From owner-ngtrans@sunroof.eng.sun.com  Mon May  1 16:22:48 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05181
	for <ngtrans-archive@odin.ietf.org>; Mon, 1 May 2000 16:22:46 -0400 (EDT)
Received: from engmail3.Eng.Sun.COM ([129.144.170.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id OAA04737;
	Mon, 1 May 2000 14:20:50 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id NAA21418;
	Mon, 1 May 2000 13:19:02 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e41KGiG15307
	for ngtrans-dist; Mon, 1 May 2000 13:16:44 -0700 (PDT)
Received: from engmail4.Eng.Sun.COM (engmail4 [129.144.134.6])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e41KGU715300
	for <ngtrans@sunroof.eng.sun.com>; Mon, 1 May 2000 13:16:35 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id NAA08981;
	Mon, 1 May 2000 13:16:28 -0700 (PDT)
Received: from mail-gw.hursley.ibm.com (mail-gw.hursley.ibm.com [194.196.110.15])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id NAA17662;
	Mon, 1 May 2000 13:16:27 -0700 (PDT)
Received: from sp3at21.hursley.ibm.com (sp3at21.hursley.ibm.com [9.20.45.21]) by mail-gw.hursley.ibm.com (AIX4.3/UCB 8.8.8/8.8.8) with ESMTP id VAA21080; Mon, 1 May 2000 21:16:25 +0100
Received: from hursley.ibm.com (gsine02.us.sine.ibm.com [9.14.6.42]) by sp3at21.hursley.ibm.com (AIX4.2/UCB 8.7/8.7.3) with ESMTP id VAA24218; Mon, 1 May 2000 21:16:22 +0100 (BST)
Message-ID: <390DE0AE.6B7E02DB@hursley.ibm.com>
Date: Mon, 01 May 2000 14:53:18 -0500
From: Brian E Carpenter <brian@hursley.ibm.com>
Organization: IBM
X-Mailer: Mozilla 4.61 [en] (Win98; I)
X-Accept-Language: en,fr
MIME-Version: 1.0
To: itojun@iijlab.net
CC: Alain Durand <Alain.Durand@eng.sun.com>,
        NGtrans List <ngtrans@sunroof.eng.sun.com>
Subject: Re: (ngtrans) 6to4 issue: link local
References: <4736.956371939@coconut.itojun.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Brian E Carpenter <brian@hursley.ibm.com>
Content-Transfer-Encoding: 7bit

I'm still struggling with this. 6to4 does exactly what draft-ietf-ngtrans-mech-06.txt
does, by reference. If this problem exists for a 6to4 router, why doesn't it exist for
any IPv6 router connected by a tunnel? In both cases the underlying Interface ID 
on the virtual interface of the router is the IPv4 address. I don't see any difference.

   Brian

itojun@iijlab.net wrote:
> 
> >> However, I'm not convinced that the benefits are worth the burden of having an
> >> extra address that will almost never be used.
> >> Could we say that a Link Local is not needed today and if we need it later,
> >> (say for multicast routing to work) we will define it then?
> >If it is needed even once, e.g. for ICMPv6 redirect as itojun suggested,
> >we need to define it.
> 
>         sorry if my analysis was unclear.
>         no, ICMPv6 redirect cannot be used even if you assign link-local
>         onto 6to4 interface.  what we really need is link-local address
>         AND mapping from link-local to IPv4 address.
> 
>         to use ICMPv6, the following conditions must be satisfied:
>         - "gateway" field on routing table needs to be link-local on
>           originating host of the packet (due to validation rules in RFC2461
>           section 8, IIRC)
>         - source address on ICMPv6 packet needs to be link-local of router
> 
>         If you configure the following routing table on the originating host:
> 
> destination     gateway         interface
> default         fe80::x         6to4
> 
>         you'll need to know mapping from fe80::x to IPv4 destination address
>         to send packet via router on fe80::x%6to4 (I love this notation :-)).
>         This is not possible since we do not have mapping between fe80::x
>         to IPv4 address, in 6to4 spec.
> 
>         if we are to use IPv4 address embedded into the last 4 bytes in
>         interface identifier field... hmm, I have never thought about it...
>         we will see two IPv4 address on 6to4 address (2002::/16).
> 
>         I'm not sure if we have real use of ICMPv6 redirect here.
>         on 6to4 cloud, most of the nodes are routers (6to4 relays),
>         not hosts.  routers do not update their routing table on inbound
>         ICMPv6 packets.
> 
>         To summarize:
>         - not sure if ICMPv6 has real use on 6to4 cloud
>         - to use ICMPv6 on 6to4 cloud, we need link-local address AND
>           mapping from link-local to IPv4 address.  ICMPv6 does not work
>           if we don't have the mapping defined.
> 
> itojun


From owner-ngtrans@sunroof.eng.sun.com  Mon May  1 16:27:57 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05398
	for <ngtrans-archive@odin.ietf.org>; Mon, 1 May 2000 16:27:56 -0400 (EDT)
Received: from engmail1.Eng.Sun.COM ([129.146.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id OAA09177;
	Mon, 1 May 2000 14:26:45 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id NAA18688;
	Mon, 1 May 2000 13:26:28 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e41KPm315345
	for ngtrans-dist; Mon, 1 May 2000 13:25:48 -0700 (PDT)
Received: from engmail4.Eng.Sun.COM (engmail4 [129.144.134.6])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e41KPf715338
	for <ngtrans@sunroof.eng.sun.com>; Mon, 1 May 2000 13:25:41 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id NAA10915
	for <ngtrans@sunroof.eng.sun.com>; Mon, 1 May 2000 13:25:39 -0700 (PDT)
Received: from mail-gw.hursley.ibm.com (mail-gw.hursley.ibm.com [194.196.110.15])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id OAA08250
	for <ngtrans@sunroof.eng.sun.com>; Mon, 1 May 2000 14:25:38 -0600 (MDT)
Received: from sp3at21.hursley.ibm.com (sp3at21.hursley.ibm.com [9.20.45.21]) by mail-gw.hursley.ibm.com (AIX4.3/UCB 8.8.8/8.8.8) with ESMTP id VAA54400; Mon, 1 May 2000 21:25:22 +0100
Received: from hursley.ibm.com (gsine02.us.sine.ibm.com [9.14.6.42]) by sp3at21.hursley.ibm.com (AIX4.2/UCB 8.7/8.7.3) with ESMTP id VAA14150; Mon, 1 May 2000 21:25:19 +0100 (BST)
Message-ID: <390DE2C0.5B40EA3@hursley.ibm.com>
Date: Mon, 01 May 2000 15:02:08 -0500
From: Brian E Carpenter <brian@hursley.ibm.com>
Organization: IBM
X-Mailer: Mozilla 4.61 [en] (Win98; I)
X-Accept-Language: en,fr
MIME-Version: 1.0
To: "Klein, Eric" <Eric_Klein@starhome.com>
CC: Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        moore@cs.utk.edu
Subject: Re: (ngtrans) draft-ietf-ngtrans-6to4-04.txt
References: <CE835E918749D21191B10060084C377E01E15B6B@ismail1.icomverse.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Brian E Carpenter <brian@hursley.ibm.com>
Content-Transfer-Encoding: 7bit

Eric,

No, we specifically forbid the use of RFC 1918 addresses to form the 6to4 prefix,
and this will be even more clear in the next version. The whole point of 6to4
is to create global prefixes, which by definition can't be done with RFC 1918
addresses.

This isn't a problem, because even if a customer only has one globally unique
IPv4 address, and all the rest are RFC 1918 addresses, that one is sufficient to 
generate all the IPv6 addresses the customer is ever going to need.

IPv6 has no equivalent of RFC 1918 addresses, and doesn't need them since
there is plenty of address space for everybody. The myth that private
addresses guarantee privacy will not be propagated in IPv6.

   Brian

"Klein, Eric" wrote:
> 
> Brian,
> 
> This draft it states:
>         5.8 Usage within Intranets
> 
>            There is nothing to stop the above scenario being deployed within
> a
>            private corporate network as part of its internal transition to
> IPv6;
>            the corporate IPv4 backbone would serve as the virtual link layer
> for
>            individual corporate sites using 2002:: prefixes.  The V4ADDR
> MUST be
>            a duly allocated global IPv4 address, which MUST be unique within
> the
>            private network. The Intranet thereby obtains globally unique
> IPv6
>            addresses even if it is internally using private IPv4 addresses
> [RFC
>            1918].
> 
> This may seem like a silly question, but is there an allocation of addresses
> to be used strictly for private networks similar to what is specified in
> RFC 1918        Address Allocation for Private Internets   February 1996
> 
> 3. Private Address Space
> 
>            The Internet Assigned Numbers Authority (IANA) has reserved the
>            following three blocks of the IP address space for private
> internets:
> 
>              10.0.0.0        -   10.255.255.255  (10/8 prefix)
>              172.16.0.0      -   172.31.255.255  (172.16/12 prefix)
>              192.168.0.0     -   192.168.255.255 (192.168/16 prefix)
> 
>            We will refer to the first block as "24-bit block", the second as
>            "20-bit block", and to the third as "16-bit" block. Note that (in
>            pre-CIDR notation) the first block is nothing but a single class
> A
>            network number, while the second block is a set of 16 contiguous
>            class B network numbers, and third block is a set of 256
> contiguous
>            class C network numbers.
> 
> This is a very valid need for private networks that are currently using
> these private networks, and wish them to remain private. There needs to be
> some method for both converting these on an IPv4 network that is
> transitioning to IPv6 network and in the final IPv6 network.


From owner-ngtrans@sunroof.eng.sun.com  Mon May  1 16:45:47 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05919
	for <ngtrans-archive@odin.ietf.org>; Mon, 1 May 2000 16:45:46 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id OAA21758;
	Mon, 1 May 2000 14:44:16 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id NAA15661;
	Mon, 1 May 2000 13:43:46 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e41KgeQ15375
	for ngtrans-dist; Mon, 1 May 2000 13:42:40 -0700 (PDT)
Received: from engmail4.Eng.Sun.COM (engmail4 [129.144.134.6])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e41KgX715368
	for <ngtrans@sunroof.eng.sun.com>; Mon, 1 May 2000 13:42:34 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id NAA15377;
	Mon, 1 May 2000 13:42:28 -0700 (PDT)
Received: from sfo.erg.sri.com (sfo.erg.sri.com [128.18.100.4])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id OAA20404;
	Mon, 1 May 2000 14:42:28 -0600 (MDT)
Received: from erg.sri.com (walleye.erg.sri.com [128.18.4.214])
	by sfo.erg.sri.com (8.9.1/8.9.0) with ESMTP id NAA21729;
	Mon, 1 May 2000 13:42:13 -0700 (PDT)
Message-ID: <390DEC26.BF4E9E67@erg.sri.com>
Date: Mon, 01 May 2000 13:42:15 -0700
From: "Fred L. Templin" <templin@erg.sri.com>
Organization: SRI International
X-Mailer: Mozilla 4.7 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Brian E Carpenter <brian@hursley.ibm.com>
CC: itojun@iijlab.net, Alain Durand <Alain.Durand@eng.sun.com>,
        NGtrans List <ngtrans@sunroof.eng.sun.com>
Subject: Re: (ngtrans) 6to4 issue: link local
References: <4736.956371939@coconut.itojun.org> <390DE0AE.6B7E02DB@hursley.ibm.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: "Fred L. Templin" <templin@erg.sri.com>
Content-Transfer-Encoding: 7bit

Hello,

I've not wanted to divert this discussion, but I'd like to point
out that my current personal draft entitled:

 draft-templin-ngtrans-v6v4compat-00.txt

presents an alternate IPv6-IPv4 compatibility address form
that *does* retain the embedded IPv4 address in its link-local
and site-local derivatives. (Some of my research efforts have
entailed using these link and site local derivatives as
source and destination IPv6 addresses for nodes attached to
a common link/site.)

A big disclaimer, however, is that this work is not yet
accepted as 'ngtrans' working group material. I am in the
process of developing a set of generalized sending rules to
show that this address form has application for automatic
IPv6-over-IPv4 tunnel creation, but until these rules are
developed and reviewed this is still just an individual
submission proposal. I believe the addressing architecture
described in my draft has the potential to complement and
augment the existing 6to4 addressing architecture, however.

Fred Templin
templin@erg.sri.com

Brian E Carpenter wrote:
> 
> I'm still struggling with this. 6to4 does exactly what draft-ietf-ngtrans-mech-06.txt
> does, by reference. If this problem exists for a 6to4 router, why doesn't it exist for
> any IPv6 router connected by a tunnel? In both cases the underlying Interface ID
> on the virtual interface of the router is the IPv4 address. I don't see any difference.
> 
>    Brian
> 
> itojun@iijlab.net wrote:
> >
> > >> However, I'm not convinced that the benefits are worth the burden of having an
> > >> extra address that will almost never be used.
> > >> Could we say that a Link Local is not needed today and if we need it later,
> > >> (say for multicast routing to work) we will define it then?
> > >If it is needed even once, e.g. for ICMPv6 redirect as itojun suggested,
> > >we need to define it.
> >
> >         sorry if my analysis was unclear.
> >         no, ICMPv6 redirect cannot be used even if you assign link-local
> >         onto 6to4 interface.  what we really need is link-local address
> >         AND mapping from link-local to IPv4 address.
> >
> >         to use ICMPv6, the following conditions must be satisfied:
> >         - "gateway" field on routing table needs to be link-local on
> >           originating host of the packet (due to validation rules in RFC2461
> >           section 8, IIRC)
> >         - source address on ICMPv6 packet needs to be link-local of router
> >
> >         If you configure the following routing table on the originating host:
> >
> > destination     gateway         interface
> > default         fe80::x         6to4
> >
> >         you'll need to know mapping from fe80::x to IPv4 destination address
> >         to send packet via router on fe80::x%6to4 (I love this notation :-)).
> >         This is not possible since we do not have mapping between fe80::x
> >         to IPv4 address, in 6to4 spec.
> >
> >         if we are to use IPv4 address embedded into the last 4 bytes in
> >         interface identifier field... hmm, I have never thought about it...
> >         we will see two IPv4 address on 6to4 address (2002::/16).
> >
> >         I'm not sure if we have real use of ICMPv6 redirect here.
> >         on 6to4 cloud, most of the nodes are routers (6to4 relays),
> >         not hosts.  routers do not update their routing table on inbound
> >         ICMPv6 packets.
> >
> >         To summarize:
> >         - not sure if ICMPv6 has real use on 6to4 cloud
> >         - to use ICMPv6 on 6to4 cloud, we need link-local address AND
> >           mapping from link-local to IPv4 address.  ICMPv6 does not work
> >           if we don't have the mapping defined.
> >
> > itojun


From owner-ngtrans@sunroof.eng.sun.com  Mon May  1 19:57:22 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA08489
	for <ngtrans-archive@odin.ietf.org>; Mon, 1 May 2000 19:57:22 -0400 (EDT)
Received: from engmail2.Eng.Sun.COM ([129.146.1.25])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id RAA26662;
	Mon, 1 May 2000 17:56:27 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id QAA24687;
	Mon, 1 May 2000 16:56:06 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e41Nt4c15529
	for ngtrans-dist; Mon, 1 May 2000 16:55:04 -0700 (PDT)
Received: from engmail4.Eng.Sun.COM (engmail4 [129.144.134.6])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e41Nsu715522
	for <ngtrans@sunroof.eng.sun.com>; Mon, 1 May 2000 16:54:56 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id QAA29201;
	Mon, 1 May 2000 16:54:54 -0700 (PDT)
Received: from coconut.itojun.org (coconut.itojun.org [210.160.95.97])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id QAA15828;
	Mon, 1 May 2000 16:54:52 -0700 (PDT)
Received: from kiwi.itojun.org (localhost.itojun.org [127.0.0.1])
	by coconut.itojun.org (8.9.3+3.2W/3.7W) with ESMTP id IAA05615;
	Tue, 2 May 2000 08:54:14 +0900 (JST)
To: Brian E Carpenter <brian@hursley.ibm.com>
cc: Alain Durand <Alain.Durand@eng.sun.com>,
        NGtrans List <ngtrans@sunroof.eng.sun.com>
In-reply-to: brian's message of Mon, 01 May 2000 14:53:18 EST.
      <390DE0AE.6B7E02DB@hursley.ibm.com>
X-Template-Reply-To: itojun@itojun.org
X-Template-Return-Receipt-To: itojun@itojun.org
X-PGP-Fingerprint: F8 24 B4 2C 8C 98 57 FD  90 5F B4 60 79 54 16 E2
Subject: Re: (ngtrans) 6to4 issue: link local
From: itojun@iijlab.net
Date: Tue, 02 May 2000 08:54:14 +0900
Message-ID: <5613.957225254@coconut.itojun.org>
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: itojun@iijlab.net


>I'm still struggling with this. 6to4 does exactly what draft-ietf-ngtrans-mech-06.txt
>does, by reference. If this problem exists for a 6to4 router, why doesn't it exist for
>any IPv6 router connected by a tunnel? In both cases the underlying Interface ID 
>on the virtual interface of the router is the IPv4 address. I don't see any difference.

	I'm not just talking about link-local address format, but also worried
	about complication due to sending mechanism.

	For other tunnelling mechanisms (like RFC1933 configured tunnel)
	we configure IPv4 address of the peer separately to the router.
	We always know the peer's IPv4 address if we determine outgoing
	interface.

	For 6to4, the only place we can get IPv4 address for a particular
	6to4 desination is from 6to4 prefix (3rd to 6th bytes).  My
	understanding is that we can transmit packets across 6to4 cloud, if
	and only if the destination address (or gateway) belongs to 6to4
	prefix (2002::/16).

itojun


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 02:37:04 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA25288
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 02:37:03 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id AAA17928;
	Tue, 2 May 2000 00:36:11 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id XAA09352;
	Mon, 1 May 2000 23:35:21 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e426YPh15778
	for ngtrans-dist; Mon, 1 May 2000 23:34:25 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e426YI715771
	for <ngtrans@sunroof.eng.sun.com>; Mon, 1 May 2000 23:34:18 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id XAA20512
	for <ngtrans@sunroof.eng.sun.com>; Mon, 1 May 2000 23:34:17 -0700 (PDT)
Received: from alpha.netvision.net.il (alpha.netvision.net.il [194.90.1.13])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id XAA28101
	for <ngtrans@sunroof.eng.sun.com>; Mon, 1 May 2000 23:34:14 -0700 (PDT)
Received: from sendout.icomverse.com (Efrat-FR3.ser.netvision.net.il [199.203.174.65])
	by alpha.netvision.net.il (8.9.3/8.8.6) with ESMTP id JAA32406;
	Tue, 2 May 2000 09:33:00 +0300 (IDT)
Received: from ismail1.icomverse.com (ismail1.icomverse.com [190.190.110.2])
	by sendout.icomverse.com (8.9.3/8.8.7) with ESMTP id IAA22198;
	Tue, 2 May 2000 08:34:34 +0300
Received: by ismail1.icomverse.com with Internet Mail Service (5.5.2650.21)
	id <JHFAWB1K>; Tue, 2 May 2000 09:33:28 +0300
Message-ID: <CE835E918749D21191B10060084C377E01E15BED@ismail1.icomverse.com>
From: "Klein, Eric" <Eric_Klein@starhome.com>
To: Brian E Carpenter <brian@hursley.ibm.com>,
        "Klein, Eric"
	 <Eric_Klein@starhome.com>
Cc: Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        moore@cs.utk.edu
Subject: RE: (ngtrans) private networks under 6to4-04
Date: Tue, 2 May 2000 09:33:26 +0300 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: "Klein, Eric" <Eric_Klein@starhome.com>

Brian, 

You misunderstand. I am talking about designs where there is no direct
registered IPv4 address. For example I am working on a project where it is
100 sites of all private network all using RFC 1918 addresses. These sites
may be connected to routers that have registered Ipv4 or IPv6 addresses used
on them, but I need to assure that the private traffic is always isolated
from the public net.

