
From internet-drafts@ietf.org  Thu Jan  3 09:31:08 2013
Return-Path: <internet-drafts@ietf.org>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B349611E80A4; Thu,  3 Jan 2013 09:31:08 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -100
X-Spam-Level: 
X-Spam-Status: No, score=-100 tagged_above=-999 required=5 tests=[USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wvhWvFnzMAUP; Thu,  3 Jan 2013 09:31:08 -0800 (PST)
Received: from ietfa.amsl.com (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 44CC411E80A5; Thu,  3 Jan 2013 09:31:08 -0800 (PST)
MIME-Version: 1.0
Content-Type: text/plain; charset="utf-8"
Content-Transfer-Encoding: quoted-printable
From: internet-drafts@ietf.org
To: i-d-announce@ietf.org
X-Test-IDTracker: no
X-IETF-IDTracker: 4.37
Message-ID: <20130103173108.20964.83652.idtracker@ietfa.amsl.com>
Date: Thu, 03 Jan 2013 09:31:08 -0800
Cc: mboned@ietf.org
Subject: [MBONED] I-D Action: draft-ietf-mboned-multrans-addr-acquisition-01.txt
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 03 Jan 2013 17:31:08 -0000

A New Internet-Draft is available from the on-line Internet-Drafts director=
ies.
 This draft is a work item of the MBONE Deployment Working Group of the IET=
F.

	Title           : Address Acquisition For Multicast Content When Source an=
d Receiver Support Differing IP Versions
	Author(s)       : Tina Tsou
                          Axel Clauberg
                          Mohamed Boucadair
                          Stig Venaas
                          Qiong Sun
	Filename        : draft-ietf-mboned-multrans-addr-acquisition-01.txt
	Pages           : 10
	Date            : 2013-01-03

Abstract:
   Each IPTV operator has their own arrangements for pre-provisioning
   program information including addresses of the multicast groups
   corresponding to broadcast programs on the subscriber receiver.
   During the transition from IPv4 to IPv6, scenarios can occur where
   the IP version supported by the receiver differs from that supported
   by the source.  This memo examines what has to be done to allow the
   receiver to acquire multicast address information in the version it
   supports in such scenarios.


The IETF datatracker status page for this draft is:
https://datatracker.ietf.org/doc/draft-ietf-mboned-multrans-addr-acquisition

There's also a htmlized version available at:
http://tools.ietf.org/html/draft-ietf-mboned-multrans-addr-acquisition-01

A diff from the previous version is available at:
http://www.ietf.org/rfcdiff?url2=3Ddraft-ietf-mboned-multrans-addr-acquisit=
ion-01


Internet-Drafts are also available by anonymous FTP at:
ftp://ftp.ietf.org/internet-drafts/


