TOC |
|
This document describes one of the network models that are designed for smooth transition to IPv6. It is called NAT444 model. NAT444 model is composed of IPv6, and IPv4 with Large Scale NAT (LSN).
NAT444 is the only scheme not to require replacing Customer Premises Equipment (CPE) even if IPv4 address exhausted. But it must be noted that NAT444 has serious restrictions i.e. it limits the number of sessions per CPE so that rich applications such as AJAX and RSS feed cannot work well.
Therefore, IPv6 which is free from such a difficulty has to be introduced into the network at the same time. In other words, NAT444 is just a tool to make IPv6 transition easy to be swallowed. It is designed for the days IPv4 and IPv6 co-existence.
This Internet-Draft is submitted in full conformance with the provisions of BCP 78 and BCP 79.
Internet-Drafts are working documents of the Internet Engineering Task Force (IETF). Note that other groups may also distribute working documents as Internet-Drafts. The list of current Internet-Drafts is at http://datatracker.ietf.org/drafts/current/.
Internet-Drafts are draft documents valid for a maximum of six months and may be updated, replaced, or obsoleted by other documents at any time. It is inappropriate to use Internet-Drafts as reference material or to cite them other than as “work in progress.”
This Internet-Draft will expire on July 22, 2011.
Copyright (c) 2011 IETF Trust and the persons identified as the document authors. All rights reserved.
This document is subject to BCP 78 and the IETF Trust's Legal Provisions Relating to IETF Documents (http://trustee.ietf.org/license-info) in effect on the date of publication of this document. Please review these documents carefully, as they describe your rights and restrictions with respect to this document. Code Components extracted from this document must include Simplified BSD License text as described in Section 4.e of the Trust Legal Provisions and are provided without warranty as described in the Simplified BSD License.
1.
Introduction
2.
Definition of NAT444 Model
3.
Behavior of NAT444 Model
4.
Pros and Cons of NAT444 Model
4.1.
Pros of NAT444 Model
4.2.
Cons of NAT444 Model
5.
Acknowledgements
6.
IANA Considerations
7.
Security Considerations
8.
References
8.1.
Normative References
8.2.
Informative References
Appendix A.
Example IPv6 Transition Scenario
§
Authors' Addresses
TOC |
The only permanent solution of the IPv4 address exhaustion is to deploy IPv6. Now, just before the exhaustion, it's time to make a transition to IPv6.
After the exhaustion, unless ISP takes any action, end users will not be able to get IPv4 address.
The servers that have only IPv4 address will continue to exist on the Internet after the IPv4 address exhaustion. In this situation, IPv6 only hosts cannot reach IPv4 only hosts.
This document explains NAT444 model that bridges the gap between the coming IPv6 Internet and the present IPv4 Internet.
TOC |
NAT444 Model is a network model that uses two Network Address and Port Translators (NAPTs) with three types of IPv4 address blocks.
The first NAPT is in CPE, and the second NAPT is in Large Scale NAT (LSN) (Yamagata, I., Miyakawa, S., Nakagawa, A., and H. Ashida, “Common requirements for IP address sharing schemes,” October 2010.) [I‑D.ietf‑behave‑lsn‑requirements]. LSN is supposed to be installed in the ISP's network.
(The IPv4 Internet) (The IPv6 Internet) | | +---------+ | IPv4 Global Address | | +--------+--------+ | | LSN | | +--------+--------+ | IPv4 | | IPv6 +-------------+ Dual Stack | +---------------+----------------+ | IPv4 NAT/IPv6 Dual Stack CPE | +---------------+----------------+ IPv4 Private Address / | IPv6 Dual Stack | +-----------+-------------+ |IPv4/IPv6 Dual Stack host| +-------------------------+
The first IPv4 address block is Private Address (Rekhter, Y., Moskowitz, R., Karrenberg, D., Groot, G., and E. Lear, “Address Allocation for Private Internets,” February 1996.) [RFC1918] inside CPE. The second one is an IPv4 Address block between CPEs and LSN. The third one is IPv4 Global Addresses that is outside LSN. The ISPs using NAT444 provide IPv6 connectivity by dual stack model.
TOC |
The IPv6 packets from the host reach the IPv6 Internet without using NAT functionality.
The following figure shows the behavior of the IPv4 packet from the host to the IPv4 server via two NATs. The first NAT in CPE overwrites the Source IP Address and Source Port from 10.0.0.2:tt to w.w.w.w:uu. Then the second NAT in LSN overwrites them from w.w.w.w:uu to y.y.y.y:vv. Destination IP Address and Port are not overwritten.