Using RFC 1918 on an IPv4 network this is no problem as even if the private
network connection should go down and routing tables try to use the default
route to the public network, those routers will discard this traffic because
it is a  RFC 1918 address as the destination.

Under IPv6 I can find no private network allocation. is this because it is
assumed that all addresses will be registered even to the desktop? If so
then the whole concept of private networks it thrown out the window. I don't
think that this is a valid assumption.

Eric

-----Original Message-----
From: Brian E Carpenter [mailto:brian@hursley.ibm.com]
Sent: Monday, May 01, 2000 11:02 PM
To: Klein, Eric
Cc: Sowmini Varadhan; ngtrans@sunroof.eng.sun.com; moore@cs.utk.edu
Subject: Re: (ngtrans) draft-ietf-ngtrans-6to4-04.txt


Eric,

No, we specifically forbid the use of RFC 1918 addresses to form the 6to4
prefix,
and this will be even more clear in the next version. The whole point of
6to4
is to create global prefixes, which by definition can't be done with RFC
1918
addresses.

This isn't a problem, because even if a customer only has one globally
unique
IPv4 address, and all the rest are RFC 1918 addresses, that one is
sufficient to 
generate all the IPv6 addresses the customer is ever going to need.

IPv6 has no equivalent of RFC 1918 addresses, and doesn't need them since
there is plenty of address space for everybody. The myth that private
addresses guarantee privacy will not be propagated in IPv6.

   Brian

"Klein, Eric" wrote:
> 
> Brian,
> 
> This draft it states:
>         5.8 Usage within Intranets
> 
>            There is nothing to stop the above scenario being deployed
within
> a
>            private corporate network as part of its internal transition to
> IPv6;
>            the corporate IPv4 backbone would serve as the virtual link
layer
> for
>            individual corporate sites using 2002:: prefixes.  The V4ADDR
> MUST be
>            a duly allocated global IPv4 address, which MUST be unique
within
> the
>            private network. The Intranet thereby obtains globally unique
> IPv6
>            addresses even if it is internally using private IPv4 addresses
> [RFC
>            1918].
> 
> This may seem like a silly question, but is there an allocation of
addresses
> to be used strictly for private networks similar to what is specified in
> RFC 1918        Address Allocation for Private Internets   February 1996
> 
> 3. Private Address Space
> 
>            The Internet Assigned Numbers Authority (IANA) has reserved the
>            following three blocks of the IP address space for private
> internets:
> 
>              10.0.0.0        -   10.255.255.255  (10/8 prefix)
>              172.16.0.0      -   172.31.255.255  (172.16/12 prefix)
>              192.168.0.0     -   192.168.255.255 (192.168/16 prefix)
> 
>            We will refer to the first block as "24-bit block", the second
as
>            "20-bit block", and to the third as "16-bit" block. Note that
(in
>            pre-CIDR notation) the first block is nothing but a single
class
> A
>            network number, while the second block is a set of 16
contiguous
>            class B network numbers, and third block is a set of 256
> contiguous
>            class C network numbers.
> 
> This is a very valid need for private networks that are currently using
> these private networks, and wish them to remain private. There needs to be
> some method for both converting these on an IPv4 network that is
> transitioning to IPv6 network and in the final IPv6 network.


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 03:09:52 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA25499
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 03:09:52 -0400 (EDT)
Received: from engmail3.Eng.Sun.COM ([129.144.170.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id BAA28719;
	Tue, 2 May 2000 01:08:49 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id AAA01559;
	Tue, 2 May 2000 00:08:20 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4277E115823
	for ngtrans-dist; Tue, 2 May 2000 00:07:14 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42778715816
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 00:07:08 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id AAA22564
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 00:07:08 -0700 (PDT)
Received: from alfa.itea.ntnu.no (alfa.itea.ntnu.no [129.241.18.10])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id AAA09334
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 00:07:07 -0700 (PDT)
Received: by alfa.itea.ntnu.no id JAA0000009063; Tue, 2 May 2000 09:04:20 +0200 (MET DST)
From: Stig Venås <venaas@alfa.itea.ntnu.no>
Date: Tue, 2 May 2000 09:04:19 +0200
To: "Klein, Eric" <Eric_Klein@starhome.com>
Cc: Brian E Carpenter <brian@hursley.ibm.com>,
        Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        moore@cs.utk.edu
Subject: Re: (ngtrans) private networks under 6to4-04
Message-ID: <20000502090419.A19838@itea.ntnu.no>
References: <CE835E918749D21191B10060084C377E01E15BED@ismail1.icomverse.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
X-Mailer: Mutt 1.0i
In-Reply-To: <CE835E918749D21191B10060084C377E01E15BED@ismail1.icomverse.com>; from Eric_Klein@starhome.com on Tue, May 02, 2000 at 09:33:26AM +0300
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Stig Venås <venaas@alfa.itea.ntnu.no>

On Tue, May 02, 2000 at 09:33:26AM +0300, Klein, Eric wrote:
> Brian, 
> 
> You misunderstand. I am talking about designs where there is no direct
> registered IPv4 address. For example I am working on a project where it is
> 100 sites of all private network all using RFC 1918 addresses. These sites
> may be connected to routers that have registered Ipv4 or IPv6 addresses used
> on them, but I need to assure that the private traffic is always isolated
> from the public net.
> 
> Using RFC 1918 on an IPv4 network this is no problem as even if the private
> network connection should go down and routing tables try to use the default
> route to the public network, those routers will discard this traffic because
> it is a  RFC 1918 address as the destination.

There are Site-Local addresses in IPv6, I think they should fulfill your
needs. See for instance RFC 2373.

Stig

-- 
Stig Venaas
UNINETT


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 11:04:40 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA02765
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 11:04:39 -0400 (EDT)
Received: from engmail1.Eng.Sun.COM ([129.146.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id JAA13895;
	Tue, 2 May 2000 09:02:28 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id IAA28529;
	Tue, 2 May 2000 08:01:04 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e42Exoh16081
	for ngtrans-dist; Tue, 2 May 2000 07:59:51 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42Exi716074
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 07:59:44 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id HAA26820
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 07:59:44 -0700 (PDT)
Received: from da1server.martin.fl.us (da1server.martin.fl.us [198.136.32.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id IAA11859
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 08:59:42 -0600 (MDT)
Received: from localhost (gmaxwell@localhost)
	by da1server.martin.fl.us (8.8.8/8.8.8) with SMTP id KAA18088;
	Tue, 2 May 2000 10:50:35 -0400 (EDT)
Date: Tue, 2 May 2000 10:50:35 -0400 (EDT)
From: Greg Maxwell <gmaxwell@martin.fl.us>
X-Sender: gmaxwell@da1server
To: "Klein, Eric" <Eric_Klein@starhome.com>
cc: Brian E Carpenter <brian@hursley.ibm.com>,
        Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        moore@cs.utk.edu
Subject: RE: (ngtrans) private networks under 6to4-04
In-Reply-To: <CE835E918749D21191B10060084C377E01E15BED@ismail1.icomverse.com>
Message-ID: <Pine.GSO.3.96.1000502104243.9125B-100000@da1server>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Greg Maxwell <gmaxwell@martin.fl.us>

On Tue, 2 May 2000, Klein, Eric wrote:

[snip]
> Under IPv6 I can find no private network allocation. is this because it is
> assumed that all addresses will be registered even to the desktop? If so
> then the whole concept of private networks it thrown out the window. I don't
> think that this is a valid assumption.

Why? 

Before Internet access was widespread and before RFC1918, large IP based
sites used registered prefixes on their disconnected networks.

You should not confuse private addressing with disconnected networks. When
you use RFC1918 your network is no more 'private' then if you were
connecting with public addressing, public connections are just obfuscated 
by the lack of global addressing.

With IPv6 you can use a public prefix on an otherwise disconnected
network. With this, you are no less private because disconnection
provides all the privacy that you are going to get via technoligical
means. Global addressing ensures that your network will continue to
function without renumbering when you need to peer with the disconnected
(or not?) network of your vendors, friends, or companies you merge with.

The importance of globally unique addressing is a core concept of IP and
is reemphasized with IPv6. If you are willing to live in the hellish world
of NATs opon NATs opon NATs then you could just as well stay with IPv4.   




From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 11:11:01 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA02954
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 11:10:59 -0400 (EDT)
Received: from engmail2.Eng.Sun.COM ([129.146.1.25])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id JAA18352;
	Tue, 2 May 2000 09:08:45 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id IAA28105;
	Tue, 2 May 2000 08:08:24 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e42F7kk16132
	for ngtrans-dist; Tue, 2 May 2000 08:07:46 -0700 (PDT)
Received: from engmail4.Eng.Sun.COM (engmail4 [129.144.134.6])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42F7a716125
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 08:07:36 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id IAA20536
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 08:07:36 -0700 (PDT)
Received: from astro.cs.utk.edu (ASTRO.CS.UTK.EDU [128.169.93.168])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id JAA17508
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 09:07:35 -0600 (MDT)
Received: from astro.cs.utk.edu (LOCALHOST [127.0.0.1])
        by astro.cs.utk.edu (cf 8.9.3) with ESMTP id LAA01094;
        Tue, 2 May 2000 11:03:36 -0400 (EDT)
Message-Id: <200005021503.LAA01094@astro.cs.utk.edu>
X-URI: http://www.cs.utk.edu/~moore/
From: Keith Moore <moore@cs.utk.edu>
To: Greg Maxwell <gmaxwell@martin.fl.us>
cc: "Klein, Eric" <Eric_Klein@starhome.com>,
        Brian E Carpenter <brian@hursley.ibm.com>,
        Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        moore@cs.utk.edu
Subject: Re: (ngtrans) private networks under 6to4-04 
In-reply-to: Your message of "Tue, 02 May 2000 10:50:35 EDT."
             <Pine.GSO.3.96.1000502104243.9125B-100000@da1server> 
X-SUBJECT-MSG-FROM: Greg Maxwell <gmaxwell@martin.fl.us> 
Date: Tue, 02 May 2000 11:03:36 -0400
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Keith Moore <moore@cs.utk.edu>

> With IPv6 you can use a public prefix on an otherwise disconnected
> network.

sure you can, but how are you going to get a public prefix?
normally you would get a prefix from your ISP, but if you're on
a disconnected network then you might not have an ISP.


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 11:43:53 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA03594
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 11:43:52 -0400 (EDT)
Received: from engmail3.Eng.Sun.COM ([129.144.170.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id JAA12897;
	Tue, 2 May 2000 09:41:10 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id IAA15244;
	Tue, 2 May 2000 08:37:37 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e42FaGi16180
	for ngtrans-dist; Tue, 2 May 2000 08:36:16 -0700 (PDT)
Received: from engmail4.Eng.Sun.COM (engmail4 [129.144.134.6])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42Fa9716173
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 08:36:09 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id IAA00048
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 08:36:08 -0700 (PDT)
Received: from alpha.netvision.net.il (alpha.netvision.net.il [194.90.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id JAA08866
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 09:35:50 -0600 (MDT)
Received: from sendout.icomverse.com (Efrat-FR3.ser.netvision.net.il [199.203.174.65])
	by alpha.netvision.net.il (8.9.3/8.8.6) with ESMTP id SAA17443;
	Tue, 2 May 2000 18:34:38 +0300 (IDT)
Received: from ismail1.icomverse.com (ismail1.icomverse.com [190.190.110.2])
	by sendout.icomverse.com (8.9.3/8.8.7) with ESMTP id RAA24558;
	Tue, 2 May 2000 17:36:11 +0300
Received: by ismail1.icomverse.com with Internet Mail Service (5.5.2650.21)
	id <JHFAWBNK>; Tue, 2 May 2000 18:35:06 +0300
Message-ID: <CE835E918749D21191B10060084C377E01E15C50@ismail1.icomverse.com>
From: "Klein, Eric" <Eric_Klein@starhome.com>
To: Keith Moore <moore@cs.utk.edu>, Greg Maxwell <gmaxwell@martin.fl.us>
Cc: "Klein, Eric" <Eric_Klein@starhome.com>,
        Brian E Carpenter
	 <brian@hursley.ibm.com>,
        Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com
Subject: RE: (ngtrans) private networks under 6to4-04 
Date: Tue, 2 May 2000 18:35:03 +0300 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: "Klein, Eric" <Eric_Klein@starhome.com>

Greg,

I agree with Keith, I am looking for private networks that will not be
connected to any ISP. Thus when I have my 100 sites I would then need to
register for my own public prefix as if I were an ISP. Rather than using a
private range for internal connectivity and registered addresses where
Interconnectivity would be required. Thus at most I would need 1 NAT
conversion with only one Registered address persite or per overall network.

Or is it your recommendation that every company go and request their own
Public prefix for every computer at their sites? If so then you loose any
gain in addresses, as every organization or company will have to request a
public prefix based on their anticipated growth at all sites. 

This would reduce the available addresses almost as fast as the current
IPv4.

Eric

-----Original Message-----
From: Keith Moore [mailto:moore@cs.utk.edu]
Sent: Tuesday, May 02, 2000 6:04 PM
To: Greg Maxwell
Cc: Klein, Eric; Brian E Carpenter; Sowmini Varadhan;
ngtrans@sunroof.eng.sun.com; moore@cs.utk.edu
Subject: Re: (ngtrans) private networks under 6to4-04 


> With IPv6 you can use a public prefix on an otherwise disconnected
> network.

sure you can, but how are you going to get a public prefix?
normally you would get a prefix from your ISP, but if you're on
a disconnected network then you might not have an ISP.


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 12:20:32 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA04292
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 12:20:31 -0400 (EDT)
Received: from engmail3.Eng.Sun.COM ([129.144.170.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id KAA12889;
	Tue, 2 May 2000 10:19:06 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id JAA24624;
	Tue, 2 May 2000 09:15:01 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e42GDjA16253
	for ngtrans-dist; Tue, 2 May 2000 09:13:45 -0700 (PDT)
Received: from engmail4.Eng.Sun.COM (engmail4 [129.144.134.6])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42GDc716246
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 09:13:39 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id JAA07942
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 09:13:37 -0700 (PDT)
Received: from mail2.microsoft.com (mail2.microsoft.com [131.107.3.124])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with SMTP id JAA26099
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 09:13:35 -0700 (PDT)
Received: from 157.54.9.104 by mail2.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 02 May 2000 09:07:26 -0700 (Pacific Daylight Time)
Received: by INET-IMC-02 with Internet Mail Service (5.5.2651.58)
	id <JW33Q2XC>; Tue, 2 May 2000 09:07:26 -0700
Message-ID: <4D0A23B3F74DD111ACCD00805F31D81022D4F894@RED-MSG-50>
From: Richard Draves <richdr@microsoft.com>
To: itojun@iijlab.net, Brian E Carpenter <brian@hursley.ibm.com>
Cc: Alain Durand <Alain.Durand@eng.sun.com>,
        NGtrans List
	 <ngtrans@sunroof.eng.sun.com>
Subject: RE: (ngtrans) 6to4 issue: link local
Date: Tue, 2 May 2000 09:07:22 -0700 
X-Mailer: Internet Mail Service (5.5.2651.58)
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Richard Draves <richdr@microsoft.com>

I agree with itojun, link-local addresses make no sense for 6to4. Sure you
could assign a link-local address to your 6to4 interface. But no one would
be able to send to it because there's no way to resolve the link-local
address into a link-layer (v4) address.

Link-local addresses do make sense for point-to-point configured tunnels,
because you know the other side's v4 address.

> -----Original Message-----
> From: itojun@iijlab.net [mailto:itojun@iijlab.net]
> Sent: Monday, May 01, 2000 4:54 PM
> To: Brian E Carpenter
> Cc: Alain Durand; NGtrans List
> Subject: Re: (ngtrans) 6to4 issue: link local
> 
> 
> 
> >I'm still struggling with this. 6to4 does exactly what 
> draft-ietf-ngtrans-mech-06.txt
> >does, by reference. If this problem exists for a 6to4 
> router, why doesn't it exist for
> >any IPv6 router connected by a tunnel? In both cases the 
> underlying Interface ID 
> >on the virtual interface of the router is the IPv4 address. 
> I don't see any difference.
> 
> 	I'm not just talking about link-local address format, 
> but also worried
> 	about complication due to sending mechanism.
> 
> 	For other tunnelling mechanisms (like RFC1933 configured tunnel)
> 	we configure IPv4 address of the peer separately to the router.
> 	We always know the peer's IPv4 address if we determine outgoing
> 	interface.
> 
> 	For 6to4, the only place we can get IPv4 address for a 
> particular
> 	6to4 desination is from 6to4 prefix (3rd to 6th bytes).  My
> 	understanding is that we can transmit packets across 
> 6to4 cloud, if
> 	and only if the destination address (or gateway) belongs to 6to4
> 	prefix (2002::/16).
> 
> itojun
> 


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 12:28:15 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA04501
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 12:28:14 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id KAA20614;
	Tue, 2 May 2000 10:26:46 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id JAA10239;
	Tue, 2 May 2000 09:23:17 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e42GLt116299
	for ngtrans-dist; Tue, 2 May 2000 09:21:55 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42GLk716292
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 09:21:46 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id JAA13634
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 09:21:46 -0700 (PDT)
Received: from astro.cs.utk.edu (ASTRO.CS.UTK.EDU [128.169.93.168])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id KAA15244
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 10:21:43 -0600 (MDT)
Received: from astro.cs.utk.edu (LOCALHOST [127.0.0.1])
        by astro.cs.utk.edu (cf 8.9.3) with ESMTP id MAA01525;
        Tue, 2 May 2000 12:17:06 -0400 (EDT)
Message-Id: <200005021617.MAA01525@astro.cs.utk.edu>
X-URI: http://www.cs.utk.edu/~moore/
From: Keith Moore <moore@cs.utk.edu>
To: "Klein, Eric" <Eric_Klein@starhome.com>
cc: Keith Moore <moore@cs.utk.edu>, Greg Maxwell <gmaxwell@martin.fl.us>,
        Brian E Carpenter <brian@hursley.ibm.com>,
        Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com
Subject: Re: (ngtrans) private networks under 6to4-04 
In-reply-to: Your message of "Tue, 02 May 2000 18:35:03 +0300."
             <CE835E918749D21191B10060084C377E01E15C50@ismail1.icomverse.com> 
Date: Tue, 02 May 2000 12:17:06 -0400
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Keith Moore <moore@cs.utk.edu>

> I agree with Keith, 

I'm don't think we agree in general.  I was just pointing out that the 
notion of "just use your public address" doesn't always work - at least,
not without some mechanism by which unconnected folks can get their
own chunk of (nonroutable) address space.

