[dhcwg] dhcpv6-24: use of anycast

Thomas Narten <narten@us.ibm.com> Wed, 08 May 2002 15:01 UTC

Received: from optimus.ietf.org (ietf.org [] (may be forged)) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA10745 for <dhcwg-archive@odin.ietf.org>; Wed, 8 May 2002 11:01:31 -0400 (EDT)
Received: (from daemon@localhost) by optimus.ietf.org (8.9.1a/8.9.1) id LAA03669 for dhcwg-archive@odin.ietf.org; Wed, 8 May 2002 11:01:40 -0400 (EDT)
Received: from optimus.ietf.org (localhost []) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA03011; Wed, 8 May 2002 10:59:18 -0400 (EDT)
Received: from ietf.org (odin []) by optimus.ietf.org (8.9.1a/8.9.1) with ESMTP id KAA02981 for <dhcwg@optimus.ietf.org>; Wed, 8 May 2002 10:59:16 -0400 (EDT)
Received: from e21.nc.us.ibm.com (e21.nc.us.ibm.com []) by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10616 for <dhcwg@ietf.org>; Wed, 8 May 2002 10:59:07 -0400 (EDT)
Received: from southrelay02.raleigh.ibm.com (southrelay02.raleigh.us.ibm.com []) by e21.nc.us.ibm.com (8.12.2/8.12.2) with ESMTP id g48ExF64029672 for <dhcwg@ietf.org>; Wed, 8 May 2002 10:59:15 -0400
Received: from rotala.raleigh.ibm.com (rotala.raleigh.ibm.com []) by southrelay02.raleigh.ibm.com (8.11.1m3/NCO/VER6.1) with ESMTP id g48Ex5Q235052 for <dhcwg@ietf.org>; Wed, 8 May 2002 10:59:05 -0400
Received: from rotala.raleigh.ibm.com (narten@localhost) by rotala.raleigh.ibm.com (8.11.6/8.11.6) with ESMTP id g48ExTa19129 for <dhcwg@ietf.org>; Wed, 8 May 2002 10:59:29 -0400
Message-Id: <200205081459.g48ExTa19129@rotala.raleigh.ibm.com>
To: dhcwg@ietf.org
Date: Wed, 08 May 2002 10:59:29 -0400
From: Thomas Narten <narten@us.ibm.com>
Subject: [dhcwg] dhcpv6-24: use of anycast
Sender: dhcwg-admin@ietf.org
Errors-To: dhcwg-admin@ietf.org
X-Mailman-Version: 1.0
Precedence: bulk
List-Id: <dhcwg.ietf.org>
X-BeenThere: dhcwg@ietf.org

Seems to me that the anycast text is a bit underspecified. It's also
not clear to me exactly what problem the anycast text is trying to

> 13. Transmission of messages by a client
>    Unless otherwise specified, a client sends DHCP messages to the
>    All_DHCP_Relay_Agents_and_Servers or the DHCP_Anycast address.

This is ambiguous. What should clients actually implement by default?
This document should make it clear what clients should implement. For
example try one and if it doesn't work, try the other? Leaving when to
use anycast to the implementor without guidelines will likely lead to
interoperability problems, with different implementations doing
different things.

>    If the client is attached to a link that supports multicast
>    transmission, the client sends DHCP messages to the
>    All_DHCP_Relay_Agents_and_Servers address.  If the client is

My understanding is that all IPv6 links support multicast by
definition. If the underlying network doesn't, it must be
emulated. Note that multicast is needed for Neighbor Discovery, which
all IPv6 nodes have to implement.

>    When a client sends a DHCP message to the DHCP_Anycast address, it
>    MUST use an address assigned to the interface for which the client
>    is interested in obtaining configuration, which is suitable for use
>    by the server in responding to the client, as the source address in
>    the header of the IP datagram.  See "Default Address Selection for
>    IPv6" [4] for more details.

Don't you also need to say that anycast can only be used if the client
already has a properly scoped address? It seems to me that anycast
isn't very useful for the *client*. I.e, let it always send to the
relay agent and let the relay agent deal with anycast vs. mcast.

Or, is this specifically restricted to the information-request
message? If so, might be good to say that.

In private mail, Ralph asked:

> Good question to ask...How about words to the effect of "if the link is 
> connected to an NBMA link, use anycast; otherwise, use multicast"?

What is an NBMA link? Who defines what that is? I'm not sure there is
a well-defined definition of NBMA link.


dhcwg mailing list