From jari.arkko@lmf.ericsson.se Mon May 3 15:50:21 2004 Received: from penguin.ericsson.se (penguin.ericsson.se [193.180.251.47]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA25432 for ; Mon, 3 May 2004 15:50:20 -0400 (EDT) Received: from esealmw142.al.sw.ericsson.se ([153.88.254.119]) by penguin.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i43JoKPA029843 for ; Mon, 3 May 2004 21:50:21 +0200 (MEST) Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118]) by esealmw142.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 3 May 2004 21:50:20 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt612.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id J884JP09; Mon, 3 May 2004 21:50:20 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i43JoCwg022698; Mon, 3 May 2004 21:50:12 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i43Jn8It025115; Mon, 3 May 2004 21:49:08 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i43Jn85x025114; Mon, 3 May 2004 21:49:08 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fep21-app.kolumbus.fi (fep21-0.kolumbus.fi [193.229.0.48]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i43Jn7It025110 for ; Mon, 3 May 2004 21:49:07 +0200 (MET DST) Received: from kolumbus.fi ([80.186.217.250]) by fep21-app.kolumbus.fi with ESMTP id <20040503194907.MUIK19565.fep21-app.kolumbus.fi@kolumbus.fi> for ; Mon, 3 May 2004 22:49:07 +0300 Message-ID: <4096A16E.3020707@kolumbus.fi> Date: Mon, 03 May 2004 22:45:50 +0300 From: Jari Arkko User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7b) Gecko/20040316 X-Accept-Language: en-us, en MIME-Version: 1.0 To: SEND WG Subject: (fwd) Last Call: 'SEcure Neighbor Discovery (SEND)' to Proposed Standard Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 03 May 2004 19:50:20.0238 (UTC) FILETIME=[DD88B6E0:01C43147] Content-Transfer-Encoding: 7bit From: The IESG Subject: Last Call: 'SEcure Neighbor Discovery (SEND)' to Proposed Standard Reply-to: iesg@ietf.org Message-Id: Sender: Nobody Date: Mon, 03 May 2004 13:15:17 -0400 The IESG has received a request from the Securing Neighbor Discovery WG to consider the following document: - 'SEcure Neighbor Discovery (SEND) ' as a Proposed Standard The IESG plans to make a decision in the next few weeks, and solicits final comments on this action. Please send any comments to the iesg@ietf.org or ietf@ietf.org mailing lists by 2004-05-17. The file can be obtained via http://www.ietf.org/internet-drafts/draft-ietf-send-ndopt-05.txt -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Sun May 9 09:59:06 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA04111 for ; Sun, 9 May 2004 09:59:06 -0400 (EDT) Received: from esealmw141.al.sw.ericsson.se ([153.88.254.120]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i49DuZAh000579 for ; Sun, 9 May 2004 15:56:35 +0200 Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw141.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Sun, 9 May 2004 15:56:35 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM9N9DHL; Sun, 9 May 2004 15:56:19 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i49DtKXA009698; Sun, 9 May 2004 15:55:20 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i49Ds9It006113; Sun, 9 May 2004 15:54:09 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i49Ds8P1006112; Sun, 9 May 2004 15:54:08 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from sequoia.muada.com (sequoia.muada.com [83.149.65.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i49Ds7It006108 for ; Sun, 9 May 2004 15:54:08 +0200 (MET DST) Received: from [127.0.0.1] (sequoia.muada.com [83.149.65.1]) by sequoia.muada.com (8.12.10/8.12.10) with ESMTP id i49Drv1C064585 for ; Sun, 9 May 2004 15:53:57 +0200 (CEST) (envelope-from iljitsch@muada.com) Mime-Version: 1.0 (Apple Message framework v613) Content-Transfer-Encoding: 7bit Message-Id: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> Content-Type: text/plain; charset=US-ASCII; format=flowed To: ietf-send@standards.ericsson.net From: Iljitsch van Beijnum Subject: CGA issues Date: Sun, 9 May 2004 15:54:10 +0200 X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 09 May 2004 13:56:35.0200 (UTC) FILETIME=[70E77000:01C435CD] Content-Transfer-Encoding: 7bit Gentlemen and gentle ladies, Let me dive right in... At the instigation of Brian Carpenter, one of the chairs of the site multihoming in IPv6 (multi6) wg, I've had a look at the CGA draft. There are two issues I want to bring up with regard to what's happening in multi6, and a third, unrelated issue. Presently, the multi6 wg is looking at multihoming mechanisms that involve a host receiving regular provider aggregatable addresses from two or more different ISPs, and switch addresses in mid-session if necessary. (As opposed to the current IPv4 way of achieving multihoming through the use of provider independent addressing, which won't scale.) The main issue is that for multi6, we are considering several mechanisms that work much more effeciently if the lower 80 or 64 bits of the IPv6 address stays the same when a rehoming event occurs. Since one of the inputs for CGA generation is the subnet prefix, it's not possible to generate an interface identifier value that is a valid CGA in two different subnets. Second, some form of cryptographically generated addresses could also be very useful for securing the interactions necessary to achieve multihoming. However, this would require those CGAs to be globally unique (at least statistically) and much longer-lived than what seems to be assumed in the current CGA draft. Finally, the draft doesn't specify the value of the universal/local and group bits in the cryptographically generated interface identifier. Obviously it's important that naive implementations don't adopt a link address that happens to have the group bit set, as will be the case in approximately 50% of all CGAs. Additionally, it is probably prudent to set the u/l bit to 1, as CGAs aren't "universal". -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Sun May 9 16:43:42 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA21106 for ; Sun, 9 May 2004 16:43:41 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i49KhdWR019847 for ; Sun, 9 May 2004 22:43:39 +0200 (MEST) Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Sun, 9 May 2004 22:43:39 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KNDH4GY4; Sun, 9 May 2004 22:43:48 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i49KhcXA012637; Sun, 9 May 2004 22:43:38 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i49KgxIt001851; Sun, 9 May 2004 22:42:59 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i49KgxJS001850; Sun, 9 May 2004 22:42:59 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from sequoia.muada.com (sequoia.muada.com [83.149.65.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i49KgwIt001846 for ; Sun, 9 May 2004 22:42:58 +0200 (MET DST) Received: from [127.0.0.1] (sequoia.muada.com [83.149.65.1]) by sequoia.muada.com (8.12.10/8.12.10) with ESMTP id i49Kgi1C071140; Sun, 9 May 2004 22:42:45 +0200 (CEST) (envelope-from iljitsch@muada.com) In-Reply-To: <409E54E0.1010900@kolumbus.fi> References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; delsp=yes; format=flowed Message-Id: <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> Content-Transfer-Encoding: 7bit Cc: ietf-send@standards.ericsson.net, Tuomas Aura From: Iljitsch van Beijnum Subject: Re: CGA issues Date: Sun, 9 May 2004 22:42:58 +0200 To: Jari Arkko X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 09 May 2004 20:43:39.0051 (UTC) FILETIME=[4EA757B0:01C43606] Content-Transfer-Encoding: 7bit On 9-mei-04, at 17:57, Jari Arkko wrote: >> The main issue is that for multi6, we are considering several >> mechanisms that work much more effeciently if the lower 80 or 64 bits >> of the IPv6 address stays the same when a rehoming event occurs. >> Since one of the inputs for CGA generation is the subnet prefix, it's >> not possible to generate an interface identifier value that is a >> valid CGA in two different subnets. > That's right, but can you explain how multi6 works better > if the interface identifier stays the same? We are considering having routers rewrite the upper 48 bits of the IPv6 address. Also, solutions that require an address mapping database are more efficient to implement if only the top 48 bits are different: this requires one entry per site, rather than one entry per host. > Presumably, a > multi6 solution would have to have some sort of a "handle" -- > interface identifier, fqdn, ephemeral key (wimp hash chain > or HIP HIT) to be able to correlate the sessions from the > multiple addresses. Where SEND (or RFC 3041) is used, the > interface identifier keeps changing. There are many proposals on the table, and they differ widely in this regard. However, when addresses change this requires communication, so it's better to avoid such a change if it's not strictly necessary. > However, if you assume > that SEND is being used, the key pair behind the CGA might > be a better alternative for the handle than the interface > identifier. I agree. See http://www.ietf.org/internet-drafts/draft-van-beijnum-multi6-cbhi -00.txt I obviously wasn't aware of CGA when I wrote this draft. > Are you trying to see how existing multi6 > solutions would work together with SEND, or are you trying > to see whether CGA could be used as a multi6 solution? CGA on it's own can't be a multihoming solution. I think it would be useful to reuse CGA as a security mechanism in a multihoming solution, but mostly I'm concerned with making SEND and multi6 work together without getting in each other's way. > Note that if you rely on the same IID across rehoming > events, you are precluding multihomed hosts, and just > talking about site multihoming. Multihomed hosts would > likely have a different IID on different interfaces. If a host has multiple interfaces, it is indeed likely that these will have different interface identifiers. However, this isn't strictly required and/or subject to change. Moreover, most hosts would have multiple addresses on one interface, in which case the same interface identifier is reused by the current IPv6 implementations I'm aware of. >> However, this would require those CGAs to be globally unique (at >> least statistically) and much longer-lived than what seems to be >> assumed in the current CGA draft. > Hmm... why? Do you want the interface identifier to be unique > or the public key to be unique? The former is much harder to > achieve, but I'm not sure it is needed. If you want to tell > your peer that you are now using a new IP address X, you'll > just sign a statement to this effect using your private > address. The peer verifies the signature and also ensures > that the public key is the same one you were using before. In this scenario you need to negotiate keys before a rehoming event occurs. This isn't very desirable, for two reasons: 1. Some applications use very many very short lived sessions (ie, HTTP) 2. Connectivity may temporarily run over an insecure link so that an attacker gets to set up malicious state that can be abused later A system where a crypto-based identifier is embedded in the IPv6 address has the advantage that after a rehoming event, the host can prove that it's still the same host as before, without the need to exchange key information before the rehoming event happens. -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Sun May 9 16:59:23 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA21570 for ; Sun, 9 May 2004 16:59:22 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i49KxNAh024563 for ; Sun, 9 May 2004 22:59:23 +0200 Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Sun, 9 May 2004 22:59:23 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM93AA15; Sun, 9 May 2004 22:59:23 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i49KxMXA012758; Sun, 9 May 2004 22:59:22 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i49KwoIt005129; Sun, 9 May 2004 22:58:50 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i49Kwob2005128; Sun, 9 May 2004 22:58:50 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fep22-app.kolumbus.fi (fep22-0.kolumbus.fi [193.229.0.60]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i49KwnIt005112 for ; Sun, 9 May 2004 22:58:49 +0200 (MET DST) Received: from kolumbus.fi ([80.186.217.250]) by fep22-app.kolumbus.fi with ESMTP id <20040509205849.VDK27645.fep22-app.kolumbus.fi@kolumbus.fi>; Sun, 9 May 2004 23:58:49 +0300 Message-ID: <409E9ABD.4040907@kolumbus.fi> Date: Sun, 09 May 2004 23:55:25 +0300 From: Jari Arkko User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7b) Gecko/20040316 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Iljitsch van Beijnum CC: ietf-send@standards.ericsson.net, Tuomas Aura Subject: Re: CGA issues References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> In-Reply-To: <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 09 May 2004 20:59:23.0641 (UTC) FILETIME=[81AC5690:01C43608] Content-Transfer-Encoding: 7bit Iljitsch van Beijnum wrote: >> Hmm... why? Do you want the interface identifier to be unique >> or the public key to be unique? The former is much harder to >> achieve, but I'm not sure it is needed. If you want to tell >> your peer that you are now using a new IP address X, you'll >> just sign a statement to this effect using your private >> address. The peer verifies the signature and also ensures >> that the public key is the same one you were using before. > > > In this scenario you need to negotiate keys before a rehoming event Not necessarily. Lets assume that you and I have communicated for a while, and that I have used the address Y. You don't know that Y = prefix1 | hash(my public key | ...), but it is. Then I move to prefix2, and adopt the address X = prefix2 | hash(my public key | ...). I'll tell you about it by sending you the following information: - Y - X - my public key - my signature At this point you will become aware that even my previous address was a CGA. And you can verify that both addresses have come from the same public key (even if the addresses are completely different). And you can verify my signature. And no information had to be communicated before the rehoming event occurred. > occurs. This isn't very desirable, for two reasons: > > 1. Some applications use very many very short lived sessions (ie, HTTP) > 2. Connectivity may temporarily run over an insecure link so that an > attacker gets to set up malicious state that can be abused later > > A system where a crypto-based identifier is embedded in the IPv6 > address has the advantage that after a rehoming event, the host can > prove that it's still the same host as before, without the need to > exchange key information before the rehoming event happens. Yes, but in the scheme that I outlined above there's no need for prior exchanges. --Jari -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Sun May 9 20:07:39 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA00039 for ; Sun, 9 May 2004 20:07:39 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4A07dWR005715 for ; Mon, 10 May 2004 02:07:39 +0200 (MEST) Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 02:07:38 +0200 Received: from esealnt614.al.sw.ericsson.se ([153.88.254.73]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KNDHVCV9; Mon, 10 May 2004 02:07:48 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KTFR4G8B; Sun, 9 May 2004 18:01:59 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i49G1fwg024500; Sun, 9 May 2004 18:01:41 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i49G0jIt003318; Sun, 9 May 2004 18:00:45 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i49G0jQW003317; Sun, 9 May 2004 18:00:45 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fep01-app.kolumbus.fi (fep01-0.kolumbus.fi [193.229.0.41]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i49G0iIt003312 for ; Sun, 9 May 2004 18:00:44 +0200 (MET DST) Received: from kolumbus.fi ([80.186.217.250]) by fep01-app.kolumbus.fi with ESMTP id <20040509160043.IZQC4346.fep01-app.kolumbus.fi@kolumbus.fi>; Sun, 9 May 2004 19:00:43 +0300 Message-ID: <409E54E0.1010900@kolumbus.fi> Date: Sun, 09 May 2004 18:57:20 +0300 From: Jari Arkko User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7b) Gecko/20040316 X-Accept-Language: en-us, en MIME-Version: 1.0 To: Iljitsch van Beijnum CC: ietf-send@standards.ericsson.net, Tuomas Aura Subject: Re: CGA issues References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> In-Reply-To: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 00:07:38.0524 (UTC) FILETIME=[CDF2B5C0:01C43622] Content-Transfer-Encoding: 7bit Iljitsch, Thanks for opening this discussion. Some comments inline: > At the instigation of Brian Carpenter, one of the chairs of the site > multihoming in IPv6 (multi6) wg, I've had a look at the CGA draft. There > are two issues I want to bring up with regard to what's happening in > multi6, and a third, unrelated issue. > > Presently, the multi6 wg is looking at multihoming mechanisms that > involve a host receiving regular provider aggregatable addresses from > two or more different ISPs, and switch addresses in mid-session if > necessary. (As opposed to the current IPv4 way of achieving multihoming > through the use of provider independent addressing, which won't scale.) > > The main issue is that for multi6, we are considering several mechanisms > that work much more effeciently if the lower 80 or 64 bits of the IPv6 > address stays the same when a rehoming event occurs. Since one of the > inputs for CGA generation is the subnet prefix, it's not possible to > generate an interface identifier value that is a valid CGA in two > different subnets. That's right, but can you explain how multi6 works better if the interface identifier stays the same? Presumably, a multi6 solution would have to have some sort of a "handle" -- interface identifier, fqdn, ephemeral key (wimp hash chain or HIP HIT) to be able to correlate the sessions from the multiple addresses. Where SEND (or RFC 3041) is used, the interface identifier keeps changing. However, if you assume that SEND is being used, the key pair behind the CGA might be a better alternative for the handle than the interface identifier. Are you trying to see how existing multi6 solutions would work together with SEND, or are you trying to see whether CGA could be used as a multi6 solution? Note that if you rely on the same IID across rehoming events, you are precluding multihomed hosts, and just talking about site multihoming. Multihomed hosts would likely have a different IID on different interfaces. > Second, some form of cryptographically generated addresses could also be > very useful for securing the interactions necessary to achieve Right. I guess this is what I was referring to above. > multihoming. However, this would require those CGAs to be globally > unique (at least statistically) and much longer-lived than what seems to > be assumed in the current CGA draft. Hmm... why? Do you want the interface identifier to be unique or the public key to be unique? The former is much harder to achieve, but I'm not sure it is needed. If you want to tell your peer that you are now using a new IP address X, you'll just sign a statement to this effect using your private address. The peer verifies the signature and also ensures that the public key is the same one you were using before. > Finally, the draft doesn't specify the value of the universal/local and > group bits in the cryptographically generated interface identifier. > Obviously it's important that naive implementations don't adopt a link > address that happens to have the group bit set, as will be the case in > approximately 50% of all CGAs. Additionally, it is probably prudent to > set the u/l bit to 1, as CGAs aren't "universal". Tuomas, can you say something about this? --Jari -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 01:39:58 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA10830 for ; Mon, 10 May 2004 01:39:57 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4A5duWR010766 for ; Mon, 10 May 2004 07:39:57 +0200 (MEST) Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 07:39:56 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KNDHW8XB; Mon, 10 May 2004 07:40:07 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4A5dmwg018598; Mon, 10 May 2004 07:39:48 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A5ctIt027282; Mon, 10 May 2004 07:38:55 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4A5ctKc027281; Mon, 10 May 2004 07:38:55 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from netcore.fi (netcore.fi [193.94.160.