> I am looking for private networks that will not be
> connected to any ISP. Thus when I have my 100 sites I would then need to
> register for my own public prefix as if I were an ISP. Rather than using a
> private range for internal connectivity and registered addresses where
> Interconnectivity would be required. Thus at most I would need 1 NAT
> conversion with only one Registered address persite or per overall network.

private address space causes major problems because eventually you 
either need to connect at least some hosts on the private network 
to the global internet, or you need to connect multiple private networks
together and you get address conflicts as a result.   

(you can use a MAC address to make a v6 address globally unique
and eliminate conflicts, but if you do connect to the outside world
you probably don't want hosts to advertise their MAC addresses to 
the outside world...

having different addresses for private vs. public access, or even
havng multiple addresses for a host where you need to use a different
source address based on which destination you are trying to contact, 
essentially requires applications to be aware of network topology.
it also frustrates development of distributed applications if you can't
use the same address everywhere to reach the same process.

address translation is even worse.  there's no way in general
that an application can work around NAT brain-damage.
 
> Or is it your recommendation that every company go and request their own
> Public prefix for every computer at their sites? If so then you loose any
> gain in addresses, as every organization or company will have to request a
> public prefix based on their anticipated growth at all sites. 
>
> This would reduce the available addresses almost as fast as the current
> IPv4.

this needs a bit more justification to make a credible argument.

Keith


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 14:30:43 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA06779
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 14:30:42 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id MAA06172;
	Tue, 2 May 2000 12:29:25 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id IAA04516;
	Tue, 2 May 2000 08:58:32 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e42FvJm16227
	for ngtrans-dist; Tue, 2 May 2000 08:57:19 -0700 (PDT)
Received: from engmail3.Eng.Sun.COM (engmail3 [129.144.170.5])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42FvC716220
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 08:57:12 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id IAA20126
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 08:57:11 -0700 (PDT)
Received: from da1server.martin.fl.us (da1server.martin.fl.us [198.136.32.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id JAA25625
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 09:57:09 -0600 (MDT)
Received: from localhost (gmaxwell@localhost)
	by da1server.martin.fl.us (8.8.8/8.8.8) with SMTP id LAA20712;
	Tue, 2 May 2000 11:48:02 -0400 (EDT)
Date: Tue, 2 May 2000 11:48:02 -0400 (EDT)
From: Greg Maxwell  <gmaxwell@martin.fl.us>
X-Sender: gmaxwell@da1server
To: "Klein, Eric" <Eric_Klein@starhome.com>
cc: Keith Moore <moore@cs.utk.edu>, Brian E Carpenter <brian@hursley.ibm.com>,
        Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com
Subject: RE: (ngtrans) private networks under 6to4-04 
In-Reply-To: <CE835E918749D21191B10060084C377E01E15C50@ismail1.icomverse.com>
Message-ID: <Pine.GSO.3.96.1000502113944.9125C-100000@da1server>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Greg Maxwell  <gmaxwell@martin.fl.us>

On Tue, 2 May 2000, Klein, Eric wrote:

> Greg,
> 
> I agree with Keith, I am looking for private networks that will not be
> connected to any ISP. 

I'm not sure of what procdure there is for obtaining non-isp-agretted
global prefixes. There should be one.. You'll need to reprefix if you want
to connected the disconnected network to the Internet if you have one, but
not to peer with other networks.

Hmm. Does anyone know if any consideration was made for non-globally
reachable sites to have global prefixes prefixes for the purposes of
peering? (i.e. "I don't want to connect my network to the internet, just a
bunch of other simmlar networks (who may or may not be on the internet)
and I don't care about route table growth among these sites so aggreation
is not important").

> Thus when I have my 100 sites I would then need to
> register for my own public prefix as if I were an ISP. Rather than using a

No. You would obtain a public prefix for every site.

> private range for internal connectivity and registered addresses where
> Interconnectivity would be required. Thus at most I would need 1 NAT
> conversion with only one Registered address persite or per overall network.

Why bother with the NAT, you get one prefix per site then all equipment at
the site gets an address, you advertise your prefix to your peers.

> Or is it your recommendation that every company go and request their own
> Public prefix for every computer at their sites? If so then you loose any

No, not one prefix every computer. I don't know what any of my computers
would do with 1,208,925,819,614,629,174,706,176 (a /48) IP addresses.

> gain in addresses, as every organization or company will have to request a
> public prefix based on their anticipated growth at all sites. 

You get one prefix per site, or perhaps one prefix per company and use the
/48 or /80 per site.

> This would reduce the available addresses almost as fast as the current
> IPv4.

Hardley. There are enough prefixes available to get every person on earth 
a prefix for the long forseeable future.



From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 17:30:38 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA09115
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 17:30:37 -0400 (EDT)
Received: from engmail2.Eng.Sun.COM ([129.146.1.25])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id PAA21571;
	Tue, 2 May 2000 15:29:35 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id OAA02308;
	Tue, 2 May 2000 14:28:16 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e42LRKg17057
	for ngtrans-dist; Tue, 2 May 2000 14:27:20 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42LRD717050
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 14:27:13 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id OAA02130;
	Tue, 2 May 2000 14:27:14 -0700 (PDT)
Received: from mail-gw.hursley.ibm.com (mail-gw.hursley.ibm.com [194.196.110.15])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id OAA18402;
	Tue, 2 May 2000 14:27:11 -0700 (PDT)
Received: from sp3at21.hursley.ibm.com (sp3at21.hursley.ibm.com [9.20.45.21]) by mail-gw.hursley.ibm.com (AIX4.3/UCB 8.8.8/8.8.8) with ESMTP id WAA54722; Tue, 2 May 2000 22:27:09 +0100
Received: from hursley.ibm.com (gsine01.us.sine.ibm.com [9.14.6.41]) by sp3at21.hursley.ibm.com (AIX4.2/UCB 8.7/8.7.3) with ESMTP id WAA24112; Tue, 2 May 2000 22:22:56 +0100 (BST)
Message-ID: <390F470B.4C90A508@hursley.ibm.com>
Date: Tue, 02 May 2000 16:22:19 -0500
From: Brian E Carpenter <brian@hursley.ibm.com>
Organization: IBM
X-Mailer: Mozilla 4.61 [en] (Win98; I)
X-Accept-Language: en,fr
MIME-Version: 1.0
To: Richard Draves <richdr@microsoft.com>
CC: itojun@iijlab.net, Alain Durand <Alain.Durand@eng.sun.com>,
        NGtrans List <ngtrans@sunroof.eng.sun.com>
Subject: Re: (ngtrans) 6to4 issue: link local
References: <4D0A23B3F74DD111ACCD00805F31D81022D4F894@RED-MSG-50>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Brian E Carpenter <brian@hursley.ibm.com>
Content-Transfer-Encoding: 7bit

Let me say it another way to see if I have it straight.

We can formally construct a link-local address (exactly as for
an IPv6-in-IPv4 tunnel) but because there is no 1:1 peer relationship
across a 6to4 exterior routing domain, the other end cannot determine
what that link-local address is.

Would you accept a phrasing in the document like:
   The link-local address of a 6to4 pseudo-interface performing 6to4
   encapsulation would be formed as described in Section 3.7 of [MECH].
   However, no scenario is known in which such an address would be useful.

The reason: this will cover the issue for future readers, and also
cover the case where a use is discovered.

Also see one more comment at the end...

    Brian  

Richard Draves wrote:
> 
> I agree with itojun, link-local addresses make no sense for 6to4. Sure you
> could assign a link-local address to your 6to4 interface. But no one would
> be able to send to it because there's no way to resolve the link-local
> address into a link-layer (v4) address.
> 
> Link-local addresses do make sense for point-to-point configured tunnels,
> because you know the other side's v4 address.
> 
> > -----Original Message-----
> > From: itojun@iijlab.net [mailto:itojun@iijlab.net]
> > Sent: Monday, May 01, 2000 4:54 PM
> > To: Brian E Carpenter
> > Cc: Alain Durand; NGtrans List
> > Subject: Re: (ngtrans) 6to4 issue: link local
> >
> >
> >
> > >I'm still struggling with this. 6to4 does exactly what
> > draft-ietf-ngtrans-mech-06.txt
> > >does, by reference. If this problem exists for a 6to4
> > router, why doesn't it exist for
> > >any IPv6 router connected by a tunnel? In both cases the
> > underlying Interface ID
> > >on the virtual interface of the router is the IPv4 address.
> > I don't see any difference.
> >
> >       I'm not just talking about link-local address format,
> > but also worried
> >       about complication due to sending mechanism.
> >
> >       For other tunnelling mechanisms (like RFC1933 configured tunnel)
> >       we configure IPv4 address of the peer separately to the router.
> >       We always know the peer's IPv4 address if we determine outgoing
> >       interface.
> >
> >       For 6to4, the only place we can get IPv4 address for a
> > particular
> >       6to4 desination is from 6to4 prefix (3rd to 6th bytes).  My
> >       understanding is that we can transmit packets across
> > 6to4 cloud, if
> >       and only if the destination address (or gateway) belongs to 6to4
> >       prefix (2002::/16).

Right, just to be clear, Sowmini noticed that it is not the destination
address that counts, it is the next hop address (i.e. the next router
*or* the destination according to circumstances).


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 17:51:17 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA09401
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 17:51:17 -0400 (EDT)
Received: from engmail3.Eng.Sun.COM ([129.144.170.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id PAA05579;
	Tue, 2 May 2000 15:50:29 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id OAA23772;
	Tue, 2 May 2000 14:49:56 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e42Lmo917196
	for ngtrans-dist; Tue, 2 May 2000 14:48:50 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42Lmg717189
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 14:48:42 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id OAA08030
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 14:48:42 -0700 (PDT)
Received: from mail-gw.hursley.ibm.com (mail-gw.hursley.ibm.com [194.196.110.15])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id OAA00276
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 14:48:41 -0700 (PDT)
Received: from sp3at21.hursley.ibm.com (sp3at21.hursley.ibm.com [9.20.45.21]) by mail-gw.hursley.ibm.com (AIX4.3/UCB 8.8.8/8.8.8) with ESMTP id WAA17832; Tue, 2 May 2000 22:48:36 +0100
Received: from hursley.ibm.com (gsine01.us.sine.ibm.com [9.14.6.41]) by sp3at21.hursley.ibm.com (AIX4.2/UCB 8.7/8.7.3) with ESMTP id WAA16336; Tue, 2 May 2000 22:48:34 +0100 (BST)
Message-ID: <390F4D06.ABFFBEB9@hursley.ibm.com>
Date: Tue, 02 May 2000 16:47:50 -0500
From: Brian E Carpenter <brian@hursley.ibm.com>
Organization: IBM
X-Mailer: Mozilla 4.61 [en] (Win98; I)
X-Accept-Language: en,fr
MIME-Version: 1.0
To: "Klein, Eric" <Eric_Klein@starhome.com>
CC: Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        moore@cs.utk.edu
Subject: Re: (ngtrans) private networks under 6to4-04
References: <CE835E918749D21191B10060084C377E01E15BED@ismail1.icomverse.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Brian E Carpenter <brian@hursley.ibm.com>
Content-Transfer-Encoding: 7bit

"Klein, Eric" wrote:
> 
> Brian,
> 
> You misunderstand. I am talking about designs where there is no direct
> registered IPv4 address. For example I am working on a project where it is
> 100 sites of all private network all using RFC 1918 addresses. These sites
> may be connected to routers that have registered Ipv4 or IPv6 addresses used
> on them, but I need to assure that the private traffic is always isolated
> from the public net.

I didn't misunderstand at all. The motivation for RFC 1918 was to allow
re-use of scarce address space re-use, and IPv6 doesn't need this, so
it is not available.

I'm quite aware of the myth that RFC 1918 addresses are necessary to ensure 
privacy. But it's a myth.

  Brian


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 18:08:00 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA09618
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 18:07:59 -0400 (EDT)
Received: from engmail1.Eng.Sun.COM ([129.146.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id QAA17221;
	Tue, 2 May 2000 16:06:44 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id PAA12883;
	Tue, 2 May 2000 15:06:23 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e42M5OK17244
	for ngtrans-dist; Tue, 2 May 2000 15:05:24 -0700 (PDT)
Received: from engmail4.Eng.Sun.COM (engmail4 [129.144.134.6])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42M5H717237
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 15:05:18 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id PAA08163
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 15:05:17 -0700 (PDT)
Received: from mail4.microsoft.com (mail4.microsoft.com [131.107.3.122])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with SMTP id PAA10499
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 15:05:15 -0700 (PDT)
Received: from 157.54.9.103 by mail4.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 02 May 2000 15:04:27 -0700 (Pacific Daylight Time)
Received: by INET-IMC-04 with Internet Mail Service (5.5.2651.58)
	id <JW3MTGVK>; Tue, 2 May 2000 15:04:25 -0700
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101CA21FF9@RED-MSG-50>
From: Richard Draves <richdr@microsoft.com>
To: "'Brian E Carpenter'" <brian@hursley.ibm.com>
Cc: itojun@iijlab.net, Alain Durand <Alain.Durand@eng.sun.com>,
        NGtrans List <ngtrans@sunroof.eng.sun.com>
Subject: RE: (ngtrans) 6to4 issue: link local
Date: Tue, 2 May 2000 15:04:20 -0700 
X-Mailer: Internet Mail Service (5.5.2651.58)
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Richard Draves <richdr@microsoft.com>

> Let me say it another way to see if I have it straight.
> 
> We can formally construct a link-local address (exactly as for
> an IPv6-in-IPv4 tunnel) but because there is no 1:1 peer relationship
> across a 6to4 exterior routing domain, the other end cannot determine
> what that link-local address is.

Better: "given a link-local (v6) address, the other end can not determine
the appropriate link-layer (v4) address"

This is really just the same situation that exists for automatic tunneling.
If you have an automatic tunneling (pseudo)interface, you don't assign it a
link-local address.

> Would you accept a phrasing in the document like:
>    The link-local address of a 6to4 pseudo-interface performing 6to4
>    encapsulation would be formed as described in Section 3.7 
> of [MECH].
>    However, no scenario is known in which such an address 
> would be useful.
> 
> The reason: this will cover the issue for future readers, and also
> cover the case where a use is discovered.

OK. You could elaborate with another sentence saying that a peer 6to4
gateway will have no way to send to such a link-local address.

Rich


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 19:47:25 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA10478
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 19:47:25 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id RAA17656;
	Tue, 2 May 2000 17:46:04 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id QAA03568;
	Tue, 2 May 2000 16:45:20 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e42NiEj17319
	for ngtrans-dist; Tue, 2 May 2000 16:44:14 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42Ni8717312
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 16:44:08 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id QAA05780
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 16:44:08 -0700 (PDT)
Received: from mail1.microsoft.com (mail1.microsoft.com [131.107.3.125])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with SMTP id QAA02780
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 16:44:07 -0700 (PDT)
Received: from 157.54.9.101 by mail1.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 02 May 2000 16:41:33 -0700 (Pacific Daylight Time)
Received: by INET-IMC-01 with Internet Mail Service (5.5.2651.58)
	id <JZTPTDXK>; Tue, 2 May 2000 16:14:50 -0700
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101CA22001@RED-MSG-50>
From: Richard Draves <richdr@microsoft.com>
To: "'Brian E Carpenter'" <brian@hursley.ibm.com>
Cc: Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        moore@cs.utk.edu, "Klein, Eric" <Eric_Klein@starhome.com>
Subject: RE: (ngtrans) draft-ietf-ngtrans-6to4-04.txt
Date: Tue, 2 May 2000 16:14:44 -0700 
X-Mailer: Internet Mail Service (5.5.2651.58)
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Richard Draves <richdr@microsoft.com>

> IPv6 has no equivalent of RFC 1918 addresses

I think of site-local addresses as being pretty similar to RFC 1918
addresses - would you disagree?

Rich


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 19:51:00 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA10521
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 19:50:59 -0400 (EDT)
Received: from engmail1.Eng.Sun.COM ([129.146.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id RAA19728;
	Tue, 2 May 2000 17:49:48 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id QAA07368;
	Tue, 2 May 2000 16:49:33 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e42Nmtt17355
	for ngtrans-dist; Tue, 2 May 2000 16:48:55 -0700 (PDT)
Received: from engmail3.Eng.Sun.COM (engmail3 [129.144.170.5])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e42Nmk717348
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 16:48:46 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id QAA24812
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 16:48:45 -0700 (PDT)
Received: from ginger.cs.utk.edu (GINGER.CS.UTK.EDU [128.169.95.11])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id RAA19158
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 17:48:43 -0600 (MDT)
Received: from ginger.cs.utk.edu (LOCALHOST [127.0.0.1])
        by ginger.cs.utk.edu (cf 8.9.3) with ESMTP id TAA08864;
        Tue, 2 May 2000 19:47:55 -0400 (EDT)
Message-Id: <200005022347.TAA08864@ginger.cs.utk.edu>
X-URI: http://www.cs.utk.edu/~moore/
From: Keith Moore <moore@cs.utk.edu>
To: Richard Draves <richdr@microsoft.com>
cc: "'Brian E Carpenter'" <brian@hursley.ibm.com>,
        Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        moore@cs.utk.edu, "Klein,
    Eric" <Eric_Klein@starhome.com>
Subject: Re: (ngtrans) draft-ietf-ngtrans-6to4-04.txt 
In-reply-to: Your message of "Tue, 02 May 2000 16:14:44 PDT."
             <4D0A23B3F74DD111ACCD00805F31D8101CA22001@RED-MSG-50> 
Date: Tue, 02 May 2000 19:47:55 -0400
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Keith Moore <moore@cs.utk.edu>

> I think of site-local addresses as being pretty similar to RFC 1918
> addresses - would you disagree?

they look similar to me.  perhaps site-local addresses should be deprecated?


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 20:04:26 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA10656
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 20:04:26 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id SAA27656;
	Tue, 2 May 2000 18:03:43 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id RAA08148;
	Tue, 2 May 2000 17:03:19 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4302K917436
	for ngtrans-dist; Tue, 2 May 2000 17:02:20 -0700 (PDT)
Received: from engmail4.Eng.Sun.COM (engmail4 [129.144.134.6])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4302D717429
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 17:02:13 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id RAA07703
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 17:02:13 -0700 (PDT)
Received: from mail3.microsoft.com (mail3.microsoft.com [131.107.3.123])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with SMTP id RAA11688
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 17:02:12 -0700 (PDT)
Received: from 157.54.9.100 by mail3.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 02 May 2000 17:00:19 -0700 (Pacific Daylight Time)
Received: by INET-IMC-03 with Internet Mail Service (5.5.2651.58)
	id <JW3NVMB6>; Tue, 2 May 2000 17:00:18 -0700
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101CA22003@RED-MSG-50>
From: Richard Draves <richdr@microsoft.com>
To: "'Keith Moore'" <moore@cs.utk.edu>
Cc: "'Brian E Carpenter'" <brian@hursley.ibm.com>,
        Sowmini Varadhan
	 <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        "Klein, Eric"
	 <Eric_Klein@starhome.com>
Subject: RE: (ngtrans) draft-ietf-ngtrans-6to4-04.txt 
Date: Tue, 2 May 2000 17:00:17 -0700 
X-Mailer: Internet Mail Service (5.5.2651.58)
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Richard Draves <richdr@microsoft.com>

> > I think of site-local addresses as being pretty similar to RFC 1918
> > addresses - would you disagree?
> 
> they look similar to me.  perhaps site-local addresses should 
> be deprecated?