+-------------+ (Port=80) | IPv4 Server | ^ x.x.x.x-> +------+------+ : | : IPv4 Global Address | : | : (The IPv4 Internet):(Dst=x.x.x.x:80/Src=y.y.y.y:vv) | : IPv4 Global Address | : | : y.y.y.y-> +----+----+ : (Port=vv) | LSN | ^ z.z.z.z-> +----+----+ : | : IPv4 Address | :(Dst=x.x.x.x:80/Src=w.w.w.w:uu) | : w.w.w.w-> +-------+-------+ : (Port=uu) | IPv4 NAT CPE | ^ 10.0.0.1-> +-------+-------+ : | : IPv4 Private Address| : | : 10.0.0.2-> +----+----+ :(Dst=x.x.x.x:80/Src=10.0.0.2:tt) (Port=tt) |IPv4 Host| +---------+
The following figure explains the behavior of returning IPv4 packet via two NATs. The first NAT in LSN overwrites the Destination IP Address and Port Number from y.y.y.y:vv to w.w.w.w:uu. Then the second NAT in CPE overwrites them from w.w.w.w:u to 10.0.0.2:tt.
+-------------+ (Port=80) | IPv4 Server | : x.x.x.x-> +------+------+ : | : IPv4 Global Address | : | : (The IPv4 Internet):(Dst=y.y.y.y:vv/Src=x.x.x.x:80) | : IPv4 Global Address | : | : y.y.y.y-> +----+----+ : (Port=vv) | LSN | v z.z.z.z-> +----+----+ : | : IPv4 Address | :(Dst=w.w.w.w:uu/Src=x.x.x.x:80) | : w.w.w.w-> +-------+-------+ : (Port=uu) | IPv4 NAT CPE | v 10.0.0.1-> +-------+-------+ : | : IPv4 Private Address | :(Dst=10.0.0.2:tt/Src=x.x.x.x:80) | : 10.0.0.2-> +----+----+ : (Port=tt) |IPv4 Host| v +---------+
TOC |
TOC |
This network model has following advantages.
- This is the only network model that doesn't require replacing CPEs those are owned by customers.
- This network model is composed of the present technology.
- This network model doesn't require address family translation.
- This network model doesn't require DNS rewriting.
- This network model doesn't require additional fragment for the packets because it doesn't use tunneling technology.
TOC |
This network model has some technical restrictions.
- Some application such as SIP requires special treatment,
because IP address is written in the payload of the packet.
Special treatment means application itself aware double NAPT
or both of two NAPTs support inspecting and rewriting the packets.
- Because both IPv4 route and IPv6 route exist,
it doubles the number of IGP route inside the LSN.
- UPnP doesn't work with double NAPTs.
TOC |
Thanks for the input and review by Shin Miyakawa, Shirou Niinobe, Takeshi Tomochika, Tomohiro Fujisaki, Dai Nishino, JP address community members, AP address community members and JPNIC members.
TOC |
There are no IANA considerations.
TOC |
Each customer inside a LSN looks using the same Global Address from outside an ISP. In case of incidents, the ISP must have the function to trace back the record of each customer's access without using only IP address.
If a Global Address of the LSN is listed on the blacklist, other customers who share the same address could be affected.
TOC |
TOC |
[RFC1918] | Rekhter, Y., Moskowitz, R., Karrenberg, D., Groot, G., and E. Lear, “Address Allocation for Private Internets,” BCP 5, RFC 1918, February 1996 (TXT). |
[RFC4925] | Li, X., Dawkins, S., Ward, D., and A. Durand, “Softwire Problem Statement,” RFC 4925, July 2007 (TXT). |
[I-D.ietf-behave-lsn-requirements] | Yamagata, I., Miyakawa, S., Nakagawa, A., and H. Ashida, “Common requirements for IP address sharing schemes,” draft-ietf-behave-lsn-requirements-00 (work in progress), October 2010 (TXT). |
TOC |
[I-D.shirasaki-isp-shared-addr] | Yamagata, I., Miyakawa, S., Nakagawa, A., Yamaguchi, J., and H. Ashida, “ISP Shared Address,” draft-shirasaki-isp-shared-addr-05 (work in progress), September 2010 (TXT). |
[I-D.shirasaki-nat444-isp-shared-addr] | Shirasaki, Y., Miyakawa, S., Nakagawa, A., Yamaguchi, J., and H. Ashida, “NAT444 addressing models,” draft-shirasaki-nat444-isp-shared-addr-04 (work in progress), July 2010 (TXT). |
TOC |
The steps of IPv6 transition are as follows.
