[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]


To: Pekka Savola <pekkas@netcore.fi>
CC: Ralph Droms <rdroms@cisco.com>, dnsop@cafax.se
From: Masataka Ohta <mohta@necom830.hpcl.titech.ac.jp>
Date: Sat, 9 Aug 2003 08:08:48 +0859 ()
In-Reply-To: <Pine.LNX.4.44.0308081438500.2900-100000@netcore.fi> from PekkaSavola at "Aug 8, 2003 02:43:50 pm"
Sender: owner-dnsop@cafax.se
Subject: Re: the well-known address approach

Pekka Savola;

> The argument I've heard a lot is one of architectural purity, as the 
> address architecture states (for example):

The proper interpretation of the address architecture RFC
is that IPv6 WG does not know anything about address (not only
anycast one) architecture.

The RFC says:

> An IPv6 unicast address refers to a single interface.

> When a unicast address is assigned to more than one interface,

without defining "assign".

The confusion, seemingly, is a result of an attempot to purposelessly
distinguish a subnet and a link.

Purposeful requirements are that a subnet, which is identical to
a link, may have multiple address prefixes and that a link may
be divided into virtual links such as VLAN, which is purely
an L2 issue.

> 2.6 Anycast Addresses

>    An IPv6 anycast address is an address that is assigned to more than
>    one interface (typically belonging to different nodes),

>    are syntactically indistinguishable from unicast addresses.  When a
>    unicast address is assigned to more than one interface, thus turning
>    it into an anycast address,

If a host (or a router or a node or whatever) have a unicast address
assigned to an interface, the host must receive a packet to the address,
even if the packet is received on another interface. In addition,
if a host have a unicast address assigned to an interface, the host
may send a packet from the address, even if the packet is sent from
another interface.

That is, all the unicast addresses of a multi-interface host are
assigned to more than one interface and are anycast ones.

> Needless to say I disagree with this model... :-)

Yup.

						Masataka Ohta
#----------------------------------------------------------------------
# To unsubscribe, send a message to <dnsop-request@cafax.se>.

Home | Date list | Subject list