Not by me - I think site-local addresses have great potential. First, for
disconnected sites. But more interestingly, to make site renumbering easier.
A connected site can use site-local and global addresses simultaneously,
using site-local addresses for internal communication & configuration. I
think this will make renumbering the site's global prefix much easier,
because there will be many fewer configuration files & databases that know
about global addresses.

Rich


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 20:11:27 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA10737
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 20:11:26 -0400 (EDT)
Received: from engmail3.Eng.Sun.COM ([129.144.170.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id SAA01306;
	Tue, 2 May 2000 18:10:22 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id RAA00226;
	Tue, 2 May 2000 17:09:54 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43095Y17491
	for ngtrans-dist; Tue, 2 May 2000 17:09:05 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4308u717484
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 17:08:56 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id RAA11533
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 17:08:55 -0700 (PDT)
Received: from ginger.cs.utk.edu (GINGER.CS.UTK.EDU [128.169.95.11])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id SAA00414
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 18:08:53 -0600 (MDT)
Received: from ginger.cs.utk.edu (LOCALHOST [127.0.0.1])
        by ginger.cs.utk.edu (cf 8.9.3) with ESMTP id UAA08918;
        Tue, 2 May 2000 20:08:06 -0400 (EDT)
Message-Id: <200005030008.UAA08918@ginger.cs.utk.edu>
X-URI: http://www.cs.utk.edu/~moore/
From: Keith Moore <moore@cs.utk.edu>
To: Richard Draves <richdr@microsoft.com>
cc: "'Keith Moore'" <moore@cs.utk.edu>,
        "'Brian E Carpenter'" <brian@hursley.ibm.com>,
        Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        "Klein, Eric" <Eric_Klein@starhome.com>
Subject: Re: (ngtrans) draft-ietf-ngtrans-6to4-04.txt 
In-reply-to: Your message of "Tue, 02 May 2000 17:00:17 PDT."
             <4D0A23B3F74DD111ACCD00805F31D8101CA22003@RED-MSG-50> 
Date: Tue, 02 May 2000 20:08:06 -0400
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Keith Moore <moore@cs.utk.edu>

site-local addresses might be fine for configuration purposes.
but if you're an application writer you want to be wary about using them.
in general, you want to avoid them unless you can be sure that they will 
not be used in any context other than your own site.
and you don't want a host to ever bind an application socket
to a site-local address by default.

Keith


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 22:52:04 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA13685
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 22:52:03 -0400 (EDT)
Received: from engmail1.Eng.Sun.COM ([129.146.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id UAA02613;
	Tue, 2 May 2000 20:51:00 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id TAA04123;
	Tue, 2 May 2000 19:50:32 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e432ng417613
	for ngtrans-dist; Tue, 2 May 2000 19:49:42 -0700 (PDT)
Received: from engmail3.Eng.Sun.COM (engmail3 [129.144.170.5])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e432nZ717605
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 19:49:35 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id TAA21842
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 19:49:33 -0700 (PDT)
Received: from zmamail05.zma.compaq.com (zmamail05.zma.compaq.com [161.114.64.105])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id UAA01985
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 20:49:33 -0600 (MDT)
Received: by zmamail05.zma.compaq.com (Postfix, from userid 12345)
	id DC7F81A26; Tue,  2 May 2000 22:49:32 -0400 (EDT)
Received: from anw.zk3.dec.com (wasted.zk3.dec.com [16.140.32.3])
	by zmamail05.zma.compaq.com (Postfix) with ESMTP
	id 9B94F1A13; Tue,  2 May 2000 22:49:32 -0400 (EDT)
Received: from localhost by anw.zk3.dec.com (8.9.3/1.1.22.2/08Sep98-0251PM)
	id WAA0000724676; Tue, 2 May 2000 22:49:31 -0400 (EDT)
From: Jim Bound <bound@zk3.dec.com>
Message-Id: <200005030249.WAA0000724676@anw.zk3.dec.com>
To: Keith Moore <moore@cs.utk.edu>
Cc: Richard Draves <richdr@microsoft.com>,
        "'Brian E Carpenter'" <brian@hursley.ibm.com>,
        Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        "Klein, Eric" <Eric_Klein@starhome.com>, bound@zk3.dec.com
Subject: Re: (ngtrans) draft-ietf-ngtrans-6to4-04.txt 
In-reply-to: Your message of "Tue, 02 May 2000 20:08:06 EDT."
             <200005030008.UAA08918@ginger.cs.utk.edu> 
Date: Tue, 02 May 2000 22:49:31 -0400
X-Mts: smtp
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Jim Bound <bound@zk3.dec.com>


>site-local addresses might be fine for configuration purposes.
>but if you're an application writer you want to be wary about using them.
>in general, you want to avoid them unless you can be sure that they will 
>not be used in any context other than your own site.
>and you don't want a host to ever bind an application socket
>to a site-local address by default.

ACK to this.  Also DNS, DHCP, SLP, et al and many distributed banking
and manufacturing networking server applications should never use them.

/jim


From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 23:22:50 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA13810
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 23:22:50 -0400 (EDT)
Received: from engmail2.Eng.Sun.COM ([129.146.1.25])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id VAA14035;
	Tue, 2 May 2000 21:22:01 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id UAA05295;
	Tue, 2 May 2000 20:21:47 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e433Kv917673
	for ngtrans-dist; Tue, 2 May 2000 20:20:57 -0700 (PDT)
Received: from jurassic.eng.sun.com (jurassic [129.146.85.31] (may be forged))
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e433Kp717666
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 20:20:51 -0700 (PDT)
Received: from eng.sun.com (sappey.Eng.Sun.COM [129.146.85.69])
	by jurassic.eng.sun.com (8.10.1+Sun/8.10.1) with ESMTP id e433Kmh926135;
	Tue, 2 May 2000 20:20:48 -0700 (PDT)
Message-ID: <390F9AAD.B770F160@eng.sun.com>
Date: Tue, 02 May 2000 20:19:09 -0700
From: Alain Durand <alain.durand@eng.sun.com>
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
To: Richard Draves <richdr@microsoft.com>, "'Keith Moore'" <moore@cs.utk.edu>,
        "'Brian E Carpenter'" <brian@hursley.ibm.com>,
        ngtrans@sunroof.eng.sun.com
Subject: Re: (ngtrans) draft-ietf-ngtrans-6to4-04.txt
References: <4D0A23B3F74DD111ACCD00805F31D8101CA22003@RED-MSG-50>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Alain Durand <alain.durand@eng.sun.com>
Content-Transfer-Encoding: 7bit

Richard Draves wrote:>

> > they look similar to me.  perhaps site-local addresses should
> > be deprecated?
>
> Not by me - I think site-local addresses have great potential. First, for
> disconnected sites. But more interestingly, to make site renumbering easier.
> A connected site can use site-local and global addresses simultaneously,
> using site-local addresses for internal communication & configuration. I
> think this will make renumbering the site's global prefix much easier,
> because there will be many fewer configuration files & databases that know
> about global addresses.
>
> Rich

I partly disagree. If your topology does not change and you change ISP,
site local addresses may ease renumbering (although deprecating the old prefix
and
advertising the new one may be enough).

But another typical case is when 2 compagnies are merging and you want
to merge their 2 networks, there is a pretty good chance that both networks
were using the same SLA IDs, (say 0x0001, 0x0002,...) so you wil have conflicts

in site local /64 prefixes, and you will be forced to renumber...
If you were not using site local adresses but global adresses on both sites,
in that particular case, you would not have to renumber because the global
/64 prefixes would be differents.

    - Alain.



From owner-ngtrans@sunroof.eng.sun.com  Tue May  2 23:29:44 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA13833
	for <ngtrans-archive@odin.ietf.org>; Tue, 2 May 2000 23:29:43 -0400 (EDT)
Received: from engmail1.Eng.Sun.COM ([129.146.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id VAA16124;
	Tue, 2 May 2000 21:28:58 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id UAA07287;
	Tue, 2 May 2000 20:28:43 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e433S4x17715
	for ngtrans-dist; Tue, 2 May 2000 20:28:04 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e433Rt717708
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 20:27:55 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id UAA07202;
	Tue, 2 May 2000 20:27:49 -0700 (PDT)
Received: from coconut.itojun.org (coconut.itojun.org [210.160.95.97])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id VAA15744;
	Tue, 2 May 2000 21:27:49 -0600 (MDT)
Received: from kiwi.itojun.org (localhost.itojun.org [127.0.0.1])
	by coconut.itojun.org (8.9.3+3.2W/3.7W) with ESMTP id MAA20334;
	Wed, 3 May 2000 12:23:08 +0900 (JST)
To: Richard Draves <richdr@microsoft.com>
cc: "'Brian E Carpenter'" <brian@hursley.ibm.com>,
        Alain Durand <Alain.Durand@eng.sun.com>,
        NGtrans List <ngtrans@sunroof.eng.sun.com>
In-reply-to: richdr's message of Tue, 02 May 2000 15:04:20 MST.
      <4D0A23B3F74DD111ACCD00805F31D8101CA21FF9@RED-MSG-50>
X-Template-Reply-To: itojun@itojun.org
X-Template-Return-Receipt-To: itojun@itojun.org
X-PGP-Fingerprint: F8 24 B4 2C 8C 98 57 FD  90 5F B4 60 79 54 16 E2
Subject: Re: (ngtrans) 6to4 issue: link local
From: itojun@iijlab.net
Date: Wed, 03 May 2000 12:23:07 +0900
Message-ID: <20332.957324187@coconut.itojun.org>
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: itojun@iijlab.net


>> Would you accept a phrasing in the document like:
>>    The link-local address of a 6to4 pseudo-interface performing 6to4
>>    encapsulation would be formed as described in Section 3.7 
>> of [MECH].
>>    However, no scenario is known in which such an address 
>> would be useful.
>> 
>> The reason: this will cover the issue for future readers, and also
>> cover the case where a use is discovered.
>OK. You could elaborate with another sentence saying that a peer 6to4
>gateway will have no way to send to such a link-local address.

	Is it clear enough for readers that "since there's no use in it,
	it is not really necessary to attach link-local address to 6to4 pseudo
	interface"?
	To me RFC2373 section 2.1 looks a bit too strong that it says
	link-local address is mandatory for all interfaces.  as brian pointed
	out, we have no use in link-local for auto tunnel interface too.

itojun


From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 02:49:15 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA26893
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 02:49:14 -0400 (EDT)
Received: from engmail2.Eng.Sun.COM ([129.146.1.25])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id AAA19143;
	Wed, 3 May 2000 00:47:53 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id XAA20700;
	Tue, 2 May 2000 23:47:30 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e436kc017827
	for ngtrans-dist; Tue, 2 May 2000 23:46:38 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e436kV717820
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 23:46:32 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id XAA21576
	for <ngtrans@sunroof.eng.sun.com>; Tue, 2 May 2000 23:46:31 -0700 (PDT)
Received: from mailgw1.netvision.net.il (mailgw1.netvision.net.il [194.90.1.14])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id AAA18676
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 00:46:27 -0600 (MDT)
Received: from sendout.icomverse.com (Efrat-FR3.ser.netvision.net.il [199.203.174.65])
	by mailgw1.netvision.net.il (8.9.3/8.9.3) with ESMTP id JAA08987
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 09:46:20 +0300 (IDT)
Received: from ismail1.icomverse.com (ismail1.icomverse.com [190.190.110.2])
	by sendout.icomverse.com (8.9.3/8.8.7) with ESMTP id IAA26704
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 08:46:49 +0300
Received: by ismail1.icomverse.com with Internet Mail Service (5.5.2650.21)
	id <JHFAWBTA>; Wed, 3 May 2000 09:45:45 +0300
Message-ID: <CE835E918749D21191B10060084C377E01E15C64@ismail1.icomverse.com>
From: "Klein, Eric" <Eric_Klein@starhome.com>
To: "Klein, Eric" <Eric_Klein@starhome.com>, ngtrans@sunroof.eng.sun.com
Subject: RE: (ngtrans) private networks under 6to4-04 
Date: Wed, 3 May 2000 09:45:43 +0300 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: "Klein, Eric" <Eric_Klein@starhome.com>

Greg, 

Your answer scares the hell out of me. I am looking at a network with will
grow to over 1000 sites in under 5 years,  all of them riding on a private
network (not the internet), based on your answer:
	>> Thus when I have my 100 sites I would then need to
	>> register for my own public prefix as if I were an ISP. Rather
than using a

	>No. You would obtain a public prefix for every site.

means that I not only will have to get an address for my network (possibly
bought from an ISP without connecting to the Internet) but then do the same
for all 1000 sites. This seems to be much worse a problem than the running
out of addresses that the current IPv4 is facing. Just think that any
international company will have to register for a Public prefix for each
site. Minolta has over 50 sites in the US alone, and they are not the
largest company out there.

This needs to be considered, because many companies have one registered
address at their main site, and then act as an "ISP" to their various sites
via their Intranets. This would result in some real complicated routing
tables and purchases.

As to the availability of an address per person for the foreseeable future,
this would not be the case. It would be like phone numbers per person: one
for the home phone (computer via ISP) and one for the work phone (computer
from their company) and one for their cell phone (from their phone company
as WAP phones each get an address) this would mean that it is 3 addresses
per person. What happens once we are connecting their car, or microwave (and
other home appliances as Microsoft wants)? 

Eric

-----Original Message-----
From: Greg Maxwell [mailto:gmaxwell@martin.fl.us]
Sent: Tuesday, May 02, 2000 6:48 PM
To: Klein, Eric
Cc: Keith Moore; Brian E Carpenter; Sowmini Varadhan;
ngtrans@sunroof.eng.sun.com
Subject: RE: (ngtrans) private networks under 6to4-04 


On Tue, 2 May 2000, Klein, Eric wrote:

> Greg,
> 
> I agree with Keith, I am looking for private networks that will not be
> connected to any ISP. 

I'm not sure of what procdure there is for obtaining non-isp-agretted
global prefixes. There should be one.. You'll need to reprefix if you want
to connected the disconnected network to the Internet if you have one, but
not to peer with other networks.

Hmm. Does anyone know if any consideration was made for non-globally
reachable sites to have global prefixes prefixes for the purposes of
peering? (i.e. "I don't want to connect my network to the internet, just a
bunch of other simmlar networks (who may or may not be on the internet)
and I don't care about route table growth among these sites so aggreation
is not important").

> Thus when I have my 100 sites I would then need to
> register for my own public prefix as if I were an ISP. Rather than using a

No. You would obtain a public prefix for every site.

> private range for internal connectivity and registered addresses where
> Interconnectivity would be required. Thus at most I would need 1 NAT
> conversion with only one Registered address persite or per overall
network.

Why bother with the NAT, you get one prefix per site then all equipment at
the site gets an address, you advertise your prefix to your peers.

> Or is it your recommendation that every company go and request their own
> Public prefix for every computer at their sites? If so then you loose any

No, not one prefix every computer. I don't know what any of my computers
would do with 1,208,925,819,614,629,174,706,176 (a /48) IP addresses.

> gain in addresses, as every organization or company will have to request a
> public prefix based on their anticipated growth at all sites. 

You get one prefix per site, or perhaps one prefix per company and use the
/48 or /80 per site.

> This would reduce the available addresses almost as fast as the current
> IPv4.

Hardley. There are enough prefixes available to get every person on earth 
a prefix for the long forseeable future.


From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 08:19:23 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA00649
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 08:19:22 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id GAA12129;
	Wed, 3 May 2000 06:18:37 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id FAA11039;
	Wed, 3 May 2000 05:17:46 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43CGmV17995
	for ngtrans-dist; Wed, 3 May 2000 05:16:49 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43CGg717988
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 05:16:42 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id FAA14241
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 05:16:40 -0700 (PDT)
Received: from da1server.martin.fl.us (da1server.martin.fl.us [198.136.32.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id GAA10921
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 06:16:39 -0600 (MDT)
Received: from localhost (gmaxwell@localhost)
	by da1server.martin.fl.us (8.8.8/8.8.8) with SMTP id IAA21131;
	Wed, 3 May 2000 08:08:01 -0400 (EDT)
Date: Wed, 3 May 2000 08:08:01 -0400 (EDT)
From: Greg Maxwell  <gmaxwell@martin.fl.us>
X-Sender: gmaxwell@da1server
To: Richard Draves <richdr@microsoft.com>
cc: "'Brian E Carpenter'" <brian@hursley.ibm.com>,
        Sowmini Varadhan <varadhan@zk3.dec.com>, ngtrans@sunroof.eng.sun.com,
        moore@cs.utk.edu, "Klein, Eric" <Eric_Klein@starhome.com>
Subject: RE: (ngtrans) draft-ietf-ngtrans-6to4-04.txt
In-Reply-To: <4D0A23B3F74DD111ACCD00805F31D8101CA22001@RED-MSG-50>
Message-ID: <Pine.GSO.3.96.1000503080721.21056A-100000@da1server>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Greg Maxwell  <gmaxwell@martin.fl.us>

On Tue, 2 May 2000, Richard Draves wrote:

> > IPv6 has no equivalent of RFC 1918 addresses
> 
> I think of site-local addresses as being pretty similar to RFC 1918
> addresses - would you disagree?

They differ in that they are scoped. Have you seen how people use RFC 1918
addresses?



From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 08:50:08 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA01198
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 08:50:08 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id GAA26446;
	Wed, 3 May 2000 06:49:20 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id FAA12763;
	Wed, 3 May 2000 05:48:52 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43Clik18048
	for ngtrans-dist; Wed, 3 May 2000 05:47:44 -0700 (PDT)
Received: from engmail3.Eng.Sun.COM (engmail3 [129.144.170.5])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43Clb718041
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 05:47:37 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id FAA03516
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 05:47:38 -0700 (PDT)
Received: from da1server.martin.fl.us (da1server.martin.fl.us [198.136.32.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id GAA25224
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 06:47:37 -0600 (MDT)
Received: from localhost (gmaxwell@localhost)
	by da1server.martin.fl.us (8.8.8/8.8.8) with SMTP id IAA22583;
	Wed, 3 May 2000 08:45:40 -0400 (EDT)
Date: Wed, 3 May 2000 08:45:40 -0400 (EDT)
From: Greg Maxwell <gmaxwell@martin.fl.us>
X-Sender: gmaxwell@da1server
To: "Klein, Eric" <Eric_Klein@starhome.com>
cc: ngtrans@sunroof.eng.sun.com
Subject: RE: (ngtrans) private networks under 6to4-04 
In-Reply-To: <CE835E918749D21191B10060084C377E01E15C64@ismail1.icomverse.com>
Message-ID: <Pine.GSO.3.96.1000503083337.21056B-100000@da1server>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Greg Maxwell <gmaxwell@martin.fl.us>

On Wed, 3 May 2000, Klein, Eric wrote:

> means that I not only will have to get an address for my network (possibly
> bought from an ISP without connecting to the Internet) but then do the same
> for all 1000 sites. This seems to be much worse a problem than the running
> out of addresses that the current IPv4 is facing. Just think that any
> international company will have to register for a Public prefix for each
> site. Minolta has over 50 sites in the US alone, and they are not the
> largest company out there.