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A5csIt027277 for ; Mon, 10 May 2004 07:38:54 +0200 (MET DST) Received: from localhost (pekkas@localhost) by netcore.fi (8.11.6/8.11.6) with ESMTP id i4A5cn625125; Mon, 10 May 2004 08:38:49 +0300 Date: Mon, 10 May 2004 08:38:49 +0300 (EEST) From: Pekka Savola To: Jari Arkko cc: Iljitsch van Beijnum , , Tuomas Aura Subject: Re: CGA issues In-Reply-To: <409E9ABD.4040907@kolumbus.fi> Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 05:39:56.0780 (UTC) FILETIME=[3A1322C0:01C43651] On Sun, 9 May 2004, Jari Arkko wrote: > Iljitsch van Beijnum wrote: > >> Hmm... why? Do you want the interface identifier to be unique > >> or the public key to be unique? The former is much harder to > >> achieve, but I'm not sure it is needed. If you want to tell > >> your peer that you are now using a new IP address X, you'll > >> just sign a statement to this effect using your private > >> address. The peer verifies the signature and also ensures > >> that the public key is the same one you were using before. > > > > > > In this scenario you need to negotiate keys before a rehoming event > > Not necessarily. Lets assume that you and I have communicated > for a while, and that I have used the address Y. You don't know > that Y = prefix1 | hash(my public key | ...), but it is. Then > I move to prefix2, and adopt the address X = prefix2 | > hash(my public key | ...). I'll tell you about it by sending > you the following information: > > - Y > - X > - my public key > - my signature > > At this point you will become aware that even my previous address > was a CGA. And you can verify that both addresses have come from > the same public key (even if the addresses are completely different). > And you can verify my signature. And no information had to be > communicated before the rehoming event occurred. Doesn't this require you share a trust anchor with everyone in the Internet (the dreaded global PKI! :)? Unless I'm missing something, this has an entirely different set of assumptions than SEND currently has? -- Pekka Savola "You each name yourselves king, yet the Netcore Oy kingdom bleeds." Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 02:08:32 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA20208 for ; Mon, 10 May 2004 02:08:30 -0400 (EDT) Received: from esealmw141.al.sw.ericsson.se ([153.88.254.120]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4A68UWR016317 for ; Mon, 10 May 2004 08:08:30 +0200 (MEST) Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw141.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 08:08:30 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM93C7SY; Mon, 10 May 2004 08:08:30 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4A68TXA017132; Mon, 10 May 2004 08:08:29 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A67tIt003638; Mon, 10 May 2004 08:07:55 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4A67tQ3003637; Mon, 10 May 2004 08:07:55 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from ftmailgfi.HQ.Flarion.com (mail.flarion.com [63.103.94.23]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A67sIt003633 for ; Mon, 10 May 2004 08:07:54 +0200 (MET DST) Received: from ftmail2000.HQ.Flarion.com ([10.10.1.120]) by ftmailgfi.HQ.Flarion.com with Microsoft SMTPSVC(5.0.2195.6713); Mon, 10 May 2004 02:07:44 -0400 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Content-Class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Subject: RE: CGA issues Date: Mon, 10 May 2004 02:07:44 -0400 Message-ID: Thread-Topic: CGA issues Thread-Index: AcQ2UUDRlCGvODvZQIK52nJkQYvMOwAAjZ7w From: "Soliman Hesham" To: "Pekka Savola" , "Jari Arkko" CC: "Iljitsch van Beijnum" , , "Tuomas Aura" X-OriginalArrivalTime: 10 May 2004 06:07:44.0733 (UTC) FILETIME=[1C4090D0:01C43655] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by sw.ericsson.se id i4A67tIt003634 Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk Content-Transfer-Encoding: 8bit > > Not necessarily. Lets assume that you and I have communicated > > for a while, and that I have used the address Y. You don't know > > that Y = prefix1 | hash(my public key | ...), but it is. Then > > I move to prefix2, and adopt the address X = prefix2 | > > hash(my public key | ...). I'll tell you about it by sending > > you the following information: > > > > - Y > > - X > > - my public key > > - my signature > > > > At this point you will become aware that even my previous address > > was a CGA. And you can verify that both addresses have come from > > the same public key (even if the addresses are completely > different). > > And you can verify my signature. And no information had to be > > communicated before the rehoming event occurred. > > Doesn't this require you share a trust anchor with everyone in the > Internet (the dreaded global PKI! :)? Unless I'm missing something, > this has an entirely different set of assumptions than SEND currently > has? => Not really. This is the same problem that send solves: Address ownership. You want to prove (to the correspondent) that you are the same node even though the address changed. What Jari described will do just that. No need for PKI. Hesham > > -- > Pekka Savola "You each name yourselves king, yet the > Netcore Oy kingdom bleeds." > Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings > > -------------------------------------------------------------------- > To unsubscribe from this list, send email with "UNSUBSCRIBE" in the > body to . > Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html > -------------------------------------------------------------------- > ======================================================== This email may contain confidential and privileged material for the sole use of the intended recipient. Any review or distribution by others is strictly prohibited. If you are not the intended recipient please contact the sender and delete all copies. ======================================================== -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 02:29:59 2004 Received: from penguin.ericsson.se (penguin.ericsson.se [193.180.251.47]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA11445 for ; Mon, 10 May 2004 02:29:58 -0400 (EDT) Received: from esealmw142.al.sw.ericsson.se ([153.88.254.119]) by penguin.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4A6TwPA015986 for ; Mon, 10 May 2004 08:29:58 +0200 (MEST) Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw142.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 08:29:57 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM93DCHS; Mon, 10 May 2004 08:29:57 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4A6Tuwg020435; Mon, 10 May 2004 08:29:56 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A6TIIt007700; Mon, 10 May 2004 08:29:18 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4A6TH2o007699; Mon, 10 May 2004 08:29:17 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from netcore.fi (netcore.fi [193.94.160.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A6TGIt007695 for ; Mon, 10 May 2004 08:29:17 +0200 (MET DST) Received: from localhost (pekkas@localhost) by netcore.fi (8.11.6/8.11.6) with ESMTP id i4A6TEF25896; Mon, 10 May 2004 09:29:14 +0300 Date: Mon, 10 May 2004 09:29:14 +0300 (EEST) From: Pekka Savola To: Soliman Hesham cc: Jari Arkko , Iljitsch van Beijnum , , Tuomas Aura Subject: RE: CGA issues In-Reply-To: Message-ID: MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 06:29:57.0255 (UTC) FILETIME=[367F4970:01C43658] On Mon, 10 May 2004, Soliman Hesham wrote: > > > Not necessarily. Lets assume that you and I have communicated > > > for a while, and that I have used the address Y. You don't know > > > that Y = prefix1 | hash(my public key | ...), but it is. Then > > > I move to prefix2, and adopt the address X = prefix2 | > > > hash(my public key | ...). I'll tell you about it by sending > > > you the following information: > > > > > > - Y > > > - X > > > - my public key > > > - my signature > > > > > > At this point you will become aware that even my previous address > > > was a CGA. And you can verify that both addresses have come from > > > the same public key (even if the addresses are completely > > different). > > > And you can verify my signature. And no information had to be > > > communicated before the rehoming event occurred. > > > > Doesn't this require you share a trust anchor with everyone in the > > Internet (the dreaded global PKI! :)? Unless I'm missing something, > > this has an entirely different set of assumptions than SEND currently > > has? > > => Not really. This is the same problem that send solves: Address > ownership. You want to prove (to the correspondent) that you are > the same node even though the address changed. > What Jari described will do just that. No need for PKI. Yes -- sorry for confusing. As Jari said, you can prove that both addresses came from the same public key; you still cannot trust the public key because you don't know who it belongs to. (I.e., this only solves address ownership when you want to prove that the addresses were generated by the same node, but nothing more.) But in verification of this kind of "my address changed changed, I'm now reachable at [here]" -messages this is sufficient as you don't even want to authenticate or identify the user, just to be sure it's the addresses belong to the same guy. -- Pekka Savola "You each name yourselves king, yet the Netcore Oy kingdom bleeds." Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 03:10:54 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA14115 for ; Mon, 10 May 2004 03:10:53 -0400 (EDT) Received: from esealmw140.al.sw.ericsson.se ([153.88.254.121]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4A7AsWR001975 for ; Mon, 10 May 2004 09:10:54 +0200 (MEST) Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw140.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 09:10:53 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KNDHXRA1; Mon, 10 May 2004 09:11:04 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4A7AqXA018426; Mon, 10 May 2004 09:10:52 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A7A9It018761; Mon, 10 May 2004 09:10:09 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4A7A9Q4018760; Mon, 10 May 2004 09:10:09 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from ftmailgfi.HQ.Flarion.com (mail.flarion.com [63.103.94.23]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A7A7It018756 for ; Mon, 10 May 2004 09:10:07 +0200 (MET DST) Received: from ftmail2000.HQ.Flarion.com ([10.10.1.120]) by ftmailgfi.HQ.Flarion.com with Microsoft SMTPSVC(5.0.2195.6713); Mon, 10 May 2004 03:10:05 -0400 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Content-Class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Subject: RE: CGA issues Date: Mon, 10 May 2004 03:10:04 -0400 Message-ID: Thread-Topic: CGA issues Thread-Index: AcQ2WB3auLiwntIxTFm+J/TLrvUnDAABA3FA From: "Soliman Hesham" To: "Pekka Savola" CC: "Jari Arkko" , "Iljitsch van Beijnum" , , "Tuomas Aura" X-OriginalArrivalTime: 10 May 2004 07:10:05.0467 (UTC) FILETIME=[D1E746B0:01C4365D] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by sw.ericsson.se id i4A7A8It018757 Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk Content-Transfer-Encoding: 8bit > Yes -- sorry for confusing. > > As Jari said, you can prove that both addresses came from the same > public key; you still cannot trust the public key because you don't > know who it belongs to. (I.e., this only solves address > ownership when > you want to prove that the addresses were generated by the same node, > but nothing more.) => Right and I believe that's all we need for the multi6 part. We only trust the public key for this purpose. > > But in verification of this kind of "my address changed changed, I'm > now reachable at [here]" -messages this is sufficient as you don't > even want to authenticate or identify the user, just to be sure it's > the addresses belong to the same guy. => Correct. Hesham > > -- > Pekka Savola "You each name yourselves king, yet the > Netcore Oy kingdom bleeds." > Systems. Networks. Security. -- George R.R. Martin: A Clash of Kings > > ======================================================== This email may contain confidential and privileged material for the sole use of the intended recipient. Any review or distribution by others is strictly prohibited. If you are not the intended recipient please contact the sender and delete all copies. ======================================================== -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 04:56:48 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA18035 for ; Mon, 10 May 2004 04:56:48 -0400 (EDT) Received: from esealmw141.al.sw.ericsson.se ([153.88.254.120]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4A8unWR005097 for ; Mon, 10 May 2004 10:56:49 +0200 (MEST) Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw141.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 10:56:48 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM931LQQ; Mon, 10 May 2004 10:56:48 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4A8uhwg027347; Mon, 10 May 2004 10:56:43 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A8tuIt013251; Mon, 10 May 2004 10:55:56 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4A8tucP013250; Mon, 10 May 2004 10:55:56 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from sequoia.muada.com (sequoia.muada.com [83.149.65.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A8ttIt013246 for ; Mon, 10 May 2004 10:55:55 +0200 (MET DST) Received: from [127.0.0.1] (sequoia.muada.com [83.149.65.1]) by sequoia.muada.com (8.12.10/8.12.10) with ESMTP id i4A8tf1C082502; Mon, 10 May 2004 10:55:41 +0200 (CEST) (envelope-from iljitsch@muada.com) In-Reply-To: <409E9ABD.4040907@kolumbus.fi> References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <409E9ABD.4040907@kolumbus.fi> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: Content-Transfer-Encoding: 7bit Cc: ietf-send@standards.ericsson.net, Tuomas Aura From: Iljitsch van Beijnum Subject: Re: CGA issues Date: Mon, 10 May 2004 10:55:56 +0200 To: Jari Arkko X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 08:56:48.0819 (UTC) FILETIME=[BA994030:01C4366C] Content-Transfer-Encoding: 7bit On 9-mei-04, at 22:55, Jari Arkko wrote: >> In this scenario you need to negotiate keys before a rehoming event > Not necessarily. Lets assume that you and I have communicated > for a while, and that I have used the address Y. You don't know > that Y = prefix1 | hash(my public key | ...), but it is. Then > I move to prefix2, and adopt the address X = prefix2 | > hash(my public key | ...). I'll tell you about it by sending > you the following information: > - Y > - X > - my public key > - my signature > At this point you will become aware that even my previous address > was a CGA. And you can verify that both addresses have come from > the same public key (even if the addresses are completely different). > And you can verify my signature. And no information had to be > communicated before the rehoming event occurred. Excellent, this alleviates the "negotiate in advance" issue. Still, I don't see the value in having the subnet prefix in the hash, and there's still the issue of address rewriting. Are there any attacks that can be fended off this way? If it's just a privacy thing I don't think that's enough justification, as it's already possible to generate a new hash based on an existing key by selecting a new "modifier" value. The only problem here is that this is more expensive when a "Sec" is in effect. But that's pretty much a feature rather than a downside. BTW, I'm no statistician and I don't play one on tv, but isn't there a significant chance that you're going to run into statistical anomalies when searching for a SHA-1 hash with a certain number of 0 bits in it? I mean, the chance of getting a SHA-1 with all 0 bits in the lowest 16 bits would be 1 in 65536. But that doesn't automatically mean that if you try 65536 times you hit one. Has anyone done the math on this? -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 05:03:59 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA18260 for ; Mon, 10 May 2004 05:03:59 -0400 (EDT) Received: from esealmw140.al.sw.ericsson.se ([153.88.254.121]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4A940Ah009288 for ; Mon, 10 May 2004 11:04:00 +0200 Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120]) by esealmw140.