[v6ops] prefix delegation maintenance

"Templin, Fred L" <Fred.L.Templin@boeing.com> Thu, 29 October 2015 22:43 UTC

From: "Templin, Fred L" <Fred.L.Templin@boeing.com>
To: "v6ops@ietf.org" <v6ops@ietf.org>
Thread-Topic: prefix delegation maintenance
Thread-Index: AdESmiZnsvYnpgYqToKdQ/Y16A41Hw==
Date: Thu, 29 Oct 2015 22:43:15 +0000
Message-ID: <2134F8430051B64F815C691A62D9831832F34040@XCH-BLV-504.nw.nos.boeing.com>
Accept-Language: en-US
Content-Language: en-US
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: base64
MIME-Version: 1.0
Archived-At: <http://mailarchive.ietf.org/arch/msg/v6ops/MkV0AZAubxxlk2oov9SYTjQGhq4>
Subject: [v6ops] prefix delegation maintenance
Precedence: list

Hi, there seems to be a notion that an IPv6 prefix can be delegated via
IPv6 ND by having the router include PIOs in RAs that have (A=1; L=0)
instead of using DHCPv6 PD. But, the two are not the same.

A properly delegated prefix is one where the network grants a prefix that
the recipient obtains for its own exclusive use, and the network establishes
routing state that associates the prefix with the recipient. But, receiving the
prefix is only the initial phase of an ongoing relationship the recipient must
have with the network.

In particular, the recipient must take periodic and/or event-driven  actions
to ensure that the prefix delegation remains alive. DHCPv6 PD provides
messages such as Renew and Rebind for this. The recipient must also tell
the network when the prefix is no longer needed, and DHCPv6 PD has a
Release message for this. The recipient is further responsible for reliability
and must retransmit if the network does not respond to initial messages,
so the recipient acts on its own behalf to make sure the prefix delegation
persists. I don't see any of these facilities specified for IPv6 ND.

So, I would rather see us use the tool that was meant for the job rather
than reinvent the wheel.

Thanks - Fred
fred.l.templin@boeing.com

[v6ops] prefix delegation maintenance Templin, Fred L