So.. You are telling me that you have 1000 sites which are disconnected
from the Internet? That Minolta has no internet access in their over 50
sites?

Excuse me, but I find that very hard to believe.
 
> This needs to be considered, because many companies have one registered
> address at their main site, and then act as an "ISP" to their various sites
> via their Intranets. This would result in some real complicated routing
> tables and purchases.

Their main site recieves a /48 from their provider. They assign out /64s
to their subnets (sites) (using 16bits for internal structure). Tada.

IPv6 stops the distruction of globally unique addressing. You do not have
private addressing, and you do not want private addressing, unless you
happen do be a networking support vendor who makes big bucks supporting
unnessassarly complex and fragile networks.

> As to the availability of an address per person for the foreseeable future,
> this would not be the case. It would be like phone numbers per person: one
> for the home phone (computer via ISP) and one for the work phone (computer
> from their company) and one for their cell phone (from their phone company
> as WAP phones each get an address) this would mean that it is 3 addresses
> per person. What happens once we are connecting their car, or microwave (and
> other home appliances as Microsoft wants)? 

No I said: "There are enough prefixes available to get every person on
earth a prefix for the long forseeable future."

Thus I ment giving each person a /64. Unless single people typically have
more then 18,446,744,073,709,551,616 net phones, tosters,
wap phones, etc you will not have a problem.




From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 09:22:41 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01692
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 09:22:40 -0400 (EDT)
Received: from engmail1.Eng.Sun.COM ([129.146.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id HAA13153;
	Wed, 3 May 2000 07:21:38 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id GAA19654;
	Wed, 3 May 2000 06:21:01 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43DJr318438
	for ngtrans-dist; Wed, 3 May 2000 06:19:53 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43DJe718421
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 06:19:41 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id GAA16251
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 06:19:40 -0700 (PDT)
Received: from ztxmail03.ztx.compaq.com (ztxmail03.ztx.compaq.com [161.114.1.207])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id HAA11919
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 07:19:39 -0600 (MDT)
Received: by ztxmail03.ztx.compaq.com (Postfix, from userid 12345)
	id 73BE1270; Wed,  3 May 2000 08:19:38 -0500 (CDT)
Received: from anw.zk3.dec.com (wasted.zk3.dec.com [16.140.32.3])
	by ztxmail03.ztx.compaq.com (Postfix) with ESMTP
	id D09D83D3; Wed,  3 May 2000 08:19:37 -0500 (CDT)
Received: by anw.zk3.dec.com (8.9.3/1.1.22.2/08Sep98-0251PM)
	id JAA0000803845; Wed, 3 May 2000 09:19:37 -0400 (EDT)
Date: Wed, 3 May 2000 09:19:37 -0400 (EDT)
From: Sowmini Varadhan <varadhan@zk3.dec.com>
Message-Id: <200005031319.JAA0000803845@anw.zk3.dec.com>
To: brian@hursley.ibm.com, richdr@microsoft.com
Subject: RE: (ngtrans) 6to4 issue: link local
Cc: Alain.Durand@eng.sun.com, gilligan@freegate.com,
        ngtrans@sunroof.eng.sun.com, nordmark@eng.sun.com
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Sowmini Varadhan <varadhan@zk3.dec.com>

> From: Richard Draves <richdr@microsoft.com>
> 
> > Let me say it another way to see if I have it straight.
> > 
> > We can formally construct a link-local address (exactly as for
> > an IPv6-in-IPv4 tunnel) but because there is no 1:1 peer relationship
> > across a 6to4 exterior routing domain, the other end cannot determine
> > what that link-local address is.
> 
> Better: "given a link-local (v6) address, the other end can not determine
> the appropriate link-layer (v4) address"
> 
> This is really just the same situation that exists for automatic tunneling.
> If you have an automatic tunneling (pseudo)interface, you don't assign it a
> link-local address.

Note, however, draft-ietf-ngtrans-mech-06.txt requires you to do just that:

"3.7.  Link-Local Addresses

   Both the configured and automatic tunnels are IPv6 interfaces (over
   the IPv4 "link layer") thus MUST have link-local addresses.  The
   link-local addresses are used by routing protocols operating over the
   tunnels."


--Sowmini


From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 10:15:06 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA02870
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 10:15:05 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id IAA11023;
	Wed, 3 May 2000 08:13:26 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id HAA26522;
	Wed, 3 May 2000 07:13:00 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43EBiE18784
	for ngtrans-dist; Wed, 3 May 2000 07:11:44 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43EBb718777
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 07:11:37 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id HAA24609;
	Wed, 3 May 2000 07:11:37 -0700 (PDT)
Received: from mail-gw.hursley.ibm.com (mail-gw.hursley.ibm.com [194.196.110.15])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id HAA19908;
	Wed, 3 May 2000 07:11:30 -0700 (PDT)
Received: from sp3at21.hursley.ibm.com (sp3at21.hursley.ibm.com [9.20.45.21]) by mail-gw.hursley.ibm.com (AIX4.3/UCB 8.8.8/8.8.8) with ESMTP id PAA47798; Wed, 3 May 2000 15:11:25 +0100
Received: from hursley.ibm.com (gsine02.us.sine.ibm.com [9.14.6.42]) by sp3at21.hursley.ibm.com (AIX4.2/UCB 8.7/8.7.3) with ESMTP id PAA17054; Wed, 3 May 2000 15:11:18 +0100 (BST)
Message-ID: <39103388.4E188322@hursley.ibm.com>
Date: Wed, 03 May 2000 09:11:20 -0500
From: Brian E Carpenter <brian@hursley.ibm.com>
Organization: IBM
X-Mailer: Mozilla 4.61 [en] (Win98; I)
X-Accept-Language: en,fr
MIME-Version: 1.0
To: itojun@iijlab.net
CC: Richard Draves <richdr@microsoft.com>,
        Alain Durand <Alain.Durand@eng.sun.com>,
        NGtrans List <ngtrans@sunroof.eng.sun.com>
Subject: Re: (ngtrans) 6to4 issue: link local
References: <20332.957324187@coconut.itojun.org>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Brian E Carpenter <brian@hursley.ibm.com>
Content-Transfer-Encoding: 7bit

Actually it was Rich who pointed it out for auto tunnels. What was
confusing me is that draft-ietf-ngtrans-mech-06.txt defines
link-local for both configured and auto tunnels. It makes a lot
more sense to me that it is meaningless for auto tunnels as well
as for 6to4.

I will adjust the wording in the 6to4 draft. Maybe 
draft-ietf-ngtrans-mech-06.txt should be tweaked as well?

   Brian

itojun@iijlab.net wrote:
> 
> >> Would you accept a phrasing in the document like:
> >>    The link-local address of a 6to4 pseudo-interface performing 6to4
> >>    encapsulation would be formed as described in Section 3.7
> >> of [MECH].
> >>    However, no scenario is known in which such an address
> >> would be useful.
> >>
> >> The reason: this will cover the issue for future readers, and also
> >> cover the case where a use is discovered.
> >OK. You could elaborate with another sentence saying that a peer 6to4
> >gateway will have no way to send to such a link-local address.
> 
>         Is it clear enough for readers that "since there's no use in it,
>         it is not really necessary to attach link-local address to 6to4 pseudo
>         interface"?
>         To me RFC2373 section 2.1 looks a bit too strong that it says
>         link-local address is mandatory for all interfaces.  as brian pointed
>         out, we have no use in link-local for auto tunnel interface too.
> 
> itojun

-- 
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Brian E Carpenter 
Program Director, Internet Standards & Technology, IBM 
On assignment for IBM at http://www.iCAIR.org 
Attend INET 2000: http://www.isoc.org/inet2000
Non-IBM email: brian@icair.org


From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 10:22:55 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA02991
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 10:22:54 -0400 (EDT)
Received: from engmail2.Eng.Sun.COM ([129.146.1.25])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id IAA14914;
	Wed, 3 May 2000 08:19:47 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id HAA22371;
	Wed, 3 May 2000 07:19:29 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43EIru18814
	for ngtrans-dist; Wed, 3 May 2000 07:18:53 -0700 (PDT)
Received: from engmail3.Eng.Sun.COM (engmail3 [129.144.170.5])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43EIl718807
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 07:18:47 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id HAA13498
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 07:18:47 -0700 (PDT)
Received: from mail-gw.hursley.ibm.com (mail-gw.hursley.ibm.com [194.196.110.15])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id IAA14276
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 08:18:42 -0600 (MDT)
Received: from sp3at21.hursley.ibm.com (sp3at21.hursley.ibm.com [9.20.45.21]) by mail-gw.hursley.ibm.com (AIX4.3/UCB 8.8.8/8.8.8) with ESMTP id PAA21498; Wed, 3 May 2000 15:18:40 +0100
Received: from hursley.ibm.com (gsine02.us.sine.ibm.com [9.14.6.42]) by sp3at21.hursley.ibm.com (AIX4.2/UCB 8.7/8.7.3) with ESMTP id PAA18842; Wed, 3 May 2000 15:18:33 +0100 (BST)
Message-ID: <3910353B.47811C6D@hursley.ibm.com>
Date: Wed, 03 May 2000 09:18:36 -0500
From: Brian E Carpenter <brian@hursley.ibm.com>
Organization: IBM
X-Mailer: Mozilla 4.61 [en] (Win98; I)
X-Accept-Language: en,fr
MIME-Version: 1.0
To: "Klein, Eric" <Eric_Klein@starhome.com>
CC: ngtrans@sunroof.eng.sun.com
Subject: Re: (ngtrans) private networks under 6to4-04
References: <CE835E918749D21191B10060084C377E01E15C64@ismail1.icomverse.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Brian E Carpenter <brian@hursley.ibm.com>
Content-Transfer-Encoding: 7bit

Eric,

IPv6 has enough address space for the scenarios you describe. Run the numbers
and you will see; this was one of the design criteria.

As for your problem, it's a valid point that should be brought up with
the IP address registries at some point. There is no shortage of address
space; it's just that the current assignment policies assume that 
addresses are always delegated via ISPs. We don't have a policy for
address assignment to disconnected networks, and maybe we should.

   Brian

"Klein, Eric" wrote:
> 
> Greg,
> 
> Your answer scares the hell out of me. I am looking at a network with will
> grow to over 1000 sites in under 5 years,  all of them riding on a private
> network (not the internet), based on your answer:
>         >> Thus when I have my 100 sites I would then need to
>         >> register for my own public prefix as if I were an ISP. Rather
> than using a
> 
>         >No. You would obtain a public prefix for every site.
> 
> means that I not only will have to get an address for my network (possibly
> bought from an ISP without connecting to the Internet) but then do the same
> for all 1000 sites. This seems to be much worse a problem than the running
> out of addresses that the current IPv4 is facing. Just think that any
> international company will have to register for a Public prefix for each
> site. Minolta has over 50 sites in the US alone, and they are not the
> largest company out there.
> 
> This needs to be considered, because many companies have one registered
> address at their main site, and then act as an "ISP" to their various sites
> via their Intranets. This would result in some real complicated routing
> tables and purchases.
> 
> As to the availability of an address per person for the foreseeable future,
> this would not be the case. It would be like phone numbers per person: one
> for the home phone (computer via ISP) and one for the work phone (computer
> from their company) and one for their cell phone (from their phone company
> as WAP phones each get an address) this would mean that it is 3 addresses
> per person. What happens once we are connecting their car, or microwave (and
> other home appliances as Microsoft wants)?
> 
> Eric
> 
> -----Original Message-----
> From: Greg Maxwell [mailto:gmaxwell@martin.fl.us]
> Sent: Tuesday, May 02, 2000 6:48 PM
> To: Klein, Eric
> Cc: Keith Moore; Brian E Carpenter; Sowmini Varadhan;
> ngtrans@sunroof.eng.sun.com
> Subject: RE: (ngtrans) private networks under 6to4-04
> 
> On Tue, 2 May 2000, Klein, Eric wrote:
> 
> > Greg,
> >
> > I agree with Keith, I am looking for private networks that will not be
> > connected to any ISP.
> 
> I'm not sure of what procdure there is for obtaining non-isp-agretted
> global prefixes. There should be one.. You'll need to reprefix if you want
> to connected the disconnected network to the Internet if you have one, but
> not to peer with other networks.
> 
> Hmm. Does anyone know if any consideration was made for non-globally
> reachable sites to have global prefixes prefixes for the purposes of
> peering? (i.e. "I don't want to connect my network to the internet, just a
> bunch of other simmlar networks (who may or may not be on the internet)
> and I don't care about route table growth among these sites so aggreation
> is not important").
> 
> > Thus when I have my 100 sites I would then need to
> > register for my own public prefix as if I were an ISP. Rather than using a
> 
> No. You would obtain a public prefix for every site.
> 
> > private range for internal connectivity and registered addresses where
> > Interconnectivity would be required. Thus at most I would need 1 NAT
> > conversion with only one Registered address persite or per overall
> network.
> 
> Why bother with the NAT, you get one prefix per site then all equipment at
> the site gets an address, you advertise your prefix to your peers.
> 
> > Or is it your recommendation that every company go and request their own
> > Public prefix for every computer at their sites? If so then you loose any
> 
> No, not one prefix every computer. I don't know what any of my computers
> would do with 1,208,925,819,614,629,174,706,176 (a /48) IP addresses.
> 
> > gain in addresses, as every organization or company will have to request a
> > public prefix based on their anticipated growth at all sites.
> 
> You get one prefix per site, or perhaps one prefix per company and use the
> /48 or /80 per site.
> 
> > This would reduce the available addresses almost as fast as the current
> > IPv4.
> 
> Hardley. There are enough prefixes available to get every person on earth
> a prefix for the long forseeable future.

-- 
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Brian E Carpenter 
Program Director, Internet Standards & Technology, IBM 
On assignment for IBM at http://www.iCAIR.org 
Attend INET 2000: http://www.isoc.org/inet2000
Non-IBM email: brian@icair.org


From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 10:31:46 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA03219
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 10:31:46 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id IAA21772;
	Wed, 3 May 2000 08:30:39 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id HAA28883;
	Wed, 3 May 2000 07:29:58 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43ESCX18845
	for ngtrans-dist; Wed, 3 May 2000 07:28:13 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43ES6718838
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 07:28:06 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id HAA26407;
	Wed, 3 May 2000 07:28:06 -0700 (PDT)
Received: from mail-gw.hursley.ibm.com (mail-gw.hursley.ibm.com [194.196.110.15])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id IAA20175;
	Wed, 3 May 2000 08:28:04 -0600 (MDT)
Received: from sp3at21.hursley.ibm.com (sp3at21.hursley.ibm.com [9.20.45.21]) by mail-gw.hursley.ibm.com (AIX4.3/UCB 8.8.8/8.8.8) with ESMTP id PAA29812; Wed, 3 May 2000 15:28:01 +0100
Received: from hursley.ibm.com (gsine02.us.sine.ibm.com [9.14.6.42]) by sp3at21.hursley.ibm.com (AIX4.2/UCB 8.7/8.7.3) with ESMTP id PAA11338; Wed, 3 May 2000 15:27:57 +0100 (BST)
Message-ID: <3910376F.FB91E53F@hursley.ibm.com>
Date: Wed, 03 May 2000 09:27:59 -0500
From: Brian E Carpenter <brian@hursley.ibm.com>
Organization: IBM
X-Mailer: Mozilla 4.61 [en] (Win98; I)
X-Accept-Language: en,fr
MIME-Version: 1.0
To: Alain Durand <alain.durand@eng.sun.com>
CC: Richard Draves <richdr@microsoft.com>, "'Keith Moore'" <moore@cs.utk.edu>,
        ngtrans@sunroof.eng.sun.com
Subject: disconnected networks [was Re: (ngtrans) draft-ietf-ngtrans-6to4-04.txt
References: <4D0A23B3F74DD111ACCD00805F31D8101CA22003@RED-MSG-50> <390F9AAD.B770F160@eng.sun.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Brian E Carpenter <brian@hursley.ibm.com>
Content-Transfer-Encoding: 7bit

Alain Durand wrote:
> 
> Richard Draves wrote:>
> 
> > > they look similar to me.  perhaps site-local addresses should
> > > be deprecated?
> >
> > Not by me - I think site-local addresses have great potential. First, for
> > disconnected sites. But more interestingly, to make site renumbering easier.
> > A connected site can use site-local and global addresses simultaneously,
> > using site-local addresses for internal communication & configuration. I
> > think this will make renumbering the site's global prefix much easier,
> > because there will be many fewer configuration files & databases that know
> > about global addresses.
> >
> > Rich
> 
> I partly disagree. If your topology does not change and you change ISP,
> site local addresses may ease renumbering (although deprecating the old prefix
> and
> advertising the new one may be enough).
> 
> But another typical case is when 2 compagnies are merging and you want
> to merge their 2 networks, there is a pretty good chance that both networks
> were using the same SLA IDs, (say 0x0001, 0x0002,...) so you wil have conflicts
> 
> in site local /64 prefixes, and you will be forced to renumber...
> If you were not using site local adresses but global adresses on both sites,
> in that particular case, you would not have to renumber because the global
> /64 prefixes would be differents.

Right. I see no reason to deprecate site-local, but we should start to consider
how to give globally unique prefixes to disconnected networks. All our experience
tells us that such networks *will* ultimately connect, and we still haven't
made renumbering painless.

Yes, I *know* what somebody is going to say about aggregation but if only very,
very large networks get their own prefixes this shouldn't be too big a concern.
And those are exactly the ones that renumbering hurts most, and are the most
likely to be seriously multi-homed.

   Brian


From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 11:02:08 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA03804
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 11:02:08 -0400 (EDT)
Received: from engmail2.Eng.Sun.COM ([129.146.1.25])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id JAA12885;
	Wed, 3 May 2000 09:00:38 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id IAA03061;
	Wed, 3 May 2000 08:00:10 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43Ex8e18889
	for ngtrans-dist; Wed, 3 May 2000 07:59:08 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43Ex1718882
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 07:59:01 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id HAA02891;
	Wed, 3 May 2000 07:59:00 -0700 (PDT)
Received: from da1server.martin.fl.us (da1server.martin.fl.us [198.136.32.5])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id HAA18383;
	Wed, 3 May 2000 07:58:58 -0700 (PDT)
Received: from localhost (gmaxwell@localhost)
	by da1server.martin.fl.us (8.8.8/8.8.8) with SMTP id KAA28763;
	Wed, 3 May 2000 10:57:00 -0400 (EDT)
Date: Wed, 3 May 2000 10:56:59 -0400 (EDT)
From: Greg Maxwell  <gmaxwell@martin.fl.us>
X-Sender: gmaxwell@da1server
To: Brian E Carpenter <brian@hursley.ibm.com>
cc: Alain Durand <alain.durand@eng.sun.com>,
        Richard Draves <richdr@microsoft.com>,
        "'Keith Moore'" <moore@cs.utk.edu>, ngtrans@sunroof.eng.sun.com
Subject: Re: disconnected networks [was Re: (ngtrans) draft-ietf-ngtrans-6to4-04.txt
In-Reply-To: <3910376F.FB91E53F@hursley.ibm.com>
Message-ID: <Pine.GSO.3.96.1000503105243.21056C-100000@da1server>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Greg Maxwell  <gmaxwell@martin.fl.us>

On Wed, 3 May 2000, Brian E Carpenter wrote:

> Right. I see no reason to deprecate site-local, but we should start to consider
> how to give globally unique prefixes to disconnected networks. All our experience
> tells us that such networks *will* ultimately connect, and we still haven't
> made renumbering painless.