From gbumgard@gmail.com  Tue Jan  8 18:26:57 2013
Return-Path: <gbumgard@gmail.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AF47E21F848B; Tue,  8 Jan 2013 18:26:57 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.997
X-Spam-Level: 
X-Spam-Status: No, score=-2.997 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HTML_MESSAGE=0.001, J_CHICKENPOX_62=0.6, NORMAL_HTTP_TO_IP=0.001, RCVD_IN_DNSWL_LOW=-1]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id XhkN9rAXS5kb; Tue,  8 Jan 2013 18:26:54 -0800 (PST)
Received: from mail-bk0-f41.google.com (mail-bk0-f41.google.com [209.85.214.41]) by ietfa.amsl.com (Postfix) with ESMTP id 4CDC621F8487; Tue,  8 Jan 2013 18:26:50 -0800 (PST)
Received: by mail-bk0-f41.google.com with SMTP id jg9so622743bkc.28 for <multiple recipients>; Tue, 08 Jan 2013 18:26:49 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-type; bh=FCUxm2IlQkXFhipTHWDyxvr9pAmRLMwwzL5aqUDdGwU=; b=vtT2oK8vDHyDeC3Hd4EB4WZRyBQORfsteKmbIacaFpCUlJomcCCHlnLryhC7Obpr7i /yzHkRXMzsjTfzXok+uUMI1xlpIf62SsBppkmyB7rsIdPFbWCWH0uUKz9FxKnTC1iOUv KWzrTz9Lu8u0ZXXoJB4aE+/eBWa1Ki993hnGtkXfLTDyGkkSzL58GXFPcTJM8ZoC/HrG Ql/KbiHnfs6JvQ59jpeDXXDMBErzPiCHhV5u4HJoIHxLhJ3yq76m8oyezoSTyroVjajs /uxn3DNDBPDLKirKt6B0fmJGRH8OxOwcD5MTYk7d2QvQ7IhSgZLQTSUMdTw1yZiEt+oM aXfA==
MIME-Version: 1.0
Received: by 10.204.147.78 with SMTP id k14mr32489808bkv.7.1357698409271; Tue, 08 Jan 2013 18:26:49 -0800 (PST)
Received: by 10.204.253.139 with HTTP; Tue, 8 Jan 2013 18:26:49 -0800 (PST)
In-Reply-To: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se>
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se>
Date: Tue, 8 Jan 2013 18:26:49 -0800
Message-ID: <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com>
From: Greg Bumgardner <gbumgard@gmail.com>
To: Magnus Westerlund <magnus.westerlund@ericsson.com>
Content-Type: multipart/alternative; boundary=0015174be2cc65925604d2d1cbab
X-Mailman-Approved-At: Tue, 08 Jan 2013 21:31:42 -0800
Cc: "mboned@ietf.org" <mboned@ietf.org>, "iesg@ietf.org" <iesg@ietf.org>, "tsv-dir@ietf.org" <tsv-dir@ietf.org>
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Jan 2013 02:26:57 -0000

--0015174be2cc65925604d2d1cbab
Content-Type: text/plain; charset=ISO-8859-1

Magnus,

I would like to thank you for providing valuable feedback regarding the AMT
spec.

I would first like to state that many of your concerns and criticisms
regarding the protocol were raised during MBONED group discussions. Many
alternative solutions were identified but ultimately rejected by the group,
primarily due to the reluctance of group members to make significant
changes to the protocol after several years of slow development. My
comments are inlined below...

Thanks,

-g.b.


On Wed, Dec 26, 2012 at 5:01 AM, Magnus Westerlund <
magnus.westerlund@ericsson.com> wrote:

>  Hi,
>
> Sorry for being late with this IETF last call comments. I will partly
> blame the ADs requesting this Transport Directorate review a bit late, the
> other part is all mine and the holidays. Anyway, I do hope you will
> consider these issues and comments as I believe I found some serious ones
> in addition to a number of clarifications that should be made.
>
>
> Significant Issues:
>
> 1. Congestion Control
> This is clearly a tunnel establishment protocol of something that is IP
> traffic. Thus normally the responsibility for congestion control is with
> the tunnelled traffic. However, I would like argue that this does not apply
> in this case due to the nature of the tunnelled traffic, i.e. multicast
> traffic and secondary due to limitations in the tunnel protocol.
>
> Lets start with the second part. This protocol claims to support ASM still
> don't provide a upstream delivery mechanism, i.e. an ASM receiver is not
> capable of sending as it should. This prevents several existing mechanism
> for congestion control that exist in protocols supporting multicast. The
> first is using RTCP for congestion control in ASM [RFC3550], the second is TCP-Friendly
> Multicast Congestion Control [RFC4654] that can be used in the RMT suite
> of protocols, and I know has been implemented in some NORM
> implementation. Thus only strictly receiver based mechanisms, such as Wave
> and Equation Based Rate Control [RFC3738] are available in this context.
>

You are correct, the AMT protocol provides no method for injecting
multicast traffic into a multicast enabled network, and thus does not
provide the functionality required for something like ASM RTP/RTCP
teleconferencing. The protocol is primarily intended for use in receiving
SSM traffic. Because the AMT protocol uses IGMP/MLD messages to establish
reception/subscription state, the protocol MAY be used to subscribe to ASM
broadcasts, but cannot generate them.