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 11:04:00 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KT7B7JKC; Mon, 10 May 2004 11:04:01 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4A93xXA020753; Mon, 10 May 2004 11:03:59 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A93kIt015471; Mon, 10 May 2004 11:03:46 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4A93ksa015469; Mon, 10 May 2004 11:03:46 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from ftmailgfi.HQ.Flarion.com (mail.flarion.com [63.103.94.23]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A93iIt015461 for ; Mon, 10 May 2004 11:03:45 +0200 (MET DST) Received: from ftmail2000.HQ.Flarion.com ([10.10.1.120]) by ftmailgfi.HQ.Flarion.com with Microsoft SMTPSVC(5.0.2195.6713); Mon, 10 May 2004 05:03:43 -0400 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Content-Class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Subject: RE: CGA issues Date: Mon, 10 May 2004 05:03:42 -0400 Message-ID: Thread-Topic: CGA issues Thread-Index: AcQ2bMEc25Zp+GcrTzu8zwk0ygvavAAAP5bw From: "Soliman Hesham" To: "Iljitsch van Beijnum" , "Jari Arkko" CC: , "Tuomas Aura" X-OriginalArrivalTime: 10 May 2004 09:03:43.0046 (UTC) FILETIME=[B17F4660:01C4366D] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by sw.ericsson.se id i4A93jIt015466 Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk Content-Transfer-Encoding: 8bit > Still, I don't see the value in having the subnet prefix in > the hash, > and there's still the issue of address rewriting. > > Are there any attacks that can be fended off this way? If > it's just a > privacy thing I don't think that's enough justification, as it's > already possible to generate a new hash based on an existing key by > selecting a new "modifier" value. The only problem here is > that this is > more expensive when a "Sec" is in effect. But that's pretty much a > feature rather than a downside. => I don't see the value of adding the prefix for SEND. However, if the intention is to define CGAs for general purpose security wherever address ownership is an issue then I do see a value of adding the prefix. This was discussed in the context of using CGAs to secure MIPv6 signalling. Of course there is a danger in having two different ways of generating a hash (one with the prefix and one without). A node using CGAs for SEND might want to use the same address as a MIPv6 CoA (most likely of course). There is no point in having one address for ND security and another for MIPv6 and another for multi6 ;) So we might need to somehow find the lowest common denominator here in order to use CGAs in a generic manner. Hesham > > BTW, I'm no statistician and I don't play one on tv, but > isn't there a > significant chance that you're going to run into statistical > anomalies > when searching for a SHA-1 hash with a certain number of 0 > bits in it? > I mean, the chance of getting a SHA-1 with all 0 bits in the > lowest 16 > bits would be 1 in 65536. But that doesn't automatically > mean that if > you try 65536 times you hit one. Has anyone done the math on this? > > -------------------------------------------------------------------- > To unsubscribe from this list, send email with "UNSUBSCRIBE" in the > body to . > Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html > -------------------------------------------------------------------- > ======================================================== This email may contain confidential and privileged material for the sole use of the intended recipient. Any review or distribution by others is strictly prohibited. If you are not the intended recipient please contact the sender and delete all copies. ======================================================== -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 05:25:36 2004 Received: from penguin.ericsson.se (penguin.ericsson.se [193.180.251.47]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA19135 for ; Mon, 10 May 2004 05:25:35 -0400 (EDT) Received: from esealmw142.al.sw.ericsson.se ([153.88.254.119]) by penguin.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4A9PaPA006324 for ; Mon, 10 May 2004 11:25:37 +0200 (MEST) Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw142.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 11:25:36 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM931V7H; Mon, 10 May 2004 11:25:35 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4A9PYwg028856; Mon, 10 May 2004 11:25:34 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A9OiIt019716; Mon, 10 May 2004 11:24:44 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4A9Oi21019715; Mon, 10 May 2004 11:24:44 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from sequoia.muada.com (sequoia.muada.com [83.149.65.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4A9OhIt019711 for ; Mon, 10 May 2004 11:24:43 +0200 (MET DST) Received: from [127.0.0.1] (sequoia.muada.com [83.149.65.1]) by sequoia.muada.com (8.12.10/8.12.10) with ESMTP id i4A9OT1C083008; Mon, 10 May 2004 11:24:29 +0200 (CEST) (envelope-from iljitsch@muada.com) In-Reply-To: References: Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: Content-Transfer-Encoding: 7bit Cc: "" , Tuomas Aura From: Iljitsch van Beijnum Subject: Re: CGA issues Date: Mon, 10 May 2004 11:24:43 +0200 To: "Soliman Hesham" X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 09:25:36.0265 (UTC) FILETIME=[C03C9790:01C43670] Content-Transfer-Encoding: 7bit On 10-mei-04, at 11:03, Soliman Hesham wrote: > => I don't see the value of adding the prefix for SEND. > However, if the intention is to define CGAs for general > purpose security wherever address ownership is an issue > then I do see a value of adding the prefix. This was discussed > in the context of using CGAs to secure MIPv6 signalling. So how does this help MIP? > There is no point in having one address for ND security and another > for MIPv6 and another for multi6 ;) > So we might need to somehow find the lowest common denominator > here in order to use CGAs in a generic manner. Actually the objectives and requirements are slightly different so it's not completely without merit to consider having different mechanisms. But a unified one would be much, much, MUCH better. For instance, for multi6 it is useful to be able to look up an identifier value to learn the associated locator values. Obviously this is going to be very hard, if not impossible, in a flat 59 bit space. That's why I proposed doing something that's otherwise quite similar to what's in the CGA draft using a site key rather than a host key, so that the number of values in the flat identifier space is limited to a number that isn't very different from the number of delegations in the largest DNS zones that exist today. BTW, what's the implementation status of SEND/CGA? -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 06:37:11 2004 Received: from penguin.ericsson.se (penguin.ericsson.se [193.180.251.47]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA23110 for ; Mon, 10 May 2004 06:37:11 -0400 (EDT) Received: from esealmw142.al.sw.ericsson.se ([153.88.254.119]) by penguin.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4AAbCPA024596 for ; Mon, 10 May 2004 12:37:12 +0200 (MEST) Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw142.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 12:37:11 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM93FKRW; Mon, 10 May 2004 12:37:11 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4AAbAXA022829; Mon, 10 May 2004 12:37:10 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AAaHIt005777; Mon, 10 May 2004 12:36:17 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4AAaHFH005776; Mon, 10 May 2004 12:36:17 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fep19-app.kolumbus.fi (fep19-0.kolumbus.fi [193.229.0.45]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AAaGIt005772 for ; Mon, 10 May 2004 12:36:16 +0200 (MET DST) Received: from mta.imail.kolumbus.fi ([193.229.5.114]) by fep19-app.kolumbus.fi with ESMTP id <20040510103615.JOPT26758.fep19-app.kolumbus.fi@mta.imail.kolumbus.fi>; Mon, 10 May 2004 13:36:15 +0300 X-Mailer: Openwave WebEngine, version 2.8.10 (webedge20-101-191-20030113) From: To: CC: Subject: Re: CGA issues Date: Mon, 10 May 2004 13:36:15 +0300 MIME-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1 Content-Transfer-Encoding: 7bit Message-Id: <20040510103615.JOPT26758.fep19-app.kolumbus.fi@mta.imail.kolumbus.fi> Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 10:37:11.0648 (UTC) FILETIME=[C07C0600:01C4367A] Content-Transfer-Encoding: 7bit > Still, I don't see the value in having the subnet prefix > in the hash, and there's still the issue of address > rewriting. > > Are there any attacks that can be fended off this way? > If it's just a privacy thing I don't think that's enough > justification, as it's already possible to generate a new > hash based on an existing key by selecting a new > "modifier" value. The only problem here is that this is > more expensive when a "Sec" is in effect. But that's > pretty much a feature rather than a downside. There is a specific attack that the prefix counters. Basically, the problem is in the number of available bits for the IID, which is pretty large but quite as much as one would perhaps hope to have in the cryptographic sense. Tuomas' "Sec" bits defend against brute force attacks, and the inclusion of the prefix defends against precomputation. Basically, if the prefix is not included in the address genereration, the bad guys -- like some three letter organization -- could generate a table of all CGAs. I did the math about a couple of years ago, and it turned out that at that time, you would have needed about 10 WTCs worth of office space to house the tape machines that hold the table. But technology is advancing, so it might have become feasible at some point. Now, with the prefix included in the hash, you'd have to have this much data for _each prefix_. > BTW, I'm no statistician and I don't play one on tv, > but isn't there a significant chance that you're going > to run into statistical anomalies when searching for a > SHA-1 hash with a certain number of 0 bits in it? I > mean, the chance of getting a SHA-1 with all 0 bits > in the lowest 16 bits would be 1 in 65536. But that > doesn't automatically mean that if you try 65536 > times you hit one. Has anyone done the math on this? Another question for Tuomas... --Jari -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 06:46:13 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA23366 for ; Mon, 10 May 2004 06:46:13 -0400 (EDT) Received: from esealmw140.al.sw.ericsson.se ([153.88.254.121]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4AAkEWR008187 for ; Mon, 10 May 2004 12:46:15 +0200 (MEST) Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118]) by esealmw140.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 12:46:11 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt612.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM0DL1PH; Mon, 10 May 2004 12:46:00 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4AAjwXA023003; Mon, 10 May 2004 12:45:58 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AAjhIt008221; Mon, 10 May 2004 12:45:43 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4AAjhp1008220; Mon, 10 May 2004 12:45:43 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from ftmailgfi.HQ.Flarion.com (mail.flarion.com [63.103.94.23]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AAjfIt008182 for ; Mon, 10 May 2004 12:45:42 +0200 (MET DST) Received: from ftmail2000.HQ.Flarion.com ([10.10.1.120]) by ftmailgfi.HQ.Flarion.com with Microsoft SMTPSVC(5.0.2195.6713); Mon, 10 May 2004 06:45:39 -0400 X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1409 Content-Class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Subject: RE: CGA issues Date: Mon, 10 May 2004 06:45:39 -0400 Message-ID: Thread-Topic: CGA issues Thread-Index: AcQ2cKAD/Ds7dK9STT2xYMsNYAEbwQACUGrA From: "Soliman Hesham" To: "Iljitsch van Beijnum" CC: , "Tuomas Aura" X-OriginalArrivalTime: 10 May 2004 10:45:39.0671 (UTC) FILETIME=[EF4A2E70:01C4367B] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by sw.ericsson.se id i4AAjhIt008204 Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk Content-Transfer-Encoding: 8bit > > => I don't see the value of adding the prefix for SEND. > > However, if the intention is to define CGAs for general > > purpose security wherever address ownership is an issue > > then I do see a value of adding the prefix. This was discussed > > in the context of using CGAs to secure MIPv6 signalling. > > So how does this help MIP? => Sorry for confusing you with this, I had a solution in mind but after thinking about it a bit more I don't think it works. So forget this part. > > There is no point in having one address for ND security and another > > for MIPv6 and another for multi6 ;) > > So we might need to somehow find the lowest common denominator > > here in order to use CGAs in a generic manner. > > Actually the objectives and requirements are slightly > different so it's > not completely without merit to consider having different > mechanisms. > But a unified one would be much, much, MUCH better. => I haven't followed multi6 in detail but from a security point of view the problem seems very similar, if not the same. The question is how to fit it into the non-existent multi6 framework ;) > > For instance, for multi6 it is useful to be able to look up an > identifier value to learn the associated locator values. > Obviously this > is going to be very hard, if not impossible, in a flat 59 bit space. > That's why I proposed doing something that's otherwise quite > similar to > what's in the CGA draft using a site key rather than a host key, so > that the number of values in the flat identifier space is > limited to a > number that isn't very different from the number of > delegations in the > largest DNS zones that exist today. => ok. I suppose it all depends on the solution details for multi6. Without that it's difficult to know how well CGAs will fit as is. > > BTW, what's the implementation status of SEND/CGA? => No idea. Hesham > > ======================================================== This email may contain confidential and privileged material for the sole use of the intended recipient. Any review or distribution by others is strictly prohibited. If you are not the intended recipient please contact the sender and delete all copies. ======================================================== -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 13:54:30 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA19946 for ; Mon, 10 May 2004 13:54:29 -0400 (EDT) Received: from esealmw141.al.sw.ericsson.se ([153.88.254.120]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4AHsTWR011181 for ; Mon, 10 May 2004 19:54:29 +0200 (MEST) Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw141.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 19:54:29 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KNDH79FW; Mon, 10 May 2004 19:52:17 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4AHpowg025798; Mon, 10 May 2004 19:51:50 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AHooIt009088; Mon, 10 May 2004 19:50:50 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4AHooUZ009087; Mon, 10 May 2004 19:50:50 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fridge.docomolabs-usa.com (key1.docomolabs-usa.com [216.98.102.225]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AHomIt009083 for ; Mon, 10 May 2004 19:50:49 +0200 (MET DST) Message-ID: <01b801c436b7$6a8ee040$366115ac@dcml.docomolabsusa.com> From: "James Kempf" To: "Jari Arkko" , "Iljitsch van Beijnum" Cc: , "Tuomas Aura" References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> Subject: u/l and g bits in CGAs (was: Re: CGA issues) Date: Mon, 10 May 2004 10:51:25 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 17:54:29.0540 (UTC) FILETIME=[D77C9E40:01C436B7] Content-Transfer-Encoding: 7bit > > Finally, the draft doesn't specify the value of the universal/local and > > group bits in the cryptographically generated interface identifier. > > Obviously it's important that naive implementations don't adopt a link > > address that happens to have the group bit set, as will be the case in > > approximately 50% of all CGAs. Additionally, it is probably prudent to > > set the u/l bit to 1, as CGAs aren't "universal". > I believe we agreed in the WG that these bits aren't set. The group bit isn't set for the reason Iljtsch has mentioned. The universal bit isn't set because the interface identifier (as opposed to the address itself) isn't of global scope. Unlike EUI-64 addresses, CGA identifiers aren't handed out by a controlling authority that is trying to keep them universally unique. They might be expected to be statistically unique globally, but that is different from a specific intent to make them unique, which is what the WG interpreted RFC 3513 to mean. Section 4 Step 6 in draft-ietf-send-cga.txt specifies how the 'u' and 'g' bits are handled in CGAs: 6. Form an interface identifier from Hash1 by writing the value of Sec into the three leftmost bits and by setting bits 6 and 7 (i.e., the "u" and "g" bits) both to zero. jak -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 14:17:45 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA21444 for ; Mon, 10 May 2004 14:17:45 -0400 (EDT) Received: from esealmw140.al.sw.ericsson.se ([153.88.254.121]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4AIHjAh025611 for ; Mon, 10 May 2004 20:17:45 +0200 Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw140.