It should be emphisised that they need not connect to the Internet for any
form of private addressing to be a BIG problem. Just consider the simple
case of plugging into someone else.





From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 13:15:59 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06747
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 13:15:58 -0400 (EDT)
Received: from engmail1.Eng.Sun.COM ([129.146.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id LAA28326;
	Wed, 3 May 2000 11:14:49 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id KAA00715;
	Wed, 3 May 2000 10:12:51 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43H5T118959
	for ngtrans-dist; Wed, 3 May 2000 10:05:29 -0700 (PDT)
Received: from engmail3.Eng.Sun.COM (engmail3 [129.144.170.5])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43H5M718952
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 10:05:22 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id KAA17219;
	Wed, 3 May 2000 10:05:20 -0700 (PDT)
Received: from mailhost.iprg.nokia.com (mailhost.iprg.nokia.com [205.226.5.12])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id KAA15169;
	Wed, 3 May 2000 10:05:19 -0700 (PDT)
Received: from darkstar.iprg.nokia.com (darkstar.iprg.nokia.com [205.226.5.69])
	by mailhost.iprg.nokia.com (8.9.3/8.9.3-GLGS) with ESMTP id KAA05801;
	Wed, 3 May 2000 10:05:12 -0700 (PDT)
Received: (from root@localhost)
	by darkstar.iprg.nokia.com (8.9.3/8.9.3-VIRSCAN) id KAA21925;
	Wed, 3 May 2000 10:01:57 -0700
X-Virus-Scanned:  Wed, 3 May 2000 10:01:57 -0700 Nokia Silicon Valley Email Exploit Scanner
Received: from <hinden@iprg.nokia.com> (spruce.iprg.nokia.com [205.226.1.63]) by darkstar.iprg.nokia.com  SMTP/WTS (12.69)
 xma017690; Wed, 3 May 00 10:00:16 -0700
Message-Id: <4.3.1.2.20000503095434.02129820@mailhost.iprg.nokia.com>
X-Sender: hinden@mailhost.iprg.nokia.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.1
Date: Wed, 03 May 2000 10:02:10 -0700
To: Brian E Carpenter <brian@hursley.ibm.com>
From: Bob Hinden <hinden@iprg.nokia.com>
Subject: Re: disconnected networks [was Re: (ngtrans)
  draft-ietf-ngtrans-6to4-04.txt
Cc: Alain Durand <alain.durand@eng.sun.com>,
        Richard Draves <richdr@microsoft.com>,
        "'Keith Moore'" <moore@cs.utk.edu>, ngtrans@sunroof.eng.sun.com
In-Reply-To: <3910376F.FB91E53F@hursley.ibm.com>
References: <4D0A23B3F74DD111ACCD00805F31D8101CA22003@RED-MSG-50>
 <390F9AAD.B770F160@eng.sun.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Bob Hinden <hinden@iprg.nokia.com>

Brian,


>Right. I see no reason to deprecate site-local, but we should start to 
>consider
>how to give globally unique prefixes to disconnected networks. All our 
>experience
>tells us that such networks *will* ultimately connect, and we still haven't
>made renumbering painless.
>
>Yes, I *know* what somebody is going to say about aggregation but if only 
>very,
>very large networks get their own prefixes this shouldn't be too big a 
>concern.
>And those are exactly the ones that renumbering hurts most, and are the most
>likely to be seriously multi-homed.

I tend to agree.  Were you thinking about a mechanisms where any site could 
get a global prefix (perhaps use a TLA like 6to4) but that these prefix 
would never be routed on the Internet?  Similar rules to the site-local 
prefix, but everyone can get their own.  When the site does connect they 
get a new prefix and we make sure that the address selection rules pick a 
routable prefix when both types exist.

Or were you thinking about routable prefixes independent of an ISP allocation?

Bob
  



From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 13:35:01 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA07170
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 13:35:00 -0400 (EDT)
Received: from engmail3.Eng.Sun.COM ([129.144.170.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id LAA13576;
	Wed, 3 May 2000 11:33:44 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id KAA24439;
	Wed, 3 May 2000 10:33:06 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43HR5b19030
	for ngtrans-dist; Wed, 3 May 2000 10:27:05 -0700 (PDT)
Received: from engmail3.Eng.Sun.COM (engmail3 [129.144.170.5])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43HQw719023
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 10:26:59 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id KAA22559
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 10:26:57 -0700 (PDT)
Received: from fwns1.raleigh.ibm.com (fwns1d.raleigh.ibm.com [204.146.167.235])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id KAA29303
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 10:26:56 -0700 (PDT)
Received: from rtpmail01.raleigh.ibm.com (rtpmail01.raleigh.ibm.com [9.37.172.24])
	by fwns1.raleigh.ibm.com (8.9.0/8.9.0/RTP-FW-1.2) with ESMTP id NAA24910;
	Wed, 3 May 2000 13:26:23 -0400
Received: from ludwigia.raleigh.ibm.com (ludwigia.raleigh.ibm.com [9.37.60.3])
	by rtpmail01.raleigh.ibm.com (8.8.5/8.8.5/RTP-ral-1.1) with ESMTP id NAA25196;
	Wed, 3 May 2000 13:26:22 -0400
Received: from ludwigia.raleigh.ibm.com (localhost [127.0.0.1]) by ludwigia.raleigh.ibm.com (8.9.3/8.7/RTP-ral-1.0) with ESMTP id NAA10430; Wed, 3 May 2000 13:20:54 -0400
Message-Id: <200005031720.NAA10430@ludwigia.raleigh.ibm.com>
To: Bob Hinden <hinden@iprg.nokia.com>
cc: ngtrans@sunroof.eng.sun.com
Subject: Re: disconnected networks [was Re: (ngtrans) draft-ietf-ngtrans-6to4-04.txt 
In-Reply-To: Message from Bob Hinden <hinden@iprg.nokia.com> 
   of "Wed, 03 May 2000 10:02:10 PDT." <4.3.1.2.20000503095434.02129820@mailhost.iprg.nokia.com> 
Date: Wed, 03 May 2000 13:20:54 -0400
From: Thomas Narten <narten@raleigh.ibm.com>
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Thomas Narten <narten@raleigh.ibm.com>

> I tend to agree.  Were you thinking about a mechanisms where any
> site could get a global prefix (perhaps use a TLA like 6to4) but
> that these prefix would never be routed on the Internet?  Similar
> rules to the site-local prefix, but everyone can get their own.
> When the site does connect they get a new prefix and we make sure
> that the address selection rules pick a routable prefix when both
> types exist.

This scheme will only benefit those folks with the foresight to get
such addresses and then in addition not to use site-local
addresses. Would it be better to dust off the previous idea of putting
in some almost-random value in the left most bits of the site-local
prefix that would identify a particular organization? Those bits
should not overlap into the SLA bits and probably allow a couple more
bytes for post-merger routing heirarchy. If we needed even more
assurance that such addresses were likely unique, have a registry from
which folks could be assigned a specific site ID value that they would
use to put in their bits. Allocations are made first-come first-serve,
from a flat identifier space.

We could then have recommendations along the lines of:

1) leave the middle to left bits of site-local addresses zeros as they
   are now.

2) or pick a random value (and provide suggestsions on how to pick
   good values) for those who are worried

3) or get a properly assigned value, for those who really really
   worry.

Thomas


From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 13:54:36 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA07495
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 13:54:35 -0400 (EDT)
Received: from engmail1.Eng.Sun.COM ([129.146.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id LAA29156;
	Wed, 3 May 2000 11:53:27 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id KAA11829;
	Wed, 3 May 2000 10:50:56 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43Hhhn19078
	for ngtrans-dist; Wed, 3 May 2000 10:43:43 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43HhY719071
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 10:43:35 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail2.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id KAA12860;
	Wed, 3 May 2000 10:43:16 -0700 (PDT)
Received: from ginger.cs.utk.edu (GINGER.CS.UTK.EDU [128.169.95.11])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id KAA10440;
	Wed, 3 May 2000 10:43:10 -0700 (PDT)
Received: from ginger.cs.utk.edu (LOCALHOST [127.0.0.1])
        by ginger.cs.utk.edu (cf 8.9.3) with ESMTP id NAA13301;
        Wed, 3 May 2000 13:45:01 -0400 (EDT)
Message-Id: <200005031745.NAA13301@ginger.cs.utk.edu>
X-URI: http://www.cs.utk.edu/~moore/
From: Keith Moore <moore@cs.utk.edu>
To: Bob Hinden <hinden@iprg.nokia.com>
cc: Brian E Carpenter <brian@hursley.ibm.com>,
        Alain Durand <alain.durand@eng.sun.com>,
        Richard Draves <richdr@microsoft.com>,
        "'Keith Moore'" <moore@cs.utk.edu>, ngtrans@sunroof.eng.sun.com
Subject: Re: disconnected networks [was Re: (ngtrans) draft-ietf-ngtrans-6to4-04.txt 
In-reply-to: Your message of "Wed, 03 May 2000 10:02:10 PDT."
             <4.3.1.2.20000503095434.02129820@mailhost.iprg.nokia.com> 
Date: Wed, 03 May 2000 13:45:01 -0400
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Keith Moore <moore@cs.utk.edu>

> I tend to agree.  Were you thinking about a mechanisms where any site could 
> get a global prefix (perhaps use a TLA like 6to4) but that these prefix 
> would never be routed on the Internet?  Similar rules to the site-local 
> prefix, but everyone can get their own.  When the site does connect they 
> get a new prefix and we make sure that the address selection rules pick a 
> routable prefix when both types exist.

I started to suggest much the same thing...also making the mechanism
accessible via a web page, for free.  but I'm not sure how much it helps.  

It does get rid of the potential for address collisions when two 
sites using site-local addresses connect to one another.   (actually
if MAC addresses are used for the address suffix, the real problem is
more likely to be collisions in different uses of the "routing" bits 
of the site-local addresses than in the entirety of the addresses)

however when a site connects to the global Internet for the first time,
it doesn't matter whether it was using site-local addresses or 
globally-unique private addresses - it still essentially has to renumber.
so renumbering had better work well.

also, regarding address selection during renumbering - you can't count
on address-selection rules to choose the right prefix.  the host has
to know which address prefixes are in the process of being deprecated 
(and therefore must be maintained for existing connections) vs. which 
ones are available for use in new connections.  I think the same would
be true when "renumbering" from globally-unique private addresses to
public addresses (or vice versa, in case a site decided to disconnect...).
that is, address selection between globally-unique private addresses
and public addresses should not be based on the prefix alone but should
use the renumbering mechanism, and it should be explicit (dictated by the
network administrators) rather than implicit (based purely on the host's
knowledge of which prefixes are available).

There is a problem during the transition from disconnected to connected  
(or vice versa): since the same set of address prefixes might not become
visible to all hosts in the conversation at once, and since the 
globally-unique private address of a host might not be usable by all 
of the other hosts in that conversation, it is entirely possible that 
a host that is participating in a distributed conversation will choose 
an address for its connection endpoint which is not usable by all other 
hosts in that conversation.  When again argues for the transition to be
explicitly managed. 

Keith

p.s. I was tempted to suggest that ISPs could advertise routes for
globally-unique private addresses prefixes during a brief transition 
phase between initial connection to the Internet and the site's 
renumbering to use aggregatable addresses.  and of course ISPs can
do this if they want to.  but even if such routes were only advertised
for a few days, this could be really taxing of the network during 
periods where there are large numbers of new v6 nets being added.

problem is, folks who connect to the global IPv6 internet are going to
want it to start working right away.


From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 19:51:49 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA12928
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 19:51:48 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id RAA10583;
	Wed, 3 May 2000 17:50:51 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id QAA14952;
	Wed, 3 May 2000 16:48:30 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43NeAw19439
	for ngtrans-dist; Wed, 3 May 2000 16:40:10 -0700 (PDT)
Received: from jurassic.eng.sun.com (jurassic [129.146.89.31] (may be forged))
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43Ne1719432
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 16:40:01 -0700 (PDT)
Received: from awe174-18 (awe174-18.AWE.Sun.COM [192.29.174.18])
	by jurassic.eng.sun.com (8.10.1+Sun/8.10.1) with SMTP id e43Ndxh181887;
	Wed, 3 May 2000 16:39:59 -0700 (PDT)
Date: Wed, 3 May 2000 16:39:12 -0700 (PDT)
From: Erik Nordmark <Erik.Nordmark@eng.sun.com>
Subject: Re: (ngtrans) draft-ietf-ngtrans-6to4-04.txt
To: Alain Durand <alain.durand@eng.sun.com>
Cc: Richard Draves <richdr@microsoft.com>, "'Keith Moore'" <moore@cs.utk.edu>,
        "'Brian E Carpenter'" <brian@hursley.ibm.com>,
        ngtrans@sunroof.eng.sun.com
In-Reply-To: "Your message with ID" <390F9AAD.B770F160@eng.sun.com>
Message-ID: <Roam.SIMC.2.0.6.957397152.29227.nordmark@jurassic>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Erik Nordmark <Erik.Nordmark@eng.sun.com>


> But another typical case is when 2 compagnies are merging and you want
> to merge their 2 networks, there is a pretty good chance that both networks
> were using the same SLA IDs, (say 0x0001, 0x0002,...) so you wil have
> conflicts

Alain,

To be more exact the issue is when two sites are merging.
Thus it all depends on at what granulatity the two companies
define their sites. If each of them have been a single site before the merge
and they wish to be a single site after the merge there would be an
issue.

But if they have followed the loose recommendations on a site (a building or 
a campus) renumbering should only be needed if IP devices are actually moving
between buildings/campuses.

Of course, if the companies have chosen to have "one site for each Internet
connection" or "one site for the whole company even though it has
multiple internet connections" they there might be a collision space
for the site-local addresses.

However, in certain cases there can be similar "collision" issues with the SLA
field for global addresses.
If the companies have X/48 and Y/48 before the merge getting service from
two different ISPs, and they decide as part of the merge to only retain
one of the ISPs they have to fit all addresses under X/48.
If both companies used SLA 0, 1, etc they would need to renumber.

In any case, router renumbering can help with the task of any
renumbering needed to resolve conflicts as long as this is done
before the two networks are glued together.

   Erik



From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 20:07:54 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA13192
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 20:07:53 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id SAA19151;
	Wed, 3 May 2000 18:06:50 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id RAA18401;
	Wed, 3 May 2000 17:04:54 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e43Nv9819472
	for ngtrans-dist; Wed, 3 May 2000 16:57:09 -0700 (PDT)
Received: from jurassic.eng.sun.com (jurassic [129.146.84.31] (may be forged))
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e43Nv3719465
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 16:57:03 -0700 (PDT)
Received: from eng.sun.com (sappey.Eng.Sun.COM [129.146.85.69])
	by jurassic.eng.sun.com (8.10.1+Sun/8.10.1) with ESMTP id e43Nv3h184407;
	Wed, 3 May 2000 16:57:03 -0700 (PDT)
Message-ID: <3910BC6C.BDF93F64@eng.sun.com>
Date: Wed, 03 May 2000 16:55:24 -0700
From: Alain Durand <alain.durand@eng.sun.com>
X-Mailer: Mozilla 4.7 [en] (X11; I; SunOS 5.8 sun4u)
X-Accept-Language: en
MIME-Version: 1.0
To: Erik Nordmark <Erik.Nordmark@eng.sun.com>
CC: ngtrans@sunroof.eng.sun.com
Subject: Renumbering & site-local addresses (was Re: (ngtrans) 
 draft-ietf-ngtrans-6to4-04.txt)
References: <Roam.SIMC.2.0.6.957397152.29227.nordmark@jurassic>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Alain Durand <alain.durand@eng.sun.com>
Content-Transfer-Encoding: 7bit

Erik Nordmark wrote:

> However, in certain cases there can be similar "collision" issues with the SLA
> field for global addresses.
> If the companies have X/48 and Y/48 before the merge getting service from
> two different ISPs, and they decide as part of the merge to only retain
> one of the ISPs they have to fit all addresses under X/48.
> If both companies used SLA 0, 1, etc they would need to renumber.

Agree, but in that case,  if they do not use Site Local addresses, they would
also have the option to keep both prefixes and do not renumber at all.
It is then a trade-off between carrying two prefixes or renumbering.

    - Alain.





From owner-ngtrans@sunroof.eng.sun.com  Wed May  3 21:22:16 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA13825
	for <ngtrans-archive@odin.ietf.org>; Wed, 3 May 2000 21:22:15 -0400 (EDT)
Received: from engmail3.Eng.Sun.COM ([129.144.170.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id TAA24041;
	Wed, 3 May 2000 19:21:15 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id SAA12701;
	Wed, 3 May 2000 18:20:36 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e441Apa19555
	for ngtrans-dist; Wed, 3 May 2000 18:10:51 -0700 (PDT)
Received: from jurassic.eng.sun.com (jurassic [129.146.81.144] (may be forged))
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e441Ag719548
	for <ngtrans@sunroof.eng.sun.com>; Wed, 3 May 2000 18:10:42 -0700 (PDT)
Received: from awe174-18 (awe174-18.AWE.Sun.COM [192.29.174.18])
	by jurassic.eng.sun.com (8.10.1+Sun/8.10.1) with SMTP id e441Afh195671;
	Wed, 3 May 2000 18:10:41 -0700 (PDT)
Date: Wed, 3 May 2000 18:09:53 -0700 (PDT)
From: Erik Nordmark <Erik.Nordmark@eng.sun.com>
Subject: Re: Renumbering & site-local addresses (was Re: (ngtrans)  draft-ietf-ngtrans-6to4-04.txt)
To: Alain Durand <alain.durand@eng.sun.com>
Cc: Erik Nordmark <Erik.Nordmark@eng.sun.com>, ngtrans@sunroof.eng.sun.com
In-Reply-To: "Your message with ID" <3910BC6C.BDF93F64@eng.sun.com>
Message-ID: <Roam.SIMC.2.0.6.957402593.32595.nordmark@jurassic>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; CHARSET=US-ASCII
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Erik Nordmark <Erik.Nordmark@eng.sun.com>


> > However, in certain cases there can be similar "collision" issues with the
SLA > > field for global addresses.
> > If the companies have X/48 and Y/48 before the merge getting service from
> > two different ISPs, and they decide as part of the merge to only retain
> > one of the ISPs they have to fit all addresses under X/48.
> > If both companies used SLA 0, 1, etc they would need to renumber.
> 
> Agree, but in that case,  if they do not use Site Local addresses, they would
> also have the option to keep both prefixes and do not renumber at all. It is
> then a trade-off between carrying two prefixes or renumbering.