Step 1: Enabling softwire client in host
ISP provides IPv6 connectivity to customers with softwire (Li, X., Dawkins, S., Ward, D., and A. Durand, “Softwire Problem Statement,” July 2007.) [RFC4925]. ISP installs LSN and softwire concentrator in its network. A softwire client in host connects to the IPv6 internet via ISP's concentrator. ISP can use existing IPv4 equipments. Customers can just use existing CPE.
(The IPv4 Internet) (The IPv6 Internet) | | IPv6 | +-----------+-----------+ | | Softwire Concentrator | | +-----------+-----------+ +---------+----------+ ^ IPv4 Global Address | : +----------+----------+ : | LSN | : +----------+----------+ : Any IPv4 Address | : IPv6 over IPv4 Softwire (ISP Network) | : (e.g. IPv6 over IPv4 L2TP) +----------+----------+ : | IPv4 NAT only CPE | : +----------+----------+ : IPv4 Private Address | v +---------------+-----------------+ |IPv4/IPv6 Softwire Client in host| +---------------------------------+
Step 2: Enabling softwire client in CPE
A customer enables softwire client in CPE. A softwire client in CPE connects to the IPv6 internet via ISP's concentrator. A Customer's network is now dual stack.
(The IPv4 Internet) (The IPv6 Internet) | | IPv6 | +----------+------------+ | | Softwire Concentrator | | +----------+------------+ +---------+------------+ ^ IPv4 Global Address | : +----------+------------+ : | LSN | : IPv6 over IPv4 Softwire +----------+------------+ : (e.g. IPv6 over IPv4 L2TP) Any IPv4 Address | : (ISP Network) | v +---------------+--------------------+ |IPv4 NAT/IPv6 Softwire client in CPE| +---------------+--------------------+ IPv4 Private Address / | IPv6 Dual Stack | +-----------+-------------+ |IPv4/IPv6 Dual Stack host| +-------------------------+
Step 3: Moving on to dual stack
ISP provides dual stack access to CPE. A CPE uplink is now dual stack.
(The IPv4 Internet) (The IPv6 Internet) | | +---------+ | IPv4 Global Address | | +--------+--------+ | | LSN | | IPv6 +--------+--------+ | Any IPv4 Address / | | IPv6 Dual Stack +-------------+ (ISP Network) | +---------------+----------------+ | IPv4 NAT/IPv6 Dual Stack CPE | +---------------+----------------+ IPv4 Private Address / | IPv6 Dual Stack | +-----------+-------------+ |IPv4/IPv6 Dual Stack host| +-------------------------+
Step 4: Moving on to pure IPv6
IPv6 transition completes.
(The IPv6 Internet) | IPv6 | +--------+----------+ | IPv6 CPE | +--------+----------+ IPv6 | +--------+----------+ | IPv6 host | +-------------------+
TOC |
Ikuhei Yamagata | |
NTT Communications Corporation | |
Granpark Tower 17F, 3-4-1 Shibaura, Minato-ku | |
Tokyo 108-8118 | |
Japan | |
Phone: | +81 3 6733 8671 |
Email: | ikuhei@nttv6.jp |
Yasuhiro Shirasaki | |
NTT Communications Corporation | |
NTT Hibiya Bldg. 7F, 1-1-6 Uchisaiwai-cho, Chiyoda-ku | |
Tokyo 100-8019 | |
Japan | |
Phone: | +81 3 6700 8530 |
Email: | yasuhiro@nttv6.jp |
Akira Nakagawa | |
Japan Internet Exchange Co., Ltd. (JPIX) | |
Otemachi Building 21F, 1-8-1 Otemachi, Chiyoda-ku | |
Tokyo 100-0004 | |
Japan | |
Phone: | +81 90 9242 2717 |
Email: | a-nakagawa@jpix.ad.jp |
Jiro Yamaguchi | |
Internet Initiative Japan Inc. | |
Kakyoin Square Bldg., 15F, 1-1-20 Kakyoin, Aoba-ku | |
Sendai 980-0013 | |
Japan | |
Phone: | +81 22 216 5650 |
Email: | jiro-y@iij.ad.jp |
Hiroyuki Ashida | |
its communications Inc. | |
3-5-7 Hisamoto Takatsu-ku Kawasaki-shi | |
Kanagawa 213-0011 | |
Japan | |
Email: | ashida@itscom.ad.jp |