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 20:17:45 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM9320GP; Mon, 10 May 2004 20:17:45 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4AIHiXA001862; Mon, 10 May 2004 20:17:44 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AIGrIt015404; Mon, 10 May 2004 20:16:53 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4AIGrGg015403; Mon, 10 May 2004 20:16:53 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fridge.docomolabs-usa.com (key1.docomolabs-usa.com [216.98.102.225]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AIGpIt015399 for ; Mon, 10 May 2004 20:16:51 +0200 (MET DST) Message-ID: <01eb01c436bb$0e382000$366115ac@dcml.docomolabsusa.com> From: "James Kempf" To: "Iljitsch van Beijnum" , "Soliman Hesham" Cc: , "Tuomas Aura" References: Subject: Re: CGA issues Date: Mon, 10 May 2004 11:17:28 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 18:17:45.0731 (UTC) FILETIME=[17AE7530:01C436BB] Content-Transfer-Encoding: 7bit > BTW, what's the implementation status of SEND/CGA? > DoCoMo Labs did a full implementation of ndopt-00 on Linux. It isn't publically available however. Ive heard from at least one platform vendor who is interested in implementing it. On the other hand, I've also heard from one influential member of the KAME community that they will not incorporate CGA into KAME because it has IPR on it. jak -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 14:44:36 2004 Received: from penguin.ericsson.se (penguin.ericsson.se [193.180.251.47]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA23686 for ; Mon, 10 May 2004 14:44:36 -0400 (EDT) Received: from esealmw140.al.sw.ericsson.se ([153.88.254.121]) by penguin.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4AIiaPA028551 for ; Mon, 10 May 2004 20:44:36 +0200 (MEST) Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw140.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 20:44:35 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM93JDZ8; Mon, 10 May 2004 20:44:35 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4AIiWwg028313; Mon, 10 May 2004 20:44:32 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AIhtIt020344; Mon, 10 May 2004 20:43:55 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4AIhtJ5020343; Mon, 10 May 2004 20:43:55 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fridge.docomolabs-usa.com (key1.docomolabs-usa.com [216.98.102.225]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AIhqIt020339 for ; Mon, 10 May 2004 20:43:53 +0200 (MET DST) Message-ID: <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> From: "James Kempf" To: "Iljitsch van Beijnum" , "Jari Arkko" Cc: , "Tuomas Aura" References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> Subject: CGA for multihoming (was: Re: CGA issues) Date: Mon, 10 May 2004 11:44:30 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 18:44:35.0491 (UTC) FILETIME=[D72C5730:01C436BE] Content-Transfer-Encoding: 7bit Iljitsch, > > Are you trying to see how existing multi6 > > solutions would work together with SEND, or are you trying > > to see whether CGA could be used as a multi6 solution? > > CGA on it's own can't be a multihoming solution. I think it would be > useful to reuse CGA as a security mechanism in a multihoming solution, > but mostly I'm concerned with making SEND and multi6 work together > without getting in each other's way. > First off, the CG properties of addresses are not expected to be checked by hosts off of the owning host's last hop link. That is, CGAs are not expected to be used for any purpose other than SEND, which is only concerned with the last hop link. Therefore, if a router somewhere in the path is rewriting the 48 bit routing domain specific prefix, this change should not affect packets sent between hosts on the local link, unless the rewriting is being done for packets on the local link (i.e. not being forwarded) by the last hop router, which seems unlikely to me. Secondly, from a practical standpoint, CGAs are heavily IPR-encumbered. The SEND WG was able to obtain a nonRAND IPR release from the IPR owners for the purpose of SEND *only*. If the multi6 working group is interested in using CGAs for multi6, it would need to ask the IPR holders about what their licensing terms are for multi6. That process is likely to take a long time, and may run into ideological opposition from the Open Source community and others who believe that IPR is bad in principle, or even people who don't object to IPR in principle if it is nonRAND but would if RAND licensing terms are required for multi6 because the multi6 WG doesn't manage to obtain a nonRAND release (this was one of the major issues in the SEND WG). It is truely unfortunate that CGAs are so heavily encumbered, since they are really useful in a wide variety of situations and one would just like to use them where they work, but that is reality. As Jari mentioned, the prefix is included to make guessing attacks more difficult, and Tuomas can comment on the guessing probability with 0's in specific bits. The WG has thought long about how to make CGAs work for SEND and we have something that we think should work for SEND. Rather than delay SEND, I would like to take the current draft to PS as is. If the multi6 WG works through the IPR issue with CGAs and comes up with a different way to prevent guessing attacks that doesn't contain the prefix, then at that time, the draft can be revised. But, for now, delaying the draft for another 6 months to a year in order to make CGAs work for other purposes than SEND, when no WG has specific plans to use them nor has worked through the IPR issues, doesn't seem to make sense to me. Anyone else in the WG want to comment? jak -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 16:03:40 2004 Received: from penguin.ericsson.se (penguin.ericsson.se [193.180.251.47]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00128 for ; Mon, 10 May 2004 16:03:39 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by penguin.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4AK3ePA005984 for ; Mon, 10 May 2004 22:03:40 +0200 (MEST) Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 22:03:40 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KNDH8T74; Mon, 10 May 2004 22:03:52 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4AK3dXA003084; Mon, 10 May 2004 22:03:39 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AK2wIt007931; Mon, 10 May 2004 22:02:58 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4AK2wM2007930; Mon, 10 May 2004 22:02:58 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fep01-app.kolumbus.fi (fep01-0.kolumbus.fi [193.229.0.41]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AK2vIt007926 for ; Mon, 10 May 2004 22:02:57 +0200 (MET DST) Received: from kolumbus.fi ([80.186.217.250]) by fep01-app.kolumbus.fi with ESMTP id <20040510200254.ZQMV4346.fep01-app.kolumbus.fi@kolumbus.fi>; Mon, 10 May 2004 23:02:54 +0300 Message-ID: <409FDF21.1070400@kolumbus.fi> Date: Mon, 10 May 2004 22:59:29 +0300 From: Jari Arkko User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7b) Gecko/20040316 X-Accept-Language: en-us, en MIME-Version: 1.0 To: James Kempf CC: Iljitsch van Beijnum , ietf-send@standards.ericsson.net, Tuomas Aura Subject: Re: CGA for multihoming References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> In-Reply-To: <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 20:03:40.0473 (UTC) FILETIME=[E3672290:01C436C9] Content-Transfer-Encoding: 7bit James Kempf wrote: > Anyone else in the WG want to comment? I agree that SEND should not wait for Multi6 design, nor should SEND drafts pay much attention to the use of CGA technology for other purposes than SEND. I'm not sure Iljitsch was proposing that we do that though. By the way, I'm pretty sure that we can't improve the current solution for preventing precomputation. It may be an interesting topic to figure out if CGAs can be made to work for Multi6, however. And certainly any Multi6 solutions should be designed so that they can live with other IPv6 features like EUI-64 IIDs, RFC 3041, or SEND. Or at the very least the IETF should be aware of the limitations of proposed Multi6 solutions, if we can't make all features work together. Regarding IPRs, I have a nit: when you James say "non-RAND", I think someone might mistake that for non-reasonable and discriminatory licensing ;-). I think you meant that its better than RAND. If someone wants to look at the license conditions, they are below. I'm not a lawyer but I think they say roalty free unless you demand money for your own patents on the same thing. Or something like that. Too bad that's not good enough for KAME, maybe it has something to do with allowing the use of their code for any purpose and not just SEND? Or the requirements of their copyleft? Or just ideological opposition, even when the use would be allowed? It would be interesting to know if there have been cases like this in the past with PD, and what happened in them. http://www.ietf.org/ietf/IPR/ericsson-send-ipsec.txt http://www.ietf.org/ietf/IPR/microsoft-ipr-draft-ietf-send-ipsec.txt --Jari -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 16:06:58 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00327 for ; Mon, 10 May 2004 16:06:57 -0400 (EDT) Received: from esealmw140.al.sw.ericsson.se ([153.88.254.121]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4AK6wWR026917 for ; Mon, 10 May 2004 22:06:58 +0200 (MEST) Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw140.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 22:06:58 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM93JQYP; Mon, 10 May 2004 22:06:58 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4AK6qwg002216; Mon, 10 May 2004 22:06:52 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AK6hIt008463; Mon, 10 May 2004 22:06:43 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4AK6hAR008462; Mon, 10 May 2004 22:06:43 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from sequoia.muada.com (sequoia.muada.com [83.149.65.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AK6gIt008446 for ; Mon, 10 May 2004 22:06:42 +0200 (MET DST) Received: from [127.0.0.1] (sequoia.muada.com [83.149.65.1]) by sequoia.muada.com (8.12.10/8.12.10) with ESMTP id i4AK6Q1C093572; Mon, 10 May 2004 22:06:26 +0200 (CEST) (envelope-from iljitsch@muada.com) In-Reply-To: <01b801c436b7$6a8ee040$366115ac@dcml.docomolabsusa.com> References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <01b801c436b7$6a8ee040$366115ac@dcml.docomolabsusa.com> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <8A5283B9-A2BD-11D8-B6DD-000A95CD987A@muada.com> Content-Transfer-Encoding: 7bit Cc: "" , Tuomas Aura From: Iljitsch van Beijnum Subject: Re: u/l and g bits in CGAs (was: Re: CGA issues) Date: Mon, 10 May 2004 22:06:36 +0200 To: "James Kempf" X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 20:06:58.0226 (UTC) FILETIME=[5945D920:01C436CA] Content-Transfer-Encoding: 7bit On 10-mei-04, at 19:51, James Kempf wrote: > I believe we agreed in the WG that these bits aren't set. The group bit > isn't set for the reason Iljtsch has mentioned. Actually I don't think it's necessary to specifically clear the bit per se, as long as there is a good strong warning that the generated interface identifier may not be used as a link address. But currently the interface identifier namespace is used up for all instances where either the u/l bit is set to local, or the u/l bit is set to universal and the group bit is cleared. This leaves the instances where the u/l bit is universal and the group bit is set free for future use, which is probably a good thing. > The universal bit isn't set > because the interface identifier (as opposed to the address itself) > isn't of > global scope. Unlike EUI-64 addresses, CGA identifiers aren't handed > out by > a controlling authority that is trying to keep them universally > unique. They > might be expected to be statistically unique globally, but that is > different > from a specific intent to make them unique, which is what the WG > interpreted > RFC 3513 to mean. Agree. > Section 4 Step 6 in draft-ietf-send-cga.txt specifies how the 'u' and > 'g' > bits are handled in CGAs: > 6. Form an interface identifier from Hash1 by writing the value of > Sec into the three leftmost bits and by setting bits 6 and 7 > (i.e., the "u" and "g" bits) both to zero. I think it's important to mention that this procedure results in an interface identifier without the intermediate steps mentioned in RFC 2464 (= flipping the u/l bit in the MAC address which has the opposite meaning in the IEEE OUI). -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 16:14:01 2004 Received: from penguin.ericsson.se (penguin.ericsson.se [193.180.251.47]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA00589 for ; Mon, 10 May 2004 16:14:01 -0400 (EDT) Received: from esealmw141.al.sw.ericsson.se ([153.88.254.120]) by penguin.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4AKE2PA007329 for ; Mon, 10 May 2004 22:14:02 +0200 (MEST) Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120]) by esealmw141.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 22:14:01 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KT7CBFWT; Mon, 10 May 2004 22:14:04 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4AKDwwg002568; Mon, 10 May 2004 22:13:58 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AKDNIt009513; Mon, 10 May 2004 22:13:23 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4AKDNot009512; Mon, 10 May 2004 22:13:23 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fridge.docomolabs-usa.com (key1.docomolabs-usa.com [216.98.102.225]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AKDLIt009508 for ; Mon, 10 May 2004 22:13:22 +0200 (MET DST) Message-ID: <025e01c436cb$54cfc4e0$366115ac@dcml.docomolabsusa.com> From: "James Kempf" To: "Iljitsch van Beijnum" Cc: , "Tuomas Aura" References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <01b801c436b7$6a8ee040$366115ac@dcml.docomolabsusa.com> <8A5283B9-A2BD-11D8-B6DD-000A95CD987A@muada.com> Subject: Re: u/l and g bits in CGAs (was: Re: CGA issues) Date: Mon, 10 May 2004 13:13:58 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 20:14:01.0453 (UTC) FILETIME=[558929D0:01C436CB] Content-Transfer-Encoding: 7bit > I think it's important to mention that this procedure results in an > interface identifier without the intermediate steps mentioned in RFC > 2464 (= flipping the u/l bit in the MAC address which has the opposite > meaning in the IEEE OUI). > > If you have any specific proposals for text and where to put that text into the draft, I believe we could accommodate a last minute change to clarify this issue. jak -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 16:27:45 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01486 for ; Mon, 10 May 2004 16:27:45 -0400 (EDT) Received: from esealmw142.al.sw.ericsson.se ([153.88.254.119]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4AKRkAh004144 for ; Mon, 10 May 2004 22:27:47 +0200 Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw142.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 22:27:46 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KNDH8YQY; Mon, 10 May 2004 22:27:59 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4AKRjXA003489; Mon, 10 May 2004 22:27:45 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AKRKIt012457; Mon, 10 May 2004 22:27:20 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4AKRKSN012456; Mon, 10 May 2004 22:27:20 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fridge.docomolabs-usa.com (key1.docomolabs-usa.com [216.98.102.225]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AKRIIt012449 for ; Mon, 10 May 2004 22:27:18 +0200 (MET DST) Message-ID: <026401c436cd$47943f70$366115ac@dcml.docomolabsusa.com> From: "James Kempf" To: "Jari Arkko" Cc: "Iljitsch van Beijnum" , , "Tuomas Aura" References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> <409FDF21.1070400@kolumbus.fi> Subject: Re: CGA for multihoming Date: Mon, 10 May 2004 13:27:55 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 20:27:46.0736 (UTC) FILETIME=[41715B00:01C436CD] Content-Transfer-Encoding: 7bit Jari, > Regarding IPRs, I have a nit: when you James say "non-RAND", > I think someone might mistake that for non-reasonable and > discriminatory licensing ;-). I think you meant that its > better than RAND. If someone wants to look at the license > conditions, they are below. I'm not a lawyer but I think > they say roalty free unless you demand money for your own > patents on the same thing. Or something like that. Too > bad that's not good enough for KAME, maybe it has something > to do with allowing the use of their code for any purpose > and not just SEND? Or the requirements of their copyleft? > Or just ideological opposition, even when the use would be > allowed? It would be interesting to know if there have > been cases like this in the past with PD, and what happened > in them. > > http://www.ietf.org/ietf/IPR/ericsson-send-ipsec.txt > http://www.ietf.org/ietf/IPR/microsoft-ipr-draft-ietf-send-ipsec.txt > Yes, sorry, I meant "better than RAND". W.r.t. KAME, I suspect that the opposition is ideological. The WG chairs managed to secure the release from both IPR holders, and went out of their way to obtain a statement from one IPR holder about their requirement for notification for implementors v.s. no requirement for users in response to a complaint from the KAME community, but it was insufficient, due to the possibility that notification might be required for derivative works (even though the release places no requirement on derivative works, just on implementors). W.r.t. past cases, the case I'm most familiar with is IBM's release of the IPR for Photuris (RFC 1822) in which IBM granted IETF the right to use a patent in IPSec/IKE. I'm not familar with exactly what the IPR was that was released, but it seems as if it hasn't stopped the Open Source community from implementing IKE and IPsec. The ideological oppositions seems to have arisen recently in response to the increased use of IPR in IETF and the increased visibility of Open Source. jak -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 16:29:08 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA01608 for ; Mon, 10 May 2004 16:29:07 -0400 (EDT) Received: from esealmw140.al.sw.ericsson.se ([153.88.254.121]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4AKT7WR029786 for ; Mon, 10 May 2004 22:29:08 +0200 (MEST) Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw140.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Mon, 10 May 2004 22:29:05 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM93JT8K; Mon, 10 May 2004 22:29:04 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4AKT3XA003522; Mon, 10 May 2004 22:29:03 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AKStIt012645; Mon, 10 May 2004 22:28:55 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4AKStsD012644; Mon, 10 May 2004 22:28:55 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from sequoia.muada.com (sequoia.muada.com [83.149.65.