If the merged company terminates the relationship with the ISP that gave
them the Y/48 prefix I think the company might not be able to continue to use
that prefix. Or it might no longer be routed.

Of course, they can ask the ISP that gave them X/48 for an
additional prefix (W/48) and then renumber Y/48 into W/48.
Their success in doing so of course depend on the allocation policies
set by the registries and might be a function of how "densly" they are 
using the SLA field in the X/48 prefix.

The above details ois all speculation. But the point is that merging two
sites into a single site can cause conflicts both for the subnet field 
in site-local addresses and in some cases the SLA field in the global
address(es).

   Erik



From owner-ngtrans@sunroof.eng.sun.com  Mon May 15 14:33:51 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA08444
	for <ngtrans-archive@odin.ietf.org>; Mon, 15 May 2000 14:33:50 -0400 (EDT)
Received: from engmail3.Eng.Sun.COM ([129.144.170.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id MAA05299;
	Mon, 15 May 2000 12:26:38 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail3.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id LAA14656;
	Mon, 15 May 2000 11:16:16 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4FIAhJ01600
	for ngtrans-dist; Mon, 15 May 2000 11:10:43 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4FIAa701593
	for <ngtrans@sunroof.Eng.Sun.COM>; Mon, 15 May 2000 11:10:36 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail1.Eng.Sun.COM (8.9.1b+Sun/8.9.1/ENSMAIL,v1.6) with ESMTP id LAA10366
	for <ngtrans@sunroof.Eng.Sun.COM>; Mon, 15 May 2000 11:10:34 -0700 (PDT)
Received: from mail-gw.hursley.ibm.com (mail-gw.hursley.ibm.com [194.196.110.15])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id LAA27931
	for <ngtrans@sunroof.Eng.Sun.COM>; Mon, 15 May 2000 11:10:32 -0700 (PDT)
Received: from sp3at21.hursley.ibm.com (sp3at21.hursley.ibm.com [9.20.45.21]) by mail-gw.hursley.ibm.com (AIX4.3/UCB 8.8.8/8.8.8) with ESMTP id TAA19048; Mon, 15 May 2000 19:09:45 +0100
Received: from hursley.ibm.com (gsine04.us.sine.ibm.com [9.14.6.44]) by sp3at21.hursley.ibm.com (AIX4.2/UCB 8.7/8.7.3) with ESMTP id TAA26868; Mon, 15 May 2000 19:09:40 +0100 (BST)
Message-ID: <39203CB1.E9D2EA2E@hursley.ibm.com>
Date: Mon, 15 May 2000 13:06:41 -0500
From: Brian E Carpenter <brian@hursley.ibm.com>
Organization: IBM
X-Mailer: Mozilla 4.61 [en] (Win98; I)
X-Accept-Language: en,fr
MIME-Version: 1.0
To: NGTRANS List <ngtrans@sunroof.eng.sun.com>, phil@nesser.com
Subject: (ngtrans) draft-ietf-ngtrans-ipv4survey-00
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Brian E Carpenter <brian@hursley.ibm.com>
Content-Transfer-Encoding: 7bit

Phil,

This draft is an excellent start on something we really have to do.

Personally I'd be inclined to add a clear statement of exactly why
we need the survey, and shorten the historical perspective a bit,
but that's a matter of taste. Apart from that, I wish you strength
in completing the document.

    Brian


From owner-ngtrans@sunroof.eng.sun.com  Fri May 19 07:11:40 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA02846
	for <ngtrans-archive@odin.ietf.org>; Fri, 19 May 2000 07:11:39 -0400 (EDT)
Received: from engmail2.Eng.Sun.COM ([129.146.1.25])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id FAA28108;
	Fri, 19 May 2000 05:10:45 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail2.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id EAA22316;
	Fri, 19 May 2000 04:06:37 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4JAq5J05599
	for ngtrans-dist; Fri, 19 May 2000 03:52:05 -0700 (PDT)
Received: from engmail3.Eng.Sun.COM (engmail3 [129.144.170.5])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4JAps705592
	for <ngtrans@sunroof.eng.sun.com>; Fri, 19 May 2000 03:51:56 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail3.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id DAA10491
	for <ngtrans@sunroof.eng.sun.com>; Fri, 19 May 2000 03:51:53 -0700 (PDT)
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id DAA29552
	for <ngtrans@sunroof.eng.sun.com>; Fri, 19 May 2000 03:51:52 -0700 (PDT)
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA02505;
	Fri, 19 May 2000 06:51:50 -0400 (EDT)
Message-Id: <200005191051.GAA02505@ietf.org>
Mime-Version: 1.0
Content-Type: Multipart/Mixed; Boundary="NextPart"
To: IETF-Announce: ;
Cc: ngtrans@sunroof.eng.sun.com
From: Internet-Drafts@ietf.org
Subject: (ngtrans) I-D ACTION:draft-ietf-ngtrans-6to4-05.txt
Date: Fri, 19 May 2000 06:51:50 -0400
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Internet-Drafts@ietf.org

--NextPart

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Next Generation Transition Working Group of the IETF.

	Title		: Connection of IPv6 Domains via IPv4 Clouds without 
                          Explicit Tunnels
	Author(s)	: B. Carpenter, K. Moore
	Filename	: draft-ietf-ngtrans-6to4-05.txt
	Pages		: 22
	Date		: 18-May-00
	
This memo specifies an optional mechanism for assigning a unique
IPv6 address prefix to any site that currently has at least one
globally unique IPv4 address, and describes scenarios for using such
a prefix during the co-existence phase of IPv4 to IPv6 transition.
Effectively it treats the IPv4 network as a unicast link layer. Note
that this is not considered to be a long term solution and that sites
should migrate in due course to native IPv6 prefixes.
The motivation for this method is to allow isolated IPv6 domains or
hosts, attached to an IPv4 network which has no native IPv6 support,
to communicate with other such IPv6 domains or hosts with minimal
manual configuration. It also automatically provides a globally
unique IPv6 address prefix to any site with at least one globally
unique IPv4 address, even if combined with an IPv4 Network Address
Translator (NAT).

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-ngtrans-6to4-05.txt

Internet-Drafts are also available by anonymous FTP. Login with the username
"anonymous" and a password of your e-mail address. After logging in,
type "cd internet-drafts" and then
	"get draft-ietf-ngtrans-6to4-05.txt".

A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html 
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt


Internet-Drafts can also be obtained by e-mail.

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-ngtrans-6to4-05.txt".
	
NOTE:	The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.
		
		
Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

--NextPart
Content-Type: Multipart/Alternative; Boundary="OtherAccess"

--OtherAccess
Content-Type: Message/External-body;
	access-type="mail-server";
	server="mailserv@ietf.org"

Content-Type: text/plain
Content-ID:	<20000518091801.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-ngtrans-6to4-05.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-ngtrans-6to4-05.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<20000518091801.I-D@ietf.org>

--OtherAccess--

--NextPart--




From owner-ngtrans@sunroof.eng.sun.com  Mon May 22 01:41:17 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA19663
	for <ngtrans-archive@odin.ietf.org>; Mon, 22 May 2000 01:41:16 -0400 (EDT)
Received: from engmail1.Eng.Sun.COM ([129.146.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id XAA22873;
	Sun, 21 May 2000 23:39:05 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail1.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id WAA16362;
	Sun, 21 May 2000 22:33:27 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4M5CcI07647
	for ngtrans-dist; Sun, 21 May 2000 22:12:38 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4M5CH707640
	for <ngtrans@sunroof.eng.sun.com>; Sun, 21 May 2000 22:12:18 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail2.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id WAA23028
	for <ngtrans@sunroof.eng.sun.com>; Sun, 21 May 2000 22:12:11 -0700 (PDT)
Received: from mail1.es.net (mail1.es.net [198.128.3.181])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with SMTP id XAA16222
	for <ngtrans@sunroof.eng.sun.com>; Sun, 21 May 2000 23:12:11 -0600 (MDT)
Received: from rlfink.lbl.gov (Truckee) [131.243.136.210] 
	by mail1.es.net with esmtp (Exim 1.81 #2)
	id 12tkVd-0005ZZ-00; Sun, 21 May 2000 22:12:10 -0700
Message-Id: <4.2.2.20000521220332.01890ea0@imap2.es.net>
X-Sender: rlfink@imap2.es.net
X-Mailer: QUALCOMM Windows Eudora Pro Version 4.2.2 
Date: Sun, 21 May 2000 22:12:04 -0700
To: NGtrans List <ngtrans@sunroof.eng.sun.com>
From: Bob Fink <fink@es.net>
Subject: (ngtrans) working group last call for 6to4-05 forwarding as PS - closes
  5 June 2000
Cc: Alain Durand <Alain.Durand@sun.com>, Tony Hain <tonyhain@microsoft.com>,
        Randy Bush <randy@psg.com>, Brian E Carpenter <brian@hursley.ibm.com>,
        Keith Moore <moore@cs.utk.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Bob Fink <fink@es.net>

NGtrans Folk,

The 6to4-05 draft was published on 18 May to address outstanding issues 
from the -04 draft. The authors believe 6to4 is done, and that it is time 
to last call it for forwarding as a Proposed Standard.

Changes from 04 to 05 version:
- text on multicast largely removed (awaiting separate draft)
- noted that relay routers only accept traffic from specific sites
- clarified that V4ADDR must be configured on an IPv4 interface
- link-local text rewritten
- added note on NUD
- strengthened security text on address filtering
- general text clarifications and response to detail comments


Thus there will be a two-week ngtrans working group last call.

Comments to the list by 5 June 2000.


Thanks,

Bob

 > A New Internet-Draft is available from the on-line Internet-Drafts 
directories.
 > This draft is a work item of the Next Generation Transition Working 
Group of the IETF.
 >
 >         Title           : Connection of IPv6 Domains via IPv4 Clouds without
 >                           Explicit Tunnels
 >         Author(s)       : B. Carpenter, K. Moore
 >         Filename        : draft-ietf-ngtrans-6to4-05.txt
 >         Pages           : 22
 >         Date            : 18-May-00
 >
 > This memo specifies an optional mechanism for assigning a unique
 > IPv6 address prefix to any site that currently has at least one
 > globally unique IPv4 address, and describes scenarios for using such
 > a prefix during the co-existence phase of IPv4 to IPv6 transition.
 > Effectively it treats the IPv4 network as a unicast link layer. Note
 > that this is not considered to be a long term solution and that sites
 > should migrate in due course to native IPv6 prefixes.
 > The motivation for this method is to allow isolated IPv6 domains or
 > hosts, attached to an IPv4 network which has no native IPv6 support,
 > to communicate with other such IPv6 domains or hosts with minimal
 > manual configuration. It also automatically provides a globally
 > unique IPv6 address prefix to any site with at least one globally
 > unique IPv4 address, even if combined with an IPv4 Network Address
 > Translator (NAT).
 >
 > A URL for this Internet-Draft is:
 > http://www.ietf.org/internet-drafts/draft-ietf-ngtrans-6to4-05.txt



From owner-ngtrans@sunroof.eng.sun.com  Mon May 22 09:30:14 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA01882
	for <ngtrans-archive@odin.ietf.org>; Mon, 22 May 2000 09:30:14 -0400 (EDT)
Received: from engmail3.Eng.Sun.COM ([129.144.170.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id HAA13093;
	Mon, 22 May 2000 07:29:00 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail3.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id GAA19891;
	Mon, 22 May 2000 06:26:07 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4MDBuv07887
	for ngtrans-dist; Mon, 22 May 2000 06:11:56 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4MDBm707880
	for <ngtrans@sunroof.Eng.Sun.COM>; Mon, 22 May 2000 06:11:49 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail1.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id GAA11906
	for <ngtrans@sunroof.Eng.Sun.COM>; Mon, 22 May 2000 06:11:46 -0700 (PDT)
Received: from mail-gw.hursley.ibm.com (mail-gw.hursley.ibm.com [194.196.110.15])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id HAA05017
	for <ngtrans@sunroof.Eng.Sun.COM>; Mon, 22 May 2000 07:11:44 -0600 (MDT)
Received: from sp3at21.hursley.ibm.com (sp3at21.hursley.ibm.com [9.20.45.21]) by mail-gw.hursley.ibm.com (AIX4.3/UCB 8.8.8/8.8.8) with ESMTP id OAA17238 for <ngtrans@sunroof.Eng.Sun.COM>; Mon, 22 May 2000 14:11:42 +0100
Received: from hursley.ibm.com (gsine04.us.sine.ibm.com [9.14.6.44]) by sp3at21.hursley.ibm.com (AIX4.2/UCB 8.7/8.7.3) with ESMTP id OAA16432 for <ngtrans@sunroof.Eng.Sun.COM>; Mon, 22 May 2000 14:11:38 +0100 (BST)
Message-ID: <392931F3.3E267909@hursley.ibm.com>
Date: Mon, 22 May 2000 08:11:16 -0500
From: Brian E Carpenter <brian@hursley.ibm.com>
Organization: IBM
X-Mailer: Mozilla 4.61 [en] (Win98; I)
X-Accept-Language: en,fr
MIME-Version: 1.0
To: NGTRANS List <ngtrans@sunroof.eng.sun.com>
Subject: (ngtrans) Re: working group last call for 6to4-05 forwarding as PS - closes5 June 
 2000
References: <4.2.2.20000521220332.01890ea0@imap2.es.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Brian E Carpenter <brian@hursley.ibm.com>
Content-Transfer-Encoding: 7bit

Note that I found two glitches during the weekend: delete the first sentence
of section 3.1, and ignore the large blank space on page 6. These will
be fixed during editing.

   Brian

Bob Fink wrote:
> 
> NGtrans Folk,
> 
> The 6to4-05 draft was published on 18 May to address outstanding issues
> from the -04 draft. The authors believe 6to4 is done, and that it is time
> to last call it for forwarding as a Proposed Standard.
> 
> Changes from 04 to 05 version:
> - text on multicast largely removed (awaiting separate draft)
> - noted that relay routers only accept traffic from specific sites
> - clarified that V4ADDR must be configured on an IPv4 interface
> - link-local text rewritten
> - added note on NUD
> - strengthened security text on address filtering
> - general text clarifications and response to detail comments
> 
> Thus there will be a two-week ngtrans working group last call.
> 
> Comments to the list by 5 June 2000.
> 
> Thanks,
> 
> Bob
> 
>  > A New Internet-Draft is available from the on-line Internet-Drafts
> directories.
>  > This draft is a work item of the Next Generation Transition Working
> Group of the IETF.
>  >
>  >         Title           : Connection of IPv6 Domains via IPv4 Clouds without
>  >                           Explicit Tunnels
>  >         Author(s)       : B. Carpenter, K. Moore
>  >         Filename        : draft-ietf-ngtrans-6to4-05.txt
>  >         Pages           : 22
>  >         Date            : 18-May-00
>  >
>  > This memo specifies an optional mechanism for assigning a unique
>  > IPv6 address prefix to any site that currently has at least one
>  > globally unique IPv4 address, and describes scenarios for using such
>  > a prefix during the co-existence phase of IPv4 to IPv6 transition.
>  > Effectively it treats the IPv4 network as a unicast link layer. Note
>  > that this is not considered to be a long term solution and that sites
>  > should migrate in due course to native IPv6 prefixes.
>  > The motivation for this method is to allow isolated IPv6 domains or
>  > hosts, attached to an IPv4 network which has no native IPv6 support,
>  > to communicate with other such IPv6 domains or hosts with minimal
>  > manual configuration. It also automatically provides a globally
>  > unique IPv6 address prefix to any site with at least one globally
>  > unique IPv4 address, even if combined with an IPv4 Network Address
>  > Translator (NAT).
>  >
>  > A URL for this Internet-Draft is:
>  > http://www.ietf.org/internet-drafts/draft-ietf-ngtrans-6to4-05.txt


From owner-ngtrans@sunroof.eng.sun.com  Mon May 22 11:23:17 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA03604
	for <ngtrans-archive@odin.ietf.org>; Mon, 22 May 2000 11:23:16 -0400 (EDT)
Received: from engmail2.Eng.Sun.COM ([129.146.1.25])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id JAA24002;
	Mon, 22 May 2000 09:21:42 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail2.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id IAA05100;
	Mon, 22 May 2000 08:20:07 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4MFBSu07960
	for ngtrans-dist; Mon, 22 May 2000 08:11:28 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4MFBH707953
	for <ngtrans@sunroof.eng.sun.com>; Mon, 22 May 2000 08:11:18 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail1.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id IAA25835
	for <ngtrans@sunroof.eng.sun.com>; Mon, 22 May 2000 08:11:15 -0700 (PDT)
Received: from jaws.cisco.com (jaws.cisco.com [198.135.0.150])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id JAA16514;
	Mon, 22 May 2000 09:11:11 -0600 (MDT)
Received: (from otroan@localhost)
	by jaws.cisco.com (8.8.8/2.6/Cisco List Logging/8.8.8) id QAA27362;
	Mon, 22 May 2000 16:09:25 +0100 (BST)
X-Authentication-Warning: jaws.cisco.com: otroan set sender to ot@cisco.com using -f
To: Brian E Carpenter <brian@hursley.ibm.com>
Cc: NGtrans List <ngtrans@sunroof.eng.sun.com>,
        Alain Durand <Alain.Durand@sun.com>,
        Tony Hain <tonyhain@microsoft.com>, Randy Bush <randy@psg.com>,
        Bob Fink <fink@es.net>, Keith Moore <moore@cs.utk.edu>
Subject: Re: (ngtrans) working group last call for 6to4-05 forwarding as PS - closes 5 June 2000
References: <4.2.2.20000521220332.01890ea0@imap2.es.net>
From: Ole Troan <ot@cisco.com>
Date: 22 May 2000 16:09:25 +0100
In-Reply-To: Bob Fink's message of "Sun, 21 May 2000 22:12:04 -0700"
Message-ID: <7t54s7qwriy.fsf@jaws.cisco.com>
Lines: 43
User-Agent: Gnus/5.0806 (Gnus v5.8.6) Emacs/20.3
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Ole Troan <ot@cisco.com>

"5.2 Mixed scenario with relay to native IPv6

[...]

2. There are two deployment options for a 6to4 exterior routing
   domain:

   2.1 No IPv6 exterior routing protocol is used. The 6to4 routers using
   a given relay router each have a default IPv6 route pointing to the
   relay router. The relay router MAY apply source address based filters
   to accept traffic only from  specific 6to4 routers.

   2.2 An IPv6 exterior routing protocol is used. The set of 6to4
   routers using a given relay router obtain native IPv6 routes from the
   relay router using a routing protocol such as BGP4+ [RFC 2283,
   BGP4+]. The relay router will advertise whatever native IPv6 routing
   prefixes are appropriate on its 6to4 pseudo-interface. These prefixes
   will indicate the regions of native IPv6 topology that the relay
   router is willing to relay to. Their choice is a matter of routing
   policy. It is necessary for network operators to carefully consider
   desirable traffic patterns and topology when choosing the scope of
   such routing advertisements. The relay router will establish BGP
   peering only with specific 6to4 routers whose traffic it is willing
   to accept.

