP/IPv6 Configuration and Management Manual (G06.29+, H06.03+, J06.03+)

IPv6 Fundamentals
HP NonStop TCP/IPv6 Configuration and Management Manual524523-012
A-18
How IPv6 Tunnels Work
Unlike ARP, neighbor discovery detects half-link failures (using neighbor
unreachability detection) and avoids sending traffic to neighbors with which two-
way connectivity is absent.
Unlike IPv4 router discovery, the router advertisement messages do not contain a
preference field. The preference field is not needed to handle routers of different
stability; the neighbor unreachability detection will detect dead routers and switch
traffic to a working one.
The use of link-local addresses to uniquely identify routers (for router
advertisement and redirect messages) allows for hosts to maintain the router
associations in the event of the site renumbering to use new global prefixes.
Placing address resolution at the ICMP layer makes the protocol more media-
independent than ARP and allows the use of standard IP authentication.
How IPv6 Tunnels Work
The TCP/IPv6 subsystem includes support for both automatic tunnels and configured
tunnels. A tunnel is configured as a SUBNET object that has its type set to TUNNEL.
Unlike SUBNETs of type ETHERNET, a tunnel SUBNET does not have a direct
association with a LAN adapter.
The tunnel SUBNET TUN is the automatic tunnel. Each TCP6MON has a copy of the
automatic tunnel SUBNET and only one automatic tunnel per TCP6MON is allowed.
The automatic tunnel’s default state is down and not configured for IPv6. To enable
IPv6 on the automatic tunnel requires that at least one other Ethernet SUBNET
supporting IPv4/IPv6 exist. When the first SUBNET configured for IPv4/IPv6 is added,
the automatic tunnel enters the STARTED state and automatically enables IPv6 on the
tunnel SUBNET.
A behavior of the automatic tunnel makes it preferable to use configured tunnels in
many cases. Automatic tunnels associate themselves with the first available SUBNET.
Because of this association with a single SUBNET, anomalies can occur whereby the
other end of a tunnel might not connect back through the source SUBNET. For this
reason, configured tunnels are preferable because both ends uniquely define the
endpoints so the same SUBNET used for transmission is used for reception.
A configured tunnel requires the local endpoint IPv4 address and the destination
endpoint’s IPv4 address as part of the ADD SUBNET command. For this command to
succeed, an IPv4 SUBNET with the local endpoint IPv4 address must first be
configured and a default route to the destination endpoint must also be configured. See
DUAL-Mode, Configured Tunnel
on page 5-5 for examples of configuring a tunnel
SUBNET.
Unlike the automatic tunnel SUBNET, more than one configured-tunnel SUBNET can
exist. Configured-tunnel SUBNET names must begin with IPT.