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4AKSsIt012640 for ; Mon, 10 May 2004 22:28:54 +0200 (MET DST) Received: from [127.0.0.1] (sequoia.muada.com [83.149.65.1]) by sequoia.muada.com (8.12.10/8.12.10) with ESMTP id i4AKSc1C093998; Mon, 10 May 2004 22:28:38 +0200 (CEST) (envelope-from iljitsch@muada.com) In-Reply-To: <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: Content-Transfer-Encoding: 7bit Cc: "" , Tuomas Aura From: Iljitsch van Beijnum Subject: Re: CGA for multihoming (was: Re: CGA issues) Date: Mon, 10 May 2004 22:28:48 +0200 To: "James Kempf" X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 10 May 2004 20:29:05.0047 (UTC) FILETIME=[701EAA70:01C436CD] Content-Transfer-Encoding: 7bit On 10-mei-04, at 20:44, James Kempf wrote: > First off, the CG properties of addresses are not expected to be > checked by > hosts off of the owning host's last hop link. In multi6 we're pretty much changing everything, details like this won't stop us. :-) > That is, CGAs are not > expected to be used for any purpose other than SEND, which is only > concerned > with the last hop link. Therefore, if a router somewhere in the path is > rewriting the 48 bit routing domain specific prefix, this change > should not > affect packets sent between hosts on the local link, unless the > rewriting is > being done for packets on the local link (i.e. not being forwarded) by > the > last hop router, which seems unlikely to me. No, it doesn't work like that. Suppose a site is multihomed, and two prefixes (one from ISP X, one from ISP Y) are available on a subnet. So host A with key K creates two addresses for itself: X.CGA(K+X) and Y.CGA(K+Y). (Simplifying, of course). Now the A starts to communicate with B. It chooses the X address for this and all is well. But after a while there the link to X fails, and now the router reroutes the packets over Y. In order to avoid problems with ingress filtering by Y, the router rewrites all addresses with the X prefix into addresses with an Y prefix. So B now gets packets from Y.CGA(K+X). The problem with this is that Y.CGA(K+X) isn't a valid SEND address so A is unable to receive packets for this address. > Secondly, from a practical standpoint, CGAs are heavily IPR-encumbered. Do you have a pointer towards more information about this? > It is > truely unfortunate that CGAs are so heavily encumbered, since they are > really useful in a wide variety of situations and one would just like > to use > them where they work, but that is reality. Is it? I find it ridiculous that someone can claim ownership of completely obvious ideas. (If someone spends a lot of time and money on developing something specific, that's different. (And I find it surprising that in these discussions the actual subject matter of the patent never comes up.)) Hasn't anyone looked into challenging this? > As Jari mentioned, the prefix is included to make guessing attacks more > difficult, Is this important enough to risk problems with multihoming? And couldn't the same results be reached by allowing all prefixes that are on-link, rather than just one? > The WG has thought long about how to make CGAs work for SEND > and we have something that we think should work for SEND. Rather than > delay > SEND, I would like to take the current draft to PS as is. The IETF has spent much more time on stuff that turned out to be problem-riddled. I appreciate your concern about delays, but a slight delay isn't going to matter in the long run, while any problems with the implementation are going to be around forever. It's important to get things right the first time if at all possible. Note that I'm only talking about the interaction between SEND and multi6 here, NOT about re-using CGA for multihoming. -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 10 21:44:16 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA21667 for ; Mon, 10 May 2004 21:44:16 -0400 (EDT) Received: from esealmw140.al.sw.ericsson.se ([153.88.254.121]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4B1iGWR029102 for ; Tue, 11 May 2004 03:44:17 +0200 (MEST) Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw140.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 03:44:16 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM93LLAJ; Tue, 11 May 2004 03:44:16 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4B1iFXA007025; Tue, 11 May 2004 03:44:15 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B1gwIt021246; Tue, 11 May 2004 03:42:58 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4B1gwtP021245; Tue, 11 May 2004 03:42:58 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fridge.docomolabs-usa.com (key1.docomolabs-usa.com [216.98.102.225]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B1guIt021239 for ; Tue, 11 May 2004 03:42:56 +0200 (MET DST) Message-ID: <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> From: "James Kempf" To: "Iljitsch van Beijnum" Cc: , "Tuomas Aura" References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> Subject: Re: CGA for multihoming (was: Re: CGA issues) Date: Mon, 10 May 2004 18:43:32 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 11 May 2004 01:44:16.0543 (UTC) FILETIME=[783FF2F0:01C436F9] Content-Transfer-Encoding: 7bit > No, it doesn't work like that. Suppose a site is multihomed, and two > prefixes (one from ISP X, one from ISP Y) are available on a subnet. So > host A with key K creates two addresses for itself: X.CGA(K+X) and > Y.CGA(K+Y). (Simplifying, of course). Now the A starts to communicate > with B. It chooses the X address for this and all is well. But after a > while there the link to X fails, and now the router reroutes the > packets over Y. In order to avoid problems with ingress filtering by Y, > the router rewrites all addresses with the X prefix into addresses with Which router? The border router or the last hop (access) router or some other router? > an Y prefix. So B now gets packets from Y.CGA(K+X). The problem with > this is that Y.CGA(K+X) isn't a valid SEND address so A is unable to > receive packets for this address. > > > Secondly, from a practical standpoint, CGAs are heavily IPR-encumbered. > > Do you have a pointer towards more information about this? > Jari sent out links to the IPR release notices. If you have access to any patent database software, you can additionally search for the patent applications, as it is more than a year since they have been filed. > > As Jari mentioned, the prefix is included to make guessing attacks more > > difficult, > > Is this important enough to risk problems with multihoming? > Yes, the WG though it serious enough (note to WG members: if anyone in the WG thinks it isn't, please speak up now). As for problems with multihoming, as I mentioned, if there are problems, then the RFC can be revised. RFCs are revised all the time. Please correct me if I am wrong, but as far as I know, multihoming is nowhere near being finalized to RFC status. Or is that not so? Is the multi6 WG, as is the case with SEND, about to finalize this proposal you've presented for address rewriting and publish it as a Proposed Standard RFC? > And couldn't the same results be reached by allowing all prefixes that > are on-link, rather than just one? > I'm not sure I understand. Could you clarify? > Note that I'm only talking about the interaction between SEND and > multi6 here, NOT about re-using CGA for multihoming. > > As I hope my questions should indicate, perhaps I still don't understand the particular multi6 scheme you're proposing to see the problem. I've studied NOID in some minor amount of depth, but not this address rewriting proposal. jak -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Tue May 11 00:48:54 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA01223 for ; Tue, 11 May 2004 00:48:53 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4B4muAh004987 for ; Tue, 11 May 2004 06:48:56 +0200 Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 06:48:56 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KND2APTF; Tue, 11 May 2004 06:49:09 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4B4mjwg018896; Tue, 11 May 2004 06:48:45 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B4llIt000689; Tue, 11 May 2004 06:47:47 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4B4lle6000688; Tue, 11 May 2004 06:47:47 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fep01-app.kolumbus.fi (fep01-0.kolumbus.fi [193.229.0.41]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B4lkIt000682 for ; Tue, 11 May 2004 06:47:46 +0200 (MET DST) Received: from kolumbus.fi ([80.186.217.250]) by fep01-app.kolumbus.fi with ESMTP id <20040511044745.GAHG4346.fep01-app.kolumbus.fi@kolumbus.fi>; Tue, 11 May 2004 07:47:45 +0300 Message-ID: <40A05A23.1050804@kolumbus.fi> Date: Tue, 11 May 2004 07:44:19 +0300 From: Jari Arkko User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7b) Gecko/20040316 X-Accept-Language: en-us, en MIME-Version: 1.0 To: James Kempf CC: Iljitsch van Beijnum , ietf-send@standards.ericsson.net, Tuomas Aura Subject: Re: CGA for multihoming References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> In-Reply-To: <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 11 May 2004 04:48:56.0018 (UTC) FILETIME=[4421AB20:01C43713] Content-Transfer-Encoding: 7bit James Kempf wrote: >>And couldn't the same results be reached by allowing all prefixes that >>are on-link, rather than just one? > > I'm not sure I understand. Could you clarify? I think Iljitsch is thinking that instead of one prefix as an input to the CGA generation, you would use all the prefixes on the link. This would work, I think, even if there's a lot of details to worry about when, e.g., one of the prefixes is deprecated. The Extension Fields part of the CGA parameters structure (Sect 3 in draft-ietf-send-cga-06.txt) can be used to add other inputs to the CGA generation. This would allow all prefixes to be included in the hash. It still doesn't make the IID part of the addresses be the same across different prefixes; the selected prefix has a special place in the hash too. Revving the CGA RFC would be needed for that. Even so, the address would presumably change when the set of prefixes changes, so I'm not sure we would end up with a practical solution this way. --Jari -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Tue May 11 03:31:49 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA08405 for ; Tue, 11 May 2004 03:31:49 -0400 (EDT) Received: from esealmw141.al.sw.ericsson.se ([153.88.254.120]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4B7VbWR028183 for ; Tue, 11 May 2004 09:31:38 +0200 (MEST) Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118]) by esealmw141.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 09:31:36 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt612.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM0DTHW2; Tue, 11 May 2004 09:31:36 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4B7VZXA010933; Tue, 11 May 2004 09:31:35 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B7UXIt007254; Tue, 11 May 2004 09:30:33 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4B7UWBi007249; Tue, 11 May 2004 09:30:32 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from mail4.microsoft.com (mail4.microsoft.com [131.107.3.122]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B7UUIt007181 for ; Tue, 11 May 2004 09:30:31 +0200 (MET DST) Received: from inet-vrs-04.redmond.corp.microsoft.com ([157.54.8.149]) by mail4.microsoft.com with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 00:30:29 -0700 Received: from 157.54.6.150 by inet-vrs-04.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 11 May 2004 00:30:29 -0700 Received: from RED-MSG-43.redmond.corp.microsoft.com ([157.54.12.203]) by inet-hub-05.redmond.corp.microsoft.com with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 00:30:29 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: CGA issues Date: Tue, 11 May 2004 00:29:32 -0700 Message-ID: <64A531765B7C8342BFA260497BE004570226D767@RED-MSG-43.redmond.corp.microsoft.com> Thread-Topic: CGA issues thread-index: AcQ13vGz9tJTvl8aTlyLfL8Yu9vyxwBR/v5A From: "Tuomas Aura" To: Cc: "Jari Arkko" , "Iljitsch van Beijnum" X-OriginalArrivalTime: 11 May 2004 07:30:29.0803 (UTC) FILETIME=[D613DFB0:01C43729] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by sw.ericsson.se id i4B7UVIt007235 Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk Content-Transfer-Encoding: 8bit Jari Arkko quoted Iljitsch van Beijnum: > > Finally, the draft doesn't specify the value of the universal/local and > > group bits in the cryptographically generated interface identifier. > > Obviously it's important that naive implementations don't adopt a link > > address that happens to have the group bit set, as will be the case in > > approximately 50% of all CGAs. Additionally, it is probably prudent to > > set the u/l bit to 1, as CGAs aren't "universal". > > Tuomas, can you say something about this? See step 6 of the GCA generation algorithm in section 4 of draft-ietf-send-cga. The "g" bit *is* set to 0. The "u" bit is set to 0 to indicate local scope. (1 would indicate universal scope.) The CGA verification algorithm in section 5 of the same draft ignores the values of the "u" and "g" bits. This is is an intentional design choice that was discussed on the list. Checking the bit values during the verification would not buy any additional security but it would close some future design options. Thus, it is better not to check them. Tuomas -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Tue May 11 05:44:46 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA15073 for ; Tue, 11 May 2004 05:44:45 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4B9ilAh020820 for ; Tue, 11 May 2004 11:44:47 +0200 Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 11:44:47 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KWS1HCB8; Tue, 11 May 2004 11:44:47 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4B9ibwg001554; Tue, 11 May 2004 11:44:37 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B9gAIt004988; Tue, 11 May 2004 11:42:10 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4B9gAv3004987; Tue, 11 May 2004 11:42:10 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from sequoia.muada.com (sequoia.muada.com [83.149.65.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B9g9It004983 for ; Tue, 11 May 2004 11:42:09 +0200 (MET DST) Received: from [127.0.0.1] (sequoia.muada.com [83.149.65.1]) by sequoia.muada.com (8.12.10/8.12.10) with ESMTP id i4B9fs1C007761; Tue, 11 May 2004 11:41:54 +0200 (CEST) (envelope-from iljitsch@muada.com) In-Reply-To: <64A531765B7C8342BFA260497BE004570226D76A@RED-MSG-43.redmond.corp.microsoft.com> References: <64A531765B7C8342BFA260497BE004570226D76A@RED-MSG-43.redmond.corp.microsoft.com> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <76821F28-A32F-11D8-B6DD-000A95CD987A@muada.com> Content-Transfer-Encoding: 7bit Cc: , "Jari Arkko" From: Iljitsch van Beijnum Subject: Re: CGA issues Date: Tue, 11 May 2004 11:42:05 +0200 To: "Tuomas Aura" X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 11 May 2004 09:44:47.0105 (UTC) FILETIME=[989A9710:01C4373C] Content-Transfer-Encoding: 7bit On 11-mei-04, at 9:50, Tuomas Aura wrote: > Supporting privacy by default is certainly not a downside. Instead, > I would think that any protocol that prevents the use of > privacy-friendly changing addresses (CGAs or RFC-3041 addresses) > is seriously flawed. I wasn't saying there is no value in supporting privacy. > I don't understand your argument about changing the modifier vs. > including the prefix in the hash. Both cause the IID to change. > Changing the modifier is potentially much more expensive. Thus, > including the prefix makes much more sense. If all else is the same, yes. But this choice can either make life for multi6 a lot harder or make it impossible to deploy SEND in certain networks in the future. >> I mean, the chance of getting a SHA-1 with all 0 bits in the lowest 16 >> bits would be 1 in 65536. But that doesn't automatically mean that if >> you try 65536 times you hit one. Has anyone done the math on this? > The median number of iterations is 0.69 * 2^(16*Sec). So where is the 99th percentile? Or the 99.99th percentile? -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Tue May 11 05:53:20 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA15504 for ; Tue, 11 May 2004 05:53:20 -0400 (EDT) Received: from esealmw142.al.sw.ericsson.se ([153.88.254.119]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4B9rLWR012048 for ; Tue, 11 May 2004 11:53:21 +0200 (MEST) Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120]) by esealmw142.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 11:53:21 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KWS1H12A; Tue, 11 May 2004 11:53:21 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4B9rKwg001976; Tue, 11 May 2004 11:53:20 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B9qxIt007563; Tue, 11 May 2004 11:52:59 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4B9qxfA007562; Tue, 11 May 2004 11:52:59 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from sequoia.muada.com (sequoia.muada.com [83.149.65.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B9qwIt007558 for ; Tue, 11 May 2004 11:52:58 +0200 (MET DST) Received: from [127.0.0.1] (sequoia.muada.com [83.149.65.1]) by sequoia.muada.com (8.12.10/8.12.10) with ESMTP id i4B9qk1C007960; Tue, 11 May 2004 11:52:46 +0200 (CEST) (envelope-from iljitsch@muada.com) In-Reply-To: <40A05A23.1050804@kolumbus.fi> References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> <40A05A23.1050804@kolumbus.fi> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: Content-Transfer-Encoding: 7bit Cc: ietf-send@standards.ericsson.net From: Iljitsch van Beijnum Subject: Re: CGA for multihoming Date: Tue, 11 May 2004 11:52:57 +0200 To: Jari Arkko X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 11 May 2004 09:53:21.0420 (UTC) FILETIME=[CB28D4C0:01C4373D] Content-Transfer-Encoding: 7bit On 11-mei-04, at 6:44, Jari Arkko wrote: >>> And couldn't the same results be reached by allowing all prefixes >>> that >>> are on-link, rather than just one? >> I'm not sure I understand. Could you clarify? > I think Iljitsch is thinking that instead of one prefix as an > input to the CGA generation, you would use all the prefixes > on the link. No. What I mean is that you still create the CGA the same way, but when checking it on another system, this system doesn't require the prefix to be the actual prefix used in the address, but also allows the prefix to be any other prefix that is on-link for this subnet. So if you have a link with prefixes 2001:0:0:1::/64 and 3ffe:0:0:1:/64, and a host generates a CGA that is 2001:0:0:1:0:dead:0:beaf, then it can also use the address 3ffe:0:0:1:0:dead:0:beaf. The current rules don't allow this as the prefix used in generating the CGA is 2001:0:0:1::/64, but the modified rule does allow this, as the prefix used in the CGA generation is on-link. This fixes the problem in instances where a host has one interface with multiple addresses, but it doesn't in the cases where a host uses multiple interfaces... -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Tue May 11 10:00:05 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA26624 for ; Tue, 11 May 2004 10:00:04 -0400 (EDT) Received: from esealmw141.al.sw.ericsson.se ([153.88.254.120]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4B7pkAh019652 for ; Tue, 11 May 2004 09:51:47 +0200 Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118]) by esealmw141.