   Although this solution is more complex, it provides effective access
   control, i.e. BGP4+ policy controls which 6to4 routers are able to
   use which relay router.
"

I don't understand how BGP policies can be used for access control. any
6to4 router can point a default route to any relay router, as long as
the relay's 6to4 address is known. BGP shouldn't matter.
the only way to provide access control is by using filters.

I'm not particularly fond of the idea of announcing just "appropriate
prefixes" to the 6to4 domain. this would lead to the situation where
the 6to4 site could only reach some parts of the IPv6 Internet.
a 6to4 site needs at least one relay which provides a default route,
or a full routing table.

/ot



From owner-ngtrans@sunroof.eng.sun.com  Mon May 22 15:17:44 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA08335
	for <ngtrans-archive@odin.ietf.org>; Mon, 22 May 2000 15:17:43 -0400 (EDT)
Received: from engmail3.Eng.Sun.COM ([129.144.170.5])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id NAA27395;
	Mon, 22 May 2000 13:15:39 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail3.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id LAA03508;
	Mon, 22 May 2000 11:56:39 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4MIphd08129
	for ngtrans-dist; Mon, 22 May 2000 11:51:43 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4MIpY708122
	for <ngtrans@sunroof.eng.sun.com>; Mon, 22 May 2000 11:51:35 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail2.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id LAA29130
	for <ngtrans@sunroof.eng.sun.com>; Mon, 22 May 2000 11:51:33 -0700 (PDT)
Received: from mail-gw.hursley.ibm.com (mail-gw.hursley.ibm.com [194.196.110.15])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id MAA08203;
	Mon, 22 May 2000 12:51:30 -0600 (MDT)
Received: from sp3at21.hursley.ibm.com (sp3at21.hursley.ibm.com [9.20.45.21]) by mail-gw.hursley.ibm.com (AIX4.3/UCB 8.8.8/8.8.8) with ESMTP id TAA30870; Mon, 22 May 2000 19:49:18 +0100
Received: from hursley.ibm.com (gsine04.us.sine.ibm.com [9.14.6.44]) by sp3at21.hursley.ibm.com (AIX4.2/UCB 8.7/8.7.3) with ESMTP id TAA21298; Mon, 22 May 2000 19:49:09 +0100 (BST)
Message-ID: <392980F0.CE29E7F9@hursley.ibm.com>
Date: Mon, 22 May 2000 13:48:16 -0500
From: Brian E Carpenter <brian@hursley.ibm.com>
Organization: IBM
X-Mailer: Mozilla 4.61 [en] (Win98; I)
X-Accept-Language: en,fr
MIME-Version: 1.0
To: Ole Troan <ot@cisco.com>
CC: NGtrans List <ngtrans@sunroof.eng.sun.com>,
        Alain Durand <Alain.Durand@sun.com>,
        Tony Hain <tonyhain@microsoft.com>, Randy Bush <randy@psg.com>,
        Bob Fink <fink@es.net>, Keith Moore <moore@cs.utk.edu>
Subject: Re: (ngtrans) working group last call for 6to4-05 forwarding as PS - 
 closes 5 June 2000
References: <4.2.2.20000521220332.01890ea0@imap2.es.net> <7t54s7qwriy.fsf@jaws.cisco.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Brian E Carpenter <brian@hursley.ibm.com>
Content-Transfer-Encoding: 7bit

See below...

Ole Troan wrote:
> 
> "5.2 Mixed scenario with relay to native IPv6
> 
> [...]
> 
> 2. There are two deployment options for a 6to4 exterior routing
>    domain:
> 
>    2.1 No IPv6 exterior routing protocol is used. The 6to4 routers using
>    a given relay router each have a default IPv6 route pointing to the
>    relay router. The relay router MAY apply source address based filters
>    to accept traffic only from  specific 6to4 routers.
> 
>    2.2 An IPv6 exterior routing protocol is used. The set of 6to4
>    routers using a given relay router obtain native IPv6 routes from the
>    relay router using a routing protocol such as BGP4+ [RFC 2283,
>    BGP4+]. The relay router will advertise whatever native IPv6 routing
>    prefixes are appropriate on its 6to4 pseudo-interface. These prefixes
>    will indicate the regions of native IPv6 topology that the relay
>    router is willing to relay to. Their choice is a matter of routing
>    policy. It is necessary for network operators to carefully consider
>    desirable traffic patterns and topology when choosing the scope of
>    such routing advertisements. The relay router will establish BGP
>    peering only with specific 6to4 routers whose traffic it is willing
>    to accept.
> 
>    Although this solution is more complex, it provides effective access
>    control, i.e. BGP4+ policy controls which 6to4 routers are able to
>    use which relay router.
> "
> 
> I don't understand how BGP policies can be used for access control. any
> 6to4 router can point a default route to any relay router, as long as
> the relay's 6to4 address is known. BGP shouldn't matter.
> the only way to provide access control is by using filters.

Perhaps "access control" is the wrong phrase. BGP policy will control
which regions of IPv6 space are accessed via which relay router, as in
the first diagram in section 5.4. That's just bog standard BGP.

> 
> I'm not particularly fond of the idea of announcing just "appropriate
> prefixes" to the 6to4 domain. this would lead to the situation where
> the 6to4 site could only reach some parts of the IPv6 Internet.
> a 6to4 site needs at least one relay which provides a default route,
> or a full routing table.

Very likely, but that is a matter of service provider policy, not something
that the IETF decides. Again, bog standard BGP.

    Brian


From owner-ngtrans@sunroof.eng.sun.com  Tue May 23 07:07:28 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA29901
	for <ngtrans-archive@odin.ietf.org>; Tue, 23 May 2000 07:07:27 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id FAA12509;
	Tue, 23 May 2000 05:04:48 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id DAA05791;
	Tue, 23 May 2000 03:59:16 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4NAa4O08547
	for ngtrans-dist; Tue, 23 May 2000 03:36:04 -0700 (PDT)
Received: from engmail1.Eng.Sun.COM (engmail1 [129.146.1.13])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4NAZp708540
	for <ngtrans@sunroof.eng.sun.com>; Tue, 23 May 2000 03:35:52 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail1.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id DAA02679
	for <ngtrans@sunroof.eng.sun.com>; Tue, 23 May 2000 03:35:45 -0700 (PDT)
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id DAA03441
	for <ngtrans@sunroof.eng.sun.com>; Tue, 23 May 2000 03:35:44 -0700 (PDT)
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA29361;
	Tue, 23 May 2000 06:35:43 -0400 (EDT)
Message-Id: <200005231035.GAA29361@ietf.org>
Mime-Version: 1.0
Content-Type: Multipart/Mixed; Boundary="NextPart"
To: IETF-Announce: ;
Cc: ngtrans@sunroof.eng.sun.com
From: Internet-Drafts@ietf.org
Subject: (ngtrans) I-D ACTION:draft-ietf-ngtrans-tcpudp-relay-01.txt
Date: Tue, 23 May 2000 06:35:33 -0400
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Internet-Drafts@ietf.org

--NextPart

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Next Generation Transition Working Group of the IETF.

	Title		: An IPv6-to-IPv4 transport relay translator
	Author(s)	: J. Hagino, K. Yamamoto
 	Filename	: draft-ietf-ngtrans-tcpudp-relay-01.txt
	Pages		: 7
	Date		: 22-May-00
	
The memo describes an IPv6-to-IPv4 transport relay translator (TRT).  It
enables IPv6-only hosts to exchange {TCP,UDP} traffic with IPv4-only
hosts.  A TRT system, which locates in the middle, translates
{TCP,UDP}/IPv6 to {TCP,UDP}/IPv4, or vice versa.

The draft talks about how to implement a TRT system using existing
technologies.  It does not define any new protocols.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-ngtrans-tcpudp-relay-01.txt

Internet-Drafts are also available by anonymous FTP. Login with the username
"anonymous" and a password of your e-mail address. After logging in,
type "cd internet-drafts" and then
	"get draft-ietf-ngtrans-tcpudp-relay-01.txt".

A list of Internet-Drafts directories can be found in
http://www.ietf.org/shadow.html 
or ftp://ftp.ietf.org/ietf/1shadow-sites.txt


Internet-Drafts can also be obtained by e-mail.

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-ngtrans-tcpudp-relay-01.txt".
	
NOTE:	The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.
		
		
Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

--NextPart
Content-Type: Multipart/Alternative; Boundary="OtherAccess"

--OtherAccess
Content-Type: Message/External-body;
	access-type="mail-server";
	server="mailserv@ietf.org"

Content-Type: text/plain
Content-ID:	<20000522111354.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-ngtrans-tcpudp-relay-01.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-ngtrans-tcpudp-relay-01.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<20000522111354.I-D@ietf.org>

--OtherAccess--

--NextPart--




From owner-ngtrans@sunroof.eng.sun.com  Tue May 23 10:42:43 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA04322
	for <ngtrans-archive@odin.ietf.org>; Tue, 23 May 2000 10:42:42 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id IAA29457;
	Tue, 23 May 2000 08:40:40 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id HAA26432;
	Tue, 23 May 2000 07:37:27 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4NEOwk08640
	for ngtrans-dist; Tue, 23 May 2000 07:24:58 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4NEOo708633
	for <ngtrans@sunroof.eng.sun.com>; Tue, 23 May 2000 07:24:52 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail2.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id HAA29450
	for <ngtrans@sunroof.eng.sun.com>; Tue, 23 May 2000 07:24:49 -0700 (PDT)
Received: from correo.teldat.es ([212.95.195.134])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id HAA11652
	for <ngtrans@sunroof.eng.sun.com>; Tue, 23 May 2000 07:24:46 -0700 (PDT)
Received: by CORREO with Internet Mail Service (5.5.2650.21)
	id <JT4XN97V>; Tue, 23 May 2000 16:26:56 +0200
Message-ID: <41FAD0CB3B6BD3118BE600C04F43DB202B4127@CORREO>
From: Noelia Garcia <NGARCIA@teldat.es>
To: NGTRANS List <ngtrans@sunroof.eng.sun.com>
Subject: (ngtrans) Question about Neighbor Discovery over tunnels
Date: Tue, 23 May 2000 16:26:56 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Noelia Garcia <NGARCIA@teldat.es>

I have developed an IPv6 stack for a router; to connect it to the 6bone  I
am implementing  tunnels  Ipv6 over Ipv4 according to the Internet Draft "
Transition Mechanisms for Ipv6 Hosts and Routers", but I have the next
problem: 
When the Ipv4 connectivity in the tunnel fails (because of an external
cause), the router does not realize of that , continues routing packests
through  the tunnel and these packets are lost. If I implement Neighbor
Unreachability Detection for the link local address of the end side of
tunnel, I could desactive this tunnel when no answer is received, and these
packets could use  an alternative path to reach the destination; however I
do not receive response to  the probe packets used by NUD I send . Therefore
I do not know how to detect that the IPv4 conectivity of one tunnel has
failed ( to desactivate it and use an alternative one) and how to detect
that is "up" again to activate it.
Could someone helping me???
Thanks.
Noelia.


From owner-ngtrans@sunroof.eng.sun.com  Tue May 23 14:16:52 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA07461
	for <ngtrans-archive@odin.ietf.org>; Tue, 23 May 2000 14:16:52 -0400 (EDT)
Received: from engmail4.Eng.Sun.COM ([129.144.134.6])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id MAA07073;
	Tue, 23 May 2000 12:14:05 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail4.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id LAA14749;
	Tue, 23 May 2000 11:12:18 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4NI5UM08870
	for ngtrans-dist; Tue, 23 May 2000 11:05:30 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4NI5M708863
	for <ngtrans@sunroof.eng.sun.com>; Tue, 23 May 2000 11:05:23 -0700 (PDT)
Received: from venus.Sun.COM (venus.EBay.Sun.COM [129.150.69.5])
	by engmail2.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id LAA20620
	for <ngtrans@sunroof.eng.sun.com>; Tue, 23 May 2000 11:05:22 -0700 (PDT)
Received: from mail2.microsoft.com (mail2.microsoft.com [131.107.3.124])
	by venus.Sun.COM (8.9.3+Sun/8.9.3) with SMTP id LAA01881
	for <ngtrans@sunroof.eng.sun.com>; Tue, 23 May 2000 11:05:21 -0700 (PDT)
Received: from 157.54.9.104 by mail2.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 23 May 2000 11:05:09 -0700 (Pacific Daylight Time)
Received: by INET-IMC-02 with Internet Mail Service (5.5.2651.58)
	id <L2FNDDPF>; Tue, 23 May 2000 11:05:09 -0700
Message-ID: <4D0A23B3F74DD111ACCD00805F31D8101CA220E2@RED-MSG-50>
From: Richard Draves <richdr@microsoft.com>
To: "'Noelia Garcia'" <NGARCIA@teldat.es>
Cc: NGTRANS List <ngtrans@sunroof.eng.sun.com>
Subject: RE: (ngtrans) Question about Neighbor Discovery over tunnels
Date: Tue, 23 May 2000 11:05:07 -0700
X-Mailer: Internet Mail Service (5.5.2651.58)
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: Richard Draves <richdr@microsoft.com>

Yes, it would be really really handy if Neighbor Unreachability Detection
worked over configured tunnels, but as you note I believe that many
implementations do not support it.

Another example is for a 6to4 gateway that knows about multiple 6to4 relays
- it wants to have NUD to detect unreachable/broken relays.

Rich

> -----Original Message-----
> From: Noelia Garcia [mailto:NGARCIA@teldat.es]
> Sent: Tuesday, May 23, 2000 7:27 AM
> To: NGTRANS List
> Subject: (ngtrans) Question about Neighbor Discovery over tunnels
> 
> 
> I have developed an IPv6 stack for a router; to connect it to 
> the 6bone  I
> am implementing  tunnels  Ipv6 over Ipv4 according to the 
> Internet Draft "
> Transition Mechanisms for Ipv6 Hosts and Routers", but I have the next
> problem: 
> When the Ipv4 connectivity in the tunnel fails (because of an external
> cause), the router does not realize of that , continues 
> routing packests
> through  the tunnel and these packets are lost. If I 
> implement Neighbor
> Unreachability Detection for the link local address of the end side of
> tunnel, I could desactive this tunnel when no answer is 
> received, and these
> packets could use  an alternative path to reach the 
> destination; however I
> do not receive response to  the probe packets used by NUD I 
> send . Therefore
> I do not know how to detect that the IPv4 conectivity of one 
> tunnel has
> failed ( to desactivate it and use an alternative one) and 
> how to detect
> that is "up" again to activate it.
> Could someone helping me???
> Thanks.
> Noelia.
> 


From owner-ngtrans@sunroof.eng.sun.com  Wed May 24 19:39:49 2000
Received: from lukla.Sun.COM (lukla.Sun.COM [192.18.98.31])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA14175
	for <ngtrans-archive@odin.ietf.org>; Wed, 24 May 2000 19:39:48 -0400 (EDT)
Received: from engmail1.Eng.Sun.COM ([129.146.1.13])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id RAA12038;
	Wed, 24 May 2000 17:38:38 -0600 (MDT)
Received: from sunroof.eng.sun.com (sunroof.Eng.Sun.COM [129.146.168.88])
	by engmail1.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id QAA28779;
	Wed, 24 May 2000 16:31:19 -0700 (PDT)
Received: (from majordomo@localhost)
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) id e4ONOUg10741
	for ngtrans-dist; Wed, 24 May 2000 16:24:30 -0700 (PDT)
Received: from engmail2.Eng.Sun.COM (engmail2 [129.146.1.25])
	by sunroof.eng.sun.com (8.10.1+Sun/8.10.1.Beta0) with ESMTP id e4ONOM710734
	for <ngtrans@sunroof.eng.sun.com>; Wed, 24 May 2000 16:24:23 -0700 (PDT)
Received: from lukla.Sun.COM (lukla.Central.Sun.COM [129.147.5.31])
	by engmail2.Eng.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id QAA06276
	for <ngtrans@sunroof.eng.sun.com>; Wed, 24 May 2000 16:24:21 -0700 (PDT)
Received: from shuttle.wide.toshiba.co.jp (shuttle.wide.toshiba.co.jp [202.249.10.124])
	by lukla.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id RAA03246
	for <ngtrans@sunroof.eng.sun.com>; Wed, 24 May 2000 17:24:19 -0600 (MDT)
Received: from localhost ([3ffe:501:100f:13ff::e])
	by shuttle.wide.toshiba.co.jp (8.9.1+3.1W/8.9.1) with ESMTP id NAA06854;
	Wed, 24 May 2000 13:49:37 +0900 (JST)
Date: Wed, 24 May 2000 13:36:55 +0900
Message-ID: <y7vhfboftso.wl@condor.isl.rdc.toshiba.co.jp>
From: JINMEI Tatuya / =?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?=
 <jinmei@isl.rdc.toshiba.co.jp>
To: NGARCIA@teldat.es
Cc: ngtrans@sunroof.eng.sun.com
Subject: Re: (ngtrans) Question about Neighbor Discovery over tunnels
In-Reply-To: In your message of "Tue, 23 May 2000 16:26:56 +0200"
	 <41FAD0CB3B6BD3118BE600C04F43DB202B4127@CORREO>
References: <41FAD0CB3B6BD3118BE600C04F43DB202B4127@CORREO>
User-Agent: Wanderlust/2.2.15 (More Than Words) Emacs/20.6 Mule/4.0 (HANANOEN)
Organization: Research & Development Center, Toshiba Corp., Kawasaki, Japan.
MIME-Version: 1.0 (generated by SEMI 1.13.7 - "Awazu")
Content-Type: text/plain; charset=US-ASCII
X-Dispatcher: imput version 980905(IM100)
Lines: 30
Sender: owner-ngtrans@sunroof.eng.sun.com
Precedence: bulk
Reply-To: JINMEI Tatuya / =?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?= <jinmei@isl.rdc.toshiba.co.jp>

>>>>> On Tue, 23 May 2000 16:26:56 +0200, 
>>>>> Noelia Garcia <NGARCIA@teldat.es> said:

> I have developed an IPv6 stack for a router; to connect it to the 6bone  I
> am implementing  tunnels  Ipv6 over Ipv4 according to the Internet Draft "
> Transition Mechanisms for Ipv6 Hosts and Routers", but I have the next
> problem: 
> When the Ipv4 connectivity in the tunnel fails (because of an external
> cause), the router does not realize of that , continues routing packests
> through  the tunnel and these packets are lost. If I implement Neighbor
> Unreachability Detection for the link local address of the end side of
> tunnel, I could desactive this tunnel when no answer is received, and these
> packets could use  an alternative path to reach the destination; however I
> do not receive response to  the probe packets used by NUD I send . Therefore
> I do not know how to detect that the IPv4 conectivity of one tunnel has
> failed ( to desactivate it and use an alternative one) and how to detect
> that is "up" again to activate it.
> Could someone helping me???

How about just running an IPv6 routing protocol such as RIPng or BGP4+
on the tunnel?

					JINMEI, Tatuya
					Communication Platform Lab.
					Corporate R&D Center, Toshiba Corp.
					jinmei@isl.rdc.toshiba.co.jp

p.s. KAME's implementation (see http://www.kame.net) supports NUD on
p2p links, so you can expect routers based on the KAME's
implementation to react to your solicitation. 