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 09:51:46 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt612.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM0DTP2P; Tue, 11 May 2004 09:51:46 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4B7pjXA011415; Tue, 11 May 2004 09:51:45 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B7p8It011410; Tue, 11 May 2004 09:51:08 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4B7p85G011409; Tue, 11 May 2004 09:51:08 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from mail1.microsoft.com (mail1.microsoft.com [131.107.3.125]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4B7p6It011404 for ; Tue, 11 May 2004 09:51:06 +0200 (MET DST) Received: from mail6.microsoft.com ([157.54.6.196]) by mail1.microsoft.com with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 00:51:06 -0700 Received: from inet-vrs-06.redmond.corp.microsoft.com ([157.54.6.181]) by mail6.microsoft.com with Microsoft SMTPSVC(6.0.3790.1069); Tue, 11 May 2004 00:51:14 -0700 Received: from 157.54.6.197 by inet-vrs-06.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Tue, 11 May 2004 00:51:04 -0700 Received: from RED-MSG-43.redmond.corp.microsoft.com ([157.54.12.203]) by INET-HUB-06.redmond.corp.microsoft.com with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 00:51:05 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: CGA issues Date: Tue, 11 May 2004 00:50:09 -0700 Message-ID: <64A531765B7C8342BFA260497BE004570226D76A@RED-MSG-43.redmond.corp.microsoft.com> Thread-Topic: CGA issues thread-index: AcQ2bJi/fP9vZUJeRi+iEuEu6HfWWQAvjnow From: "Tuomas Aura" To: "Iljitsch van Beijnum" , "Jari Arkko" Cc: X-OriginalArrivalTime: 11 May 2004 07:51:05.0545 (UTC) FILETIME=[B6A32390:01C4372C] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by sw.ericsson.se id i4B7p7It011405 Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk Content-Transfer-Encoding: 8bit Iljitsch wrote: > Still, I don't see the value in having the subnet prefix in the hash, > and there's still the issue of address rewriting. > > Are there any attacks that can be fended off this way? If it's just a > privacy thing I don't think that's enough justification, as it's > already possible to generate a new hash based on an existing key by > selecting a new "modifier" value. The only problem here is that this is > more expensive when a "Sec" is in effect. But that's pretty much a > feature rather than a downside. Supporting privacy by default is certainly not a downside. Instead, I would think that any protocol that prevents the use of privacy-friendly changing addresses (CGAs or RFC-3041 addresses) is seriously flawed. I don't understand your argument about changing the modifier vs. including the prefix in the hash. Both cause the IID to change. Changing the modifier is potentially much more expensive. Thus, including the prefix makes much more sense. > BTW, I'm no statistician and I don't play one on tv, but isn't there a > significant chance that you're going to run into statistical anomalies > when searching for a SHA-1 hash with a certain number of 0 bits in it? > I mean, the chance of getting a SHA-1 with all 0 bits in the lowest 16 > bits would be 1 in 65536. But that doesn't automatically mean that if > you try 65536 times you hit one. Has anyone done the math on this? The median number of iterations is 0.69 * 2^(16*Sec). Any "statistical anomaly" in the distribution of the hash values would indicate a flaw in the SHA-1 hash function. Tuomas -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Tue May 11 12:10:48 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA05672 for ; Tue, 11 May 2004 12:10:47 -0400 (EDT) Received: from esealmw140.al.sw.ericsson.se ([153.88.254.121]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4BGAmWR021432 for ; Tue, 11 May 2004 18:10:49 +0200 (MEST) Received: from esealnt612.al.sw.ericsson.se ([153.88.254.118]) by esealmw140.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 18:10:48 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt612.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM0DY179; Tue, 11 May 2004 18:10:48 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4BGAbwg022248; Tue, 11 May 2004 18:10:38 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4BG9eIt028896; Tue, 11 May 2004 18:09:40 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4BG9eA6028895; Tue, 11 May 2004 18:09:40 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from n97.nomadiclab.com (teldanex.hiit.fi [212.68.5.99]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4BG9dIt028891 for ; Tue, 11 May 2004 18:09:39 +0200 (MET DST) Received: from [IPv6:::1] (polle-vpn.local.pnr.iki.fi [192.168.0.193]) by n97.nomadiclab.com (Postfix) with ESMTP id 52C098; Tue, 11 May 2004 19:09:39 +0300 (EEST) In-Reply-To: <01eb01c436bb$0e382000$366115ac@dcml.docomolabsusa.com> References: <01eb01c436bb$0e382000$366115ac@dcml.docomolabsusa.com> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <9DE2BFBA-A365-11D8-AD89-000393CE1E8C@nomadiclab.com> Content-Transfer-Encoding: 7bit Cc: James Kempf , Iljitsch van Beijnum From: Pekka Nikander Subject: Re: CGA issues Date: Tue, 11 May 2004 19:09:44 +0300 To: SEND WG X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 11 May 2004 16:10:48.0686 (UTC) FILETIME=[85FB60E0:01C43772] Content-Transfer-Encoding: 7bit >> BTW, what's the implementation status of SEND/CGA? > > On the other hand, I've also heard from one influential member of the > KAME > community that they will not incorporate CGA into KAME because it has > IPR on > it. We implemented CGAs last summer on FreeBSD (and the code is available upon request, as a 150 Mb full modified source tree), but didn't implement the SEND parts. Hence, if anybody is interested in implementing (and preferably maintaining) a KAME patch, maybe our code could be used as a partial starting point. --Pekka -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Tue May 11 12:27:33 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA06736 for ; Tue, 11 May 2004 12:27:32 -0400 (EDT) Received: from esealmw142.al.sw.ericsson.se ([153.88.254.119]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4BGRYWR024197 for ; Tue, 11 May 2004 18:27:34 +0200 (MEST) Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120]) by esealmw142.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 18:27:34 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KWS1KSJP; Tue, 11 May 2004 18:27:35 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4BGRXXA022518; Tue, 11 May 2004 18:27:33 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4BGQuIt002419; Tue, 11 May 2004 18:26:56 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4BGQuAs002418; Tue, 11 May 2004 18:26:56 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from n97.nomadiclab.com (teldanex.hiit.fi [212.68.5.99]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4BGQtIt002414 for ; Tue, 11 May 2004 18:26:55 +0200 (MET DST) Received: from [IPv6:::1] (polle-vpn.local.pnr.iki.fi [192.168.0.193]) by n97.nomadiclab.com (Postfix) with ESMTP id 624D68; Tue, 11 May 2004 19:26:55 +0300 (EEST) In-Reply-To: References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> <40A05A23.1050804@kolumbus.fi> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <07C1BFF9-A368-11D8-AD89-000393CE1E8C@nomadiclab.com> Content-Transfer-Encoding: 7bit Cc: ietf-send@standards.ericsson.net, Jari Arkko From: Pekka Nikander Subject: Re: CGA for multihoming Date: Tue, 11 May 2004 19:27:00 +0300 To: Iljitsch van Beijnum X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 11 May 2004 16:27:34.0461 (UTC) FILETIME=[DD7876D0:01C43774] Content-Transfer-Encoding: 7bit Iljitsch, >>>> And couldn't the same results be reached by allowing all prefixes >>>> that >>>> are on-link, rather than just one? > > No. What I mean is that you still create the CGA the same way, but > when checking it on another system, this system doesn't require the > prefix to be the actual prefix used in the address, but also allows > the prefix to be any other prefix that is on-link for this subnet. > > So if you have a link with prefixes 2001:0:0:1::/64 and > 3ffe:0:0:1:/64, and a host generates a CGA that is > 2001:0:0:1:0:dead:0:beaf, then it can also use the address > 3ffe:0:0:1:0:dead:0:beaf. The current rules don't allow this as the > prefix used in generating the CGA is 2001:0:0:1::/64, but the modified > rule does allow this, as the prefix used in the CGA generation is > on-link. I don't understand why the host has ever to see your Y.CGA(K+X) address. If I understand correctly, the peer will definitely see it, but why has to host itself see it? Assumedly, the multi6 state at the peer could list all the possible addresses that the peer may receive, namely Y.CGA(K+Y), X.CGA(K+X) as well as Y.CGA(K+X) and X.CGA(K+Y), but mark that only the first two ones can be used as destination addresses? --Pekka -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Tue May 11 12:35:55 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA07088 for ; Tue, 11 May 2004 12:35:50 -0400 (EDT) Received: from esealmw141.al.sw.ericsson.se ([153.88.254.120]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4BGZqWR025725 for ; Tue, 11 May 2004 18:35:52 +0200 (MEST) Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw141.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Tue, 11 May 2004 18:35:52 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KND2G8RL; Tue, 11 May 2004 18:36:07 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4BGZmwg023487; Tue, 11 May 2004 18:35:48 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4BGZaIt004709; Tue, 11 May 2004 18:35:36 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4BGZZZC004708; Tue, 11 May 2004 18:35:35 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fridge.docomolabs-usa.com (key1.docomolabs-usa.com [216.98.102.225]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4BGZXIt004704 for ; Tue, 11 May 2004 18:35:34 +0200 (MET DST) Message-ID: <008c01c43776$126d5440$366115ac@dcml.docomolabsusa.com> From: "James Kempf" To: "Jari Arkko" , "Iljitsch van Beijnum" Cc: "Brian E Carpenter" , "Brian E Carpenter" , References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> <40A05A23.1050804@kolumbus.fi> Subject: Re: CGA for multihoming Date: Tue, 11 May 2004 09:36:11 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 11 May 2004 16:35:52.0355 (UTC) FILETIME=[063D0F30:01C43776] Content-Transfer-Encoding: 7bit Iljitsch, > >>> And couldn't the same results be reached by allowing all prefixes > >>> that > >>> are on-link, rather than just one? > > >> I'm not sure I understand. Could you clarify? > > > I think Iljitsch is thinking that instead of one prefix as an > > input to the CGA generation, you would use all the prefixes > > on the link. > > No. What I mean is that you still create the CGA the same way, but when > checking it on another system, this system doesn't require the prefix > to be the actual prefix used in the address, but also allows the prefix > to be any other prefix that is on-link for this subnet. > > So if you have a link with prefixes 2001:0:0:1::/64 and 3ffe:0:0:1:/64, > and a host generates a CGA that is 2001:0:0:1:0:dead:0:beaf, then it > can also use the address 3ffe:0:0:1:0:dead:0:beaf. The current rules > don't allow this as the prefix used in generating the CGA is > 2001:0:0:1::/64, but the modified rule does allow this, as the prefix > used in the CGA generation is on-link. > So I think I understand your proposal now, and I think your proposal requires deeper changes than just to SEND. To use your example, suppose a node, call it A, on a link with the prefixes 2001:0:0:1::/64 and 3ffe:0:0:1:/64 generates an address (it need not be a CGA) as 2001:0:0:1:0:dead:0:beaf. Now, suppose some router prior to the last hop router changes the prefix on the address so it become 3ffe:0:0:1:0:dead:0:beaf, as you say, in response to the ISP routing prefix 2001:0:0:1::/64 going down. The packet would arrive at the last hop router OK, but the last hop router would look into its Neighbor Cache and see that it doesn't have a mapping corresponding to 3ffe:0:0:1:0:dead:0:beaf. Then, it would do Neighbor Discovery to find out what node on the link claims that address. Why should Node A respond positively to the Neighbor Discovery request? It didn't generate 3ffe:0:0:1:0:dead:0:beaf in the first place, it generated 2001:0:0:1:0:dead:0:beaf. This problem will occur regardless of whether the address is a CGA or not. What your proposal requires is what in the IPv6 WG has been called DIID (Duplicate Interface Identifier Discovery) rather than DAD (Duplicate Address Detection). Your proposal requires that a node doing address autoconfiguration use the same interface identifier for each address configured on an interface, and to check the interface identifier rather than the address when doing duplicate detection (it need check only once for one prefix if this rule is in effect, typically the prefix will be the link local address prefix since that is the first address configured). The node would be required to respond to Neighbor Discovery messages for any on-link prefix containing the interface identifier. Probably the easiest way to do this would be to require a node to generate addresses for each prefix it sees on the link, using the same interface identifier, so that the access router has Neighbor Cache mappings for all addresses corresponding to that interface identifier. The IPv6 WG has discussed DIID a couple times (since it does provide some other benefits, like reducing how often a node needs to do duplicate detection), but I believe it has been rejected each time. Your proposal would, additionally, make RFC 3041 privacy address impossible to use, as Tuomas has stated, because RFC 3041 privacy addresses require the ability of the node to randomly generate interface identifiers. This is unlikely to be acceptable to the Security Directorate, unless you are prepared to offer an alternative that provides the same level of privacy as RFC 3041 addresses. jak -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Thu May 13 00:07:04 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA01007 for ; Thu, 13 May 2004 00:07:04 -0400 (EDT) Received: from esealmw142.al.sw.ericsson.se ([153.88.254.119]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4D475Ah028145 for ; Thu, 13 May 2004 06:07:05 +0200 Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw142.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Thu, 13 May 2004 06:07:04 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM9PACYC; Thu, 13 May 2004 06:07:04 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4D46vwg018976; Thu, 13 May 2004 06:06:58 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4D464It019172; Thu, 13 May 2004 06:06:04 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4D464OU019171; Thu, 13 May 2004 06:06:04 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fep22-app.kolumbus.fi (fep22-0.kolumbus.fi [193.229.0.60]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4D463It019167 for ; Thu, 13 May 2004 06:06:03 +0200 (MET DST) Received: from kolumbus.fi ([80.186.217.250]) by fep22-app.kolumbus.fi with ESMTP id <20040513040602.LBTL27645.fep22-app.kolumbus.fi@kolumbus.fi> for ; Thu, 13 May 2004 07:06:02 +0300 Message-ID: <40A2F35C.3090604@kolumbus.fi> Date: Thu, 13 May 2004 07:02:36 +0300 From: Jari Arkko User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7b) Gecko/20040316 X-Accept-Language: en-us, en MIME-Version: 1.0 To: SEND WG Subject: RFC 3756 on IPv6 Neighbor Discovery (ND) Trust Models and Threats Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 13 May 2004 04:07:04.0990 (UTC) FILETIME=[C044B3E0:01C4389F] Content-Transfer-Encoding: 7bit A new Request for Comments is now available in online RFC libraries. RFC 3756 Title: IPv6 Neighbor Discovery (ND) Trust Models and Threats Author(s): P. Nikander, Ed., J. Kempf, E. Nordmark Status: Informational Date: May 2004 Mailbox: pekka.nikander@nomadiclab.com, kempf@docomolabs-usa.com, erik.nordmark@sun.com Pages: 23 Characters: 56674 Updates/Obsoletes/SeeAlso: None I-D Tag: draft-ietf-send-psreq-04.txt URL: ftp://ftp.rfc-editor.org/in-notes/rfc3756.txt The existing IETF standards specify that IPv6 Neighbor Discovery (ND) and Address Autoconfiguration mechanisms may be protected with IPsec Authentication Header (AH). However, the current specifications limit the security solutions to manual keying due to practical problems faced with automatic key management. This document specifies three different trust models and discusses the threats pertinent to IPv6 Neighbor Discovery. The purpose of this discussion is to define the requirements for Securing IPv6 Neighbor Discovery. This document is a product of the Securing Neighbor Discovery Working Group of the IETF. This memo provides information for the Internet community. It does not specify an Internet standard of any kind. Distribution of this memo is unlimited. This announcement is sent to the IETF list and the RFC-DIST list. Requests to be added to or deleted from the IETF distribution list should be sent to IETF-REQUEST@IETF.ORG. Requests to be added to or deleted from the RFC-DIST distribution list should be sent to RFC-DIST-REQUEST@RFC-EDITOR.ORG. Details on obtaining RFCs via FTP or EMAIL may be obtained by sending an EMAIL message to rfc-info@RFC-EDITOR.ORG with the message body help: ways_to_get_rfcs. For example: To: rfc-info@RFC-EDITOR.ORG Subject: getting rfcs help: ways_to_get_rfcs Requests for special distribution should be addressed to either the author of the RFC in question, or to RFC-Manager@RFC-EDITOR.ORG. Unless specifically noted otherwise on the RFC itself, all RFCs are for unlimited distribution.echo Submissions for Requests for Comments should be sent to RFC-EDITOR@RFC-EDITOR.ORG. Please consult RFC 2223, Instructions to RFC Authors, for further information. Joyce K. Reynolds and Sandy Ginoza USC/Information Sciences Institute -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Thu May 13 06:49:23 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA04150 for ; Thu, 13 May 2004 06:49:22 -0400 (EDT) Received: from esealmw140.al.sw.ericsson.se ([153.88.254.121]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4DAnOAh016933 for ; Thu, 13 May 2004 12:49:24 +0200 Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw140.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Thu, 13 May 2004 12:49:23 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM9PDGLF; Thu, 13 May 2004 12:49:23 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4DAnLwg005874; Thu, 13 May 2004 12:49:21 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4DAmXIt021450; Thu, 13 May 2004 12:48:33 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4DAmXwA021449; Thu, 13 May 2004 12:48:33 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from sequoia.muada.com (sequoia.muada.com [83.149.65.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4DAmWIt021445 for ; Thu, 13 May 2004 12:48:32 +0200 (MET DST) Received: from [127.0.0.1] (sequoia.muada.com [83.149.65.1]) by sequoia.muada.com (8.12.10/8.12.10) with ESMTP id i4DAm41C055917; Thu, 13 May 2004 12:48:05 +0200 (CEST) (envelope-from iljitsch@muada.com) In-Reply-To: <07C1BFF9-A368-11D8-AD89-000393CE1E8C@nomadiclab.com> References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> <40A05A23.1050804@kolumbus.fi> <07C1BFF9-A368-11D8-AD89-000393CE1E8C@nomadiclab.com> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <0A1374B0-A4CB-11D8-B6DD-000A95CD987A@muada.com> Content-Transfer-Encoding: 7bit Cc: ietf-send@standards.ericsson.net From: Iljitsch van Beijnum Subject: Re: CGA for multihoming Date: Thu, 13 May 2004 12:48:16 +0200 To: Pekka Nikander X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 13 May 2004 10:49:23.0908 (UTC) FILETIME=[F42F4840:01C438D7] Content-Transfer-Encoding: 7bit On 11-mei-04, at 18:27, Pekka Nikander wrote: >> No. What I mean is that you still create the CGA the same way, but >> when checking it on another system, this system doesn't require the >> prefix to be the actual prefix used in the address, but also allows >> the prefix to be any other prefix that is on-link for this subnet. >> So if you have a link with prefixes 2001:0:0:1::/64 and >> 3ffe:0:0:1:/64, and a host generates a CGA that is >> 2001:0:0:1:0:dead:0:beaf, then it can also use the address >> 3ffe:0:0:1:0:dead:0:beaf. The current rules don't allow this as the >> prefix used in generating the CGA is 2001:0:0:1::/64, but the >> modified rule does allow this, as the prefix used in the CGA >> generation is on-link. > I don't understand why the host has ever to see your Y.CGA(K+X) > address. > If I understand correctly, the peer will definitely see it, but why has > to host itself see it? Assumedly, the multi6 state at the peer could > list > all the possible addresses that the peer may receive, namely > Y.CGA(K+Y), > X.CGA(K+X) as well as Y.CGA(K+X) and X.CGA(K+Y), but mark that only the > first two ones can be used as destination addresses? You make a good point. Still, I would be much happier if hosts could have the same interface identifier within different prefixes. This would make some issues such as ICMP packets generated for rewritten addresses easier to deal with. BTW, what's the deployment scenario for CGA anyway? I guess it's somewhat useful that an attacker can't spoof packets and make them seem to come from another host or router, but how does this stop a malicious system from presenting itself as a legitimate router? -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Thu May 13 07:56:53 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA06658 for ; Thu, 13 May 2004 07:56:49 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4DBumWR025007 for ; Thu, 13 May 2004 13:56:49 +0200 (MEST) Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Thu, 13 May 2004 13:56:48 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KND2WS9W; Thu, 13 May 2004 13:57:08 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4DBufwg009253; Thu, 13 May 2004 13:56:41 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4DBtsIt005798; Thu, 13 May 2004 13:55:54 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4DBtsuc005797; Thu, 13 May 2004 13:55:54 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from sequoia.muada.com (sequoia.muada.com [83.149.65.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4DBtnIt005781 for ; Thu, 13 May 2004 13:55:50 +0200 (MET DST) Received: from [127.0.0.1] (sequoia.muada.com [83.149.65.1]) by sequoia.muada.com (8.12.10/8.12.10) with ESMTP id i4DBtV1C057063; Thu, 13 May 2004 13:55:33 +0200 (CEST) (envelope-from iljitsch@muada.com) In-Reply-To: <008c01c43776$126d5440$366115ac@dcml.docomolabsusa.com> References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> <40A05A23.1050804@kolumbus.fi> <008c01c43776$126d5440$366115ac@dcml.docomolabsusa.com> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <7630EF64-A4D4-11D8-B6DD-000A95CD987A@muada.com> Content-Transfer-Encoding: 7bit Cc: Brian Carpenter , "" From: Iljitsch van Beijnum Subject: Re: CGA for multihoming Date: Thu, 13 May 2004 13:55:43 +0200 To: "James Kempf" X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 13 May 2004 11:56:48.0758 (UTC) FILETIME=[5F1A6960:01C438E1] Content-Transfer-Encoding: 7bit On 11-mei-04, at 18:36, James Kempf wrote: >> What I mean is that you still create the CGA the same way, but when >> checking it on another system, this system doesn't require the prefix >> to be the actual prefix used in the address, but also allows the >> prefix >> to be any other prefix that is on-link for this subnet. >> So if you have a link with prefixes 2001:0:0:1::/64 and >> 3ffe:0:0:1:/64, >> and a host generates a CGA that is 2001:0:0:1:0:dead:0:beaf, then it >> can also use the address 3ffe:0:0:1:0:dead:0:beaf. The current rules >> don't allow this as the prefix used in generating the CGA is >> 2001:0:0:1::/64, but the modified rule does allow this, as the prefix >> used in the CGA generation is on-link. > So I think I understand your proposal now, and I think your proposal > requires deeper changes than just to SEND. Of course. > To use your example, suppose a > node, call it A, on a link with the prefixes 2001:0:0:1::/64 and > 3ffe:0:0:1:/64 generates an address (it need not be a CGA) as > 2001:0:0:1:0:dead:0:beaf. Now, suppose some router prior to the last > hop > router changes the prefix on the address so it become > 3ffe:0:0:1:0:dead:0:beaf, as you say, in response to the ISP routing > prefix > 2001:0:0:1::/64 going down. [I think you're omitting the part that the packet is delivered at the correspondent, which then sends back a packet to the rewritten address.] > The packet would arrive at the last hop router > OK, but the last hop router would look into its Neighbor Cache and see > that > it doesn't have a mapping corresponding to 3ffe:0:0:1:0:dead:0:beaf. > Then, > it would do Neighbor Discovery to find out what node on the link > claims that > address. > Why should Node A respond positively to the Neighbor Discovery > request? It > didn't generate 3ffe:0:0:1:0:dead:0:beaf in the first place, it > generated > 2001:0:0:1:0:dead:0:beaf. This problem will occur regardless of > whether the > address is a CGA or not. Obviously the idea is that the host DOES configure the address 3ffe:0:0:1:0:dead:0:beaf for itself after generating the CGA 2001:0:0:1:0:dead:0:beaf. Note that today, hosts (at least the ones I've seen) will combine all advertised prefixes with all interface identifiers so this behavior would be fairly natural. > Your proposal would, additionally, make RFC 3041 privacy address > impossible > to use, as Tuomas has stated, because RFC 3041 privacy addresses > require the > ability of the node to randomly generate interface identifiers. How would this be impossible? The only requirement is that the interface identifier is used with all available prefixes. Win XP (for instance) already does this. > This is unlikely to be acceptable to the Security Directorate, unless > you are prepared to offer an alternative that provides the same level > of privacy as RFC 3041 addresses. Note that the privacy concern that led to the creation of RFC 3041 is that a MAC address could be tracked when it moves from site to site. As such, only changing interface identifiers when moving from site to site would be enough to alleviate privacy concerns, and this doesn't get in the way of multihoming. -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Thu May 13 11:41:29 2004 Received: from penguin.ericsson.se (penguin.ericsson.se [193.180.251.47]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA19308 for ; Thu, 13 May 2004 11:41:28 -0400 (EDT) Received: from esealmw141.al.sw.ericsson.se ([153.88.254.120]) by penguin.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4DFfSPA020806 for ; Thu, 13 May 2004 17:41:28 +0200 (MEST) Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw141.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Thu, 13 May 2004 17:41:27 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KND2YPGL; Thu, 13 May 2004 17:41:48 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4DFfQXA009584; Thu, 13 May 2004 17:41:26 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4DFeQIt024080; Thu, 13 May 2004 17:40:26 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4DFeQOW024075; Thu, 13 May 2004 17:40:26 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fridge.docomolabs-usa.com (key1.docomolabs-usa.com [216.98.102.225]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4DFeMIt024056 for ; Thu, 13 May 2004 17:40:23 +0200 (MET DST) Message-ID: <006401c43900$b16b29b0$366115ac@dcml.docomolabsusa.com> From: "James Kempf" To: "Iljitsch van Beijnum" , "Pekka Nikander" Cc: References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> <40A05A23.1050804@kolumbus.fi> <07C1BFF9-A368-11D8-AD89-000393CE1E8C@nomadiclab.com> <0A1374B0-A4CB-11D8-B6DD-000A95CD987A@muada.com> Subject: Re: CGA for multihoming Date: Thu, 13 May 2004 08:40:59 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 13 May 2004 15:41:27.0983 (UTC) FILETIME=[C158EBF0:01C43900] Content-Transfer-Encoding: 7bit > BTW, what's the deployment scenario for CGA anyway? I guess it's > somewhat useful that an attacker can't spoof packets and make them seem > to come from another host or router, but how does this stop a malicious > system from presenting itself as a legitimate router? > CGA isn't used for that. There's another part of SEND that involves routers having certificates which are used to sign Router Advertisements. The host checks the signature. jak -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Thu May 13 12:16:00 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA20973 for ; Thu, 13 May 2004 12:16:00 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4DGG1Ah023920 for ; Thu, 13 May 2004 18:16:01 +0200 Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Thu, 13 May 2004 18:16:01 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM9PF0VB; Thu, 13 May 2004 18:16:01 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4DGG0XA010075; Thu, 13 May 2004 18:16:00 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4DGFNIt002042; Thu, 13 May 2004 18:15:23 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4DGFN4Z002033; Thu, 13 May 2004 18:15:23 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fridge.docomolabs-usa.com (key1.docomolabs-usa.com [216.98.102.225]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4DGFHIt001786 for ; Thu, 13 May 2004 18:15:17 +0200 (MET DST) Message-ID: <009201c43905$92170b60$366115ac@dcml.docomolabsusa.com> From: "James Kempf" To: "Iljitsch van Beijnum" Cc: "Brian Carpenter" , References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> <40A05A23.1050804@kolumbus.fi> <008c01c43776$126d5440$366115ac@dcml.docomolabsusa.com> <7630EF64-A4D4-11D8-B6DD-000A95CD987A@muada.com> Subject: Re: CGA for multihoming Date: Thu, 13 May 2004 09:15:54 -0700 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 13 May 2004 16:16:01.0614 (UTC) FILETIME=[9553E6E0:01C43905] Content-Transfer-Encoding: 7bit > > Your proposal would, additionally, make RFC 3041 privacy address > > impossible > > to use, as Tuomas has stated, because RFC 3041 privacy addresses > > require the > > ability of the node to randomly generate interface identifiers. > > How would this be impossible? The only requirement is that the > interface identifier is used with all available prefixes. Win XP (for > instance) already does this. > > > This is unlikely to be acceptable to the Security Directorate, unless > > you are prepared to offer an alternative that provides the same level > > of privacy as RFC 3041 addresses. > > Note that the privacy concern that led to the creation of RFC 3041 is > that a MAC address could be tracked when it moves from site to site. As > such, only changing interface identifiers when moving from site to site > would be enough to alleviate privacy concerns, and this doesn't get in > the way of multihoming. > > RFC 3041 currently puts no restrictions on how the random interface identifiers are generated. Thus if a user felt that someone was watching their traffic, the user could force a change in interface identifiers at any time and without restriction to try to break the correlation. In addition, if more than one address has the same interface identifier, it provides a snooper with an additional data point about the host. While current practice on many operating systems (for example Linux) is to claim addresses with the same interface identifier for all prefixes, RFC 3041 is really only 3 years old, and widespread deployment of IPv6 has yet to occur, so ISPs and others concerned with deployment may not have faced up to the issue yet. The proposal you've described would establish a correlation between addresses having different prefixes. If a host wants to use different prefixes with different interface identifiers to obscure its traffic, for example by routing particular kinds of traffic over one prefix in order to foil monitoring, that would not be possible. This would result in a net reduction in address privacy. Speaking with my WG chair hat off and operator hat on, in Japan privacy, in particular address and location privacy, is a major concern primarily because the social and legal concern with privacy in Japan is considerably higher than in the US and perhaps even than in Europe(for example, they have nothing like a Social Security number, due to concerns about it being misused). Japan had strong regulations governing misuse of location information from mobile devices such as cell phones and laptops years before the EU, and the regulations in the US even now are laughable by comparison.* Any proposal that would result in a net reduction in privacy would not be to the benefit of DoCoMo's Japanese customers. Speaking with my WG chair hat on again, my concern is that the net effect of the proposal, both for CGA and for RFC 3041, would result in a net reduction in address privacy and security of CG addresses from guessing attacks. Should the proposal reach Last Call (which, from the lack of response to my previous question about where it is in the process, I would guess is not imminent) I or someone else is sure to bring these questions up. In addition, I know that there are alternate multi6 proposals, such as NOID, that don't have these issues. As such, I don't believe at this time that changes in the CGA draft are wise. Should the Security Directorate feel that the proposal is acceptable, and all raised questions be addressed, the CGA draft can be revise at that time. Anyone else in the WG care to comment? jak * For a short comparison of location privacy regulation in the US, Europe, and Japan see the ISOC Member Briefing at: http://www.isoc.org/briefings/015/index.shtml -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Thu May 13 12:58:54 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA23410 for ; Thu, 13 May 2004 12:58:53 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4DGwsWR004134 for ; Thu, 13 May 2004 18:58:55 +0200 (MEST) Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Thu, 13 May 2004 18:58:54 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KWS18MSP; Thu, 13 May 2004 18:59:03 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4DGwqwg025261; Thu, 13 May 2004 18:58:53 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4DGwGIt010479; Thu, 13 May 2004 18:58:16 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4DGwGoG010478; Thu, 13 May 2004 18:58:16 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from sequoia.muada.com (sequoia.muada.com [83.149.65.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4DGwBIt010471 for ; Thu, 13 May 2004 18:58:11 +0200 (MET DST) Received: from [127.0.0.1] (sequoia.muada.com [83.149.65.1]) by sequoia.muada.com (8.12.10/8.12.10) with ESMTP id i4DGvr1C062169; Thu, 13 May 2004 18:57:54 +0200 (CEST) (envelope-from iljitsch@muada.com) In-Reply-To: <009201c43905$92170b60$366115ac@dcml.docomolabsusa.com> References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> <40A05A23.1050804@kolumbus.fi> <008c01c43776$126d5440$366115ac@dcml.docomolabsusa.com> <7630EF64-A4D4-11D8-B6DD-000A95CD987A@muada.com> <009201c43905$92170b60$366115ac@dcml.docomolabsusa.com> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: Content-Transfer-Encoding: 7bit Cc: , "Brian Carpenter" From: Iljitsch van Beijnum Subject: Re: CGA for multihoming Date: Thu, 13 May 2004 18:58:05 +0200 To: "James Kempf" X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 13 May 2004 16:58:54.0569 (UTC) FILETIME=[92EDAD90:01C4390B] Content-Transfer-Encoding: 7bit On 13-mei-04, at 18:15, James Kempf wrote: > RFC 3041 currently puts no restrictions on how the random interface > identifiers are generated. If people want to change their interface identifier every minute, I don't care. (Don't expect me to jump through hoops because of this behavior, though.) But it is NOT a good idea to make impossible the current practice and possible required future requirement that host creates an address for each available prefix using a certain interface identifier. > In addition, > if more than one address has the same interface identifier, it > provides a > snooper with an additional data point about the host. Only in theory. Describe for me a realistic scenario where a host would be required to leak this information that would otherwise be unavailable. > The proposal you've described would establish a correlation between > addresses having different prefixes. If a host wants to use different > prefixes with different interface identifiers to obscure its traffic, > for > example by routing particular kinds of traffic over one prefix in > order to > foil monitoring, that would not be possible. Of course this would be possible. What you say is that it must be mandatory. That is a completely different thing. > This would result in a net reduction in address privacy. 0. End-users will in most cases be identifyable by the first 48 or 64 bits of their address 1. NOT requiring hosts to NOT use the same interface identifier for different prefixes in and of its own doesn't impact privacy 2. Multihoming may lead to leakage of much more information anyway 3. The tradeoff between privacy and other issues doesn't automatically fall towards privacy > Speaking with my WG chair hat off and operator hat on, in Japan > privacy, in > particular address and location privacy, is a major concern primarily > because the social and legal concern with privacy in Japan is > considerably > higher than in the US and perhaps even than in Europe(for example, > they have > nothing like a Social Security number, due to concerns about it being > misused). Japan had strong regulations governing misuse of location > information from mobile devices such as cell phones and laptops years > before > the EU, and the regulations in the US even now are laughable by > comparison.* Hm, strange that there is no RFC 3041 support in the KAME-derived IPv6 stacks I'm familiar with then. > Speaking with my WG chair hat on again, my concern is that the net > effect of > the proposal, both for CGA and for RFC 3041, would result in a net > reduction > in address privacy and security of CG addresses from guessing attacks. Your concern is unfounded. Please take the time to familiarize yourself with the issues. -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Thu May 13 23:07:37 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA05669 for ; Thu, 13 May 2004 23:07:34 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4E37YAh009917 for ; Fri, 14 May 2004 05:07:35 +0200 Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Fri, 14 May 2004 05:07:34 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KND26P15; Fri, 14 May 2004 05:07:56 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4E37Uwg017506; Fri, 14 May 2004 05:07:30 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4E36hIt021874; Fri, 14 May 2004 05:06:43 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4E36h4B021873; Fri, 14 May 2004 05:06:43 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from shuttle.wide.toshiba.co.jp (shuttle.wide.toshiba.co.jp [202.249.10.124]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4E36eIt021869 for ; Fri, 14 May 2004 05:06:41 +0200 (MET DST) Received: from ocean.jinmei.org (unknown [3ffe:501:100f:1048:96f:69a3:96a3:6369]) by shuttle.wide.toshiba.co.jp (Postfix) with ESMTP id 31A6015263; Fri, 14 May 2004 12:06:35 +0900 (JST) Date: Fri, 14 May 2004 12:06:49 +0900 Message-ID: From: JINMEI Tatuya / =?ISO-2022-JP?B?GyRCP0BMQEMjOkgbKEI=?= To: Iljitsch van Beijnum Cc: "James Kempf" , , "Brian Carpenter" Subject: Re: CGA for multihoming In-Reply-To: References: <591B0516-A1C0-11D8-BE01-000A95CD987A@muada.com> <409E54E0.1010900@kolumbus.fi> <74A83498-A1F9-11D8-BE01-000A95CD987A@muada.com> <020101c436be$d5317190$366115ac@dcml.docomolabsusa.com> <000901c436f9$5f1b0760$816115ac@dcml.docomolabsusa.com> <40A05A23.1050804@kolumbus.fi> <008c01c43776$126d5440$366115ac@dcml.docomolabsusa.com> <7630EF64-A4D4-11D8-B6DD-000A95CD987A@muada.com> <009201c43905$92170b60$366115ac@dcml.docomolabsusa.com> User-Agent: Wanderlust/2.10.1 (Watching The Wheels) Emacs/21.3 Mule/5.0 (SAKAKI) Organization: Research & Development Center, Toshiba Corp., Kawasaki, Japan. MIME-Version: 1.0 (generated by SEMI 1.14.5 - "Awara-Onsen") Content-Type: text/plain; charset=US-ASCII Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 14 May 2004 03:07:34.0193 (UTC) FILETIME=[9A521210:01C43960] A couple of quick clarifications. >>>>> On Thu, 13 May 2004 18:58:05 +0200, >>>>> Iljitsch van Beijnum said: >> Speaking with my WG chair hat off and operator hat on, in Japan >> privacy, in >> particular address and location privacy, is a major concern primarily >> because the social and legal concern with privacy in Japan is >> considerably >> higher than in the US and perhaps even than in Europe(for example, >> they have >> nothing like a Social Security number, due to concerns about it being >> misused). Japan had strong regulations governing misuse of location >> information from mobile devices such as cell phones and laptops years >> before >> the EU, and the regulations in the US even now are laughable by >> comparison.* > Hm, strange that there is no RFC 3041 support in the KAME-derived IPv6 > stacks I'm familiar with then. FreeBSD (which should definitely have a "KAME-derived" IPv6 stack) has had support for RFC3041 for a certain period (at least FreeBSD 4.8R has it). KAME snapshots have supported RFC3041 since February 2001. Secondly, even if no KAME-derived stacks support RFC3041, I would say it's irrelevant to what James said. KAME's target has been the entire Internet community all over the world from the beginning, and has never been limited to the Japanese community (the project does even not have Japanese web pages except developers's local ones under kame.net or a mailing list limited to Japanese users). JINMEI, Tatuya Communication Platform Lab. Corporate R&D Center, Toshiba Corp. jinmei@isl.rdc.toshiba.co.jp -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Mon May 17 23:39:08 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA26439 for ; Mon, 17 May 2004 23:39:07 -0400 (EDT) Received: from esealmw141.al.sw.ericsson.se ([153.88.254.120]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4I3d8Ah030422 for ; Tue, 18 May 2004 05:39:08 +0200 Received: from esealnt613.al.sw.ericsson.se ([153.88.254.125]) by esealmw141.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Tue, 18 May 2004 05:39:08 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt613.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id KM9QFGC1; Tue, 18 May 2004 05:39:07 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4I3d6XA019717; Tue, 18 May 2004 05:39:06 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4I3btIt013336; Tue, 18 May 2004 05:37:55 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4I3btmn013335; Tue, 18 May 2004 05:37:55 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from fep01-app.kolumbus.fi (fep01-0.kolumbus.fi [193.229.0.41]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4I3bsIt013331 for ; Tue, 18 May 2004 05:37:54 +0200 (MET DST) Received: from kolumbus.fi ([80.186.217.250]) by fep01-app.kolumbus.fi with ESMTP id <20040518033747.KWEM4346.fep01-app.kolumbus.fi@kolumbus.fi> for ; Tue, 18 May 2004 06:37:47 +0300 Message-ID: <40A98434.3000606@kolumbus.fi> Date: Tue, 18 May 2004 06:34:12 +0300 From: Jari Arkko User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.7b) Gecko/20040316 X-Accept-Language: en-us, en MIME-Version: 1.0 To: SEND WG Subject: (Fwd) Protocol Action: 'Cryptographically Generated Addresses (CGA)' to Proposed Standard Content-Type: text/plain; charset=us-ascii; format=flowed Content-Transfer-Encoding: 7bit Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 18 May 2004 03:39:08.0144 (UTC) FILETIME=[ACDB0F00:01C43C89] Content-Transfer-Encoding: 7bit The IESG has approved the following document: - 'Cryptographically Generated Addresses (CGA) ' as a Proposed Standard This document is the product of the Securing Neighbor Discovery Working Group. The IESG contact persons are Margaret Wasserman and Thomas Narten. Technical Summary This document describes a method for binding a public signature key to an IPv6 address in the Secure Neighbor Discovery (SEND) protocol. Cryptographically Generated Addresses (CGA) are IPv6 addresses where the interface identifier is generated by computing a cryptographic one-way hash function from a public key and auxiliary parameters. The binding between the public key and the address can be verified by re-computing the hash value and by comparing the hash with the interface identifier. Messages sent from an IPv6 address can be protected by attaching the public key and auxiliary parameters and by signing the message with the corresponding private key. The protection works without a certification authority or other security infrastructure. Working Group Summary This document is the work output of the SEND WG. It reflects updates to address last call review comments from Pekka Savola. Protocol Quality This document was reviewed for the IESG by Margaret Wasserman. -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Thu May 20 15:13:40 2004 Received: from eagle.ericsson.se (eagle.ericsson.se [193.180.251.53]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA13517 for ; Thu, 20 May 2004 15:13:39 -0400 (EDT) Received: from esealmw143.al.sw.ericsson.se ([153.88.254.118]) by eagle.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4KJDdAh030419 for ; Thu, 20 May 2004 21:13:40 +0200 Received: from esealnt610.al.sw.ericsson.se ([153.88.254.120]) by esealmw143.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Thu, 20 May 2004 21:13:39 +0200 Received: from tjatte.sw.ericsson.se ([153.88.242.9]) by esealnt610.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id LFSXVH91; Thu, 20 May 2004 21:13:39 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by tjatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4KJDcXA005559; Thu, 20 May 2004 21:13:38 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4KJCLIt001155; Thu, 20 May 2004 21:12:21 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4KJCK0d001154; Thu, 20 May 2004 21:12:20 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from mail4.microsoft.com (mail4.microsoft.com [131.107.3.122]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4KJCIIt001146 for ; Thu, 20 May 2004 21:12:19 +0200 (MET DST) Received: from inet-vrs-04.redmond.corp.microsoft.com ([157.54.8.149]) by mail4.microsoft.com with Microsoft SMTPSVC(6.0.3790.0); Thu, 20 May 2004 12:12:17 -0700 Received: from 157.54.8.23 by inet-vrs-04.redmond.corp.microsoft.com (InterScan E-Mail VirusWall NT); Thu, 20 May 2004 12:12:17 -0700 Received: from RED-MSG-43.redmond.corp.microsoft.com ([157.54.12.203]) by inet-hub-01.redmond.corp.microsoft.com with Microsoft SMTPSVC(6.0.3790.0); Thu, 20 May 2004 12:12:32 -0700 X-MimeOLE: Produced By Microsoft Exchange V6.5.7226.0 Content-class: urn:content-classes:message MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Subject: RE: CGA issues Date: Thu, 20 May 2004 12:12:09 -0700 Message-ID: <64A531765B7C8342BFA260497BE004570244F33A@RED-MSG-43.redmond.corp.microsoft.com> Thread-Topic: CGA issues thread-index: AcQ3PGXlYAoMfgxRT8mLxHWp4hn7vwHYHQhQ From: "Tuomas Aura" To: "Iljitsch van Beijnum" Cc: X-OriginalArrivalTime: 20 May 2004 19:12:32.0886 (UTC) FILETIME=[671C0160:01C43E9E] Content-Transfer-Encoding: 8bit X-MIME-Autoconverted: from quoted-printable to 8bit by sw.ericsson.se id i4KJCJIt001147 Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk Content-Transfer-Encoding: 8bit Iljitsch van Beijnum wrote: > >> I mean, the chance of getting a SHA-1 with all 0 bits in the lowest 16 > >> bits would be 1 in 65536. But that doesn't automatically mean that if > >> you try 65536 times you hit one. Has anyone done the math on this? > > > The median number of iterations is 0.69 * 2^(16*Sec). > > So where is the 99th percentile? Or the 99.99th percentile? In general, if one wants to find a suitable modifier with probability p, the number trial required is n = -ln(1-p) * 2^(16*Sec). For example: p=0.5 --> n = 0.69 * 2^(16^Sec) p=0.99 --> n = 4.6 * 2^(16^Sec) p=0.9999 --> n = 9.2 * 2^(16^Sec) p=0.999999 --> n = 13.8 * 2^(16^Sec) Tuomas -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From jari.arkko@lmf.ericsson.se Fri May 21 10:35:30 2004 Received: from albatross.ericsson.se (albatross.ericsson.se [193.180.251.49]) by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA05560 for ; Fri, 21 May 2004 10:35:29 -0400 (EDT) Received: from esealmw140.al.sw.ericsson.se ([153.88.254.121]) by albatross.ericsson.se (8.12.10/8.12.10/WIREfire-1.8b) with ESMTP id i4LEZUWR004348 for ; Fri, 21 May 2004 16:35:30 +0200 (MEST) Received: from esealnt611.al.sw.ericsson.se ([153.88.254.121]) by esealmw140.al.sw.ericsson.se with Microsoft SMTPSVC(6.0.3790.0); Fri, 21 May 2004 16:35:30 +0200 Received: from fnatte.sw.ericsson.se ([153.88.242.8]) by esealnt611.al.sw.ericsson.se with SMTP (Microsoft Exchange Internet Mail Service Version 5.5.2657.72) id LFSZMK0K; Fri, 21 May 2004 16:35:29 +0200 Received: from sw.ericsson.se (prdxweb.sw.ericsson.se [153.88.240.43]) by fnatte.sw.ericsson.se (8.12.10/8.12.10) with ESMTP id i4LEZPwg004092; Fri, 21 May 2004 16:35:25 +0200 (MEST) Received: from prdxweb.sw.ericsson.se (localhost [127.0.0.1]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4LEYcIt012888; Fri, 21 May 2004 16:34:38 +0200 (MET DST) Received: (from ietfmdomo@localhost) by prdxweb.sw.ericsson.se (8.12.10/8.12.10/Submit) id i4LEYc7G012887; Fri, 21 May 2004 16:34:38 +0200 (MET DST) X-Authentication-Warning: prdxweb.sw.ericsson.se: ietfmdomo set sender to owner-ietf-send@standards.ericsson.net using -f Received: from n97.nomadiclab.com (teldanex.hiit.fi [212.68.5.99]) by sw.ericsson.se (8.12.10/8.12.10/unixcenter-xnetx-1.0) with ESMTP id i4LEYZIt012877 for ; Fri, 21 May 2004 16:34:37 +0200 (MET DST) Received: from [IPv6:::1] (teldanex-vpn.local.pnr.iki.fi [192.168.0.194]) by n97.nomadiclab.com (Postfix) with ESMTP id 2D6918; Fri, 21 May 2004 17:34:35 +0300 (EEST) In-Reply-To: <64A531765B7C8342BFA260497BE004570244F33A@RED-MSG-43.redmond.corp.microsoft.com> References: <64A531765B7C8342BFA260497BE004570244F33A@RED-MSG-43.redmond.corp.microsoft.com> Mime-Version: 1.0 (Apple Message framework v613) Content-Type: text/plain; charset=US-ASCII; format=flowed Message-Id: <0208C4E3-AB34-11D8-AA2E-000393CE1E8C@nomadiclab.com> Content-Transfer-Encoding: 7bit Cc: SEND WG From: Pekka Nikander Subject: Re: CGA issues Date: Fri, 21 May 2004 17:34:46 +0300 To: "Tuomas Aura" X-Mailer: Apple Mail (2.613) Sender: owner-ietf-send@standards.ericsson.net Precedence: bulk X-OriginalArrivalTime: 21 May 2004 14:35:30.0071 (UTC) FILETIME=[DD8D7A70:01C43F40] Content-Transfer-Encoding: 7bit >>>> I mean, the chance of getting a SHA-1 with all 0 bits in >>>> the lowest 16 bits would be 1 in 65536. But that doesn't >>>> automatically mean that if you try 65536 times you hit >>>> one. Has anyone done the math on this? >> >>> The median number of iterations is 0.69 * 2^(16*Sec). >> >> So where is the 99th percentile? Or the 99.99th percentile? > > In general, if one wants to find a suitable modifier with probability > p, > the number trial required is n = -ln(1-p) * 2^(16*Sec). For example: > p=0.5 --> n = 0.69 * 2^(16^Sec) > p=0.99 --> n = 4.6 * 2^(16^Sec) > p=0.9999 --> n = 9.2 * 2^(16^Sec) > p=0.999999 --> n = 13.8 * 2^(16^Sec) But that only if you assume that the hash function has the same properties as a random function, if I understand correctly. But aren't hash functions "flatter" than random functions? (Yes, this is an academic question, but I have to admit that my math is rusty and I'd like to understand this better.) --Pekka -------------------------------------------------------------------- To unsubscribe from this list, send email with "UNSUBSCRIBE" in the body to . Archive: http://standards.ericsson.net/lists/ietf-send/maillist.html -------------------------------------------------------------------- From sani@artist.net.my Fri May 28 21:00:01 2004 Received: from zodiachost ([202.157.185.16]) by ietf.org (8.9.1a/8.9.1a) with SMTP id UAA21076 for ; Fri, 28 May 2004 20:59:59 -0400 (EDT) Received: (qmail 24534 invoked from network); 29 May 2004 00:55:59 -0000 Received: from localhost.webvis.net (HELO www.promail.net.my) (127.0.0.1) by zodiachost with SMTP; 29 May 2004 00:55:59 -0000 Received: from 216.139.180.24 (SquirrelMail authenticated user sani@artist.net.my) by www.promail.net.my with HTTP; Sat, 29 May 2004 08:55:59 +0800 (MYT) Message-ID: <1258.216.139.180.24.1085792159.squirrel@www.promail.net.my> Date: Sat, 29 May 2004 08:55:59 +0800 (MYT) Subject: from mohammed. From: "sani,hassan" To: moha2@promail.net.my User-Agent: SquirrelMail/1.4.2 MIME-Version: 1.0 Content-Type: text/html;charset=iso-8859-1 Content-Transfer-Encoding: 8bit X-Priority: 3 Importance: Normal Content-Transfer-Encoding: 8bit

Attn: Dear Sir,

 I am Mohammed Abacha, the eldest son of the late
president of Nigeria, General Sani Abacha. I was
falsely accused of  murder and as such was imprisoned,
but thanks to Allah, I have been  released for my
innocence.
 
Well dear friend I need your assistance in
transferring some  of my money into your account,
because the government is  making plans to seize them,
as they did to my fathers own.
 
Please view this site and read it's content
carefully,
http://news.bbc.co.uk/1/hi/world/africa/468903.stm
 
 
 the amount is $20 million in a Security firm Abroad.
All that is  needed is for me to instruct the company
to transfer the  funds to your account, I will
remunerate you with 20% of the  total funds
transferred to your vital bank account as compensation
for  your assistance, Five percent (5%) would be set
aside to  take care of all expenses we may incur
during the transaction.

To  indicate your interest,
contact my lawyer urgently and
confidentially for more information and the roles you
will play in this business. All the legal information
concerning  this Money will be sent to you as soon as
we agree together.
 
All my calls are  monitored, So I will get my private attorney to
get you the  needed information.
 
 If this proposal satisfies you, please contact my
lawyer  immediately with your full names, telephone
and fax numbers  to enable my lawyer, Barrister Prince Osaji Chibuzo to contact  you. He shall handle this
transaction from beginning to the  end on behalf of
the family. Alternatively you can contact
 him on his telephone numbers 234-1-475-14-39 or email
him at:  osaji@katamail.com
 
 Thanks for your cooperation.
 
 Best Regard.
 
 Mohammed Abacha.