I should note that earlier versions of the AMT spec described a method for
broadcasting SSM via a gateway/relay tunnel, but this capability was
removed due to feasibility considerations.


> Secondly, many multicast usages are in fact deployed without any
> congestion control. This based on that the deploying entity controls the
> scope and authorization for requesting multicast delivery. However, does restrictions
> does not apply to AMT delivery of multicast. If the gateway can reach using
> unicast the relay it can be delivered the multicast group from the domain
> the relay is attached to. Thus, this protocol changes the deployment
> restrictions of multicast which many non-congestion controlled delivery
> is based on. Instead the non-congestion controlled traffic can now sent
> over an IP/UDP tunnel over Internet where neither relay nor gateway may
> have any knowledge about the path the traffic may take.
>

The AMT protocol does not attempt to address the issue of congestion
control. The group decided to treat the detection and remediation of
congestion (of encapsulated multicast packets) as an application-layer
responsibility. While we might have attempted to provide some level of QoS
control on tunnel packets by copying ToS/DS bits from inner to outer
headers, this would only occur on a packet-by-packet basis on not on the
tunnel as a whole.


>

> Based on this I would like to see two changes to this protocol
> specification. First a section discussing the issue of congestion control.
>

I'm not sure what, if anything, I might add to address this. I think the
spec makes it pretty clear that the tunneling is provided by best-effort
delivery of UDP packets (perhaps restricted by relay/router configuration).
Perhaps you could suggest some text.


> Secondly, I think this protocol should have an applicability statement
> limiting its deployment to restricted environments where the relay and
> gateway deployers can provide certain resource provisions between the
> entities to avoid the multicast traffic affecting other traffic sharing
> the same bottlenecks in ways not allowed by the network provider.
>

While I understand your concern, this would essentially negate the entire
reason for the protocol, which is to provide multicast transit to
non-multicast connected receivers. However, the AMT spec does not preclude
the implementation/deployment of AMT-enabled gateway devices for which AMT
support may be enabled or disabled (or blocked via a firewall).


>
>
> 2. Security
>
> This protocols is frank with it having limited security features and says
> this is similar to the IGMP and MLD protocols being used. However, I think
> this is a failure to propoerly consider the threat model. If one uses AMT
> over general Internet it will run in a network where the one deploying the
> multicast and the relays no longer control requirements on source address
> verification or possibilities for traffic separation as they can do within
> the domains where multicast currently are deployed. The security
> vulnerabilities in IGMP and MLD are much more contained and controllable in
> a LAN environment where one has chosen to deploy multicast compared to an
> Relay exposing this to the whole Internet. Once more I think there is only
> two choices here.
>
> A) Beef up the security to general Internet threat model, i.e. at a
> minimal provide a real model for gateway authentication using identities,
> not only return routability based verifications.
>
> B) Limit the applicability of AMT to managed environments and make it
> clear that the relay will need to limit which gateways are allowed to
> access the relay based on addressing.
>
> Based on the first significant issue with congestion control I expect that
> there is little meaning to do A) unless also one is willing to beef up AMT
> to provide congestion control. Which I think is not according to the design
> wishes for the protocol designers.
>

The security topic was visited and revisited numerous times over the last
couple of years. The group made the decision to limit the protocol to
simple source address verification and leave it at that. An emphasis was
placed on protecting relays against simple DoS attacks but not gateways.


3. Use of Zero Checksum
>
> The AMT specification enables the use of Zero UDP checksum with IPv6, i.e.
> draft-ietf-6man-udpchecksums-06<http://datatracker.ietf.org/doc/draft-ietf-6man-udpchecksums/>and
> draft-ietf-6man-udpzero-08<http://datatracker.ietf.org/doc/draft-ietf-6man-udpzero/>.
> Nothing against this in principal. However, I have noticed that AMT fails
> to properly address the failure modes of using a zero-checksum. AMT is a
> typical example of a protocol that actually need active verification of
> each tunnel that zero-checksum functions. This as AMT is clearly intended
> to be deployed with its Gateway part in end-hosts and residential network
> devices or routers. This means the tunnel will pass through both firewalls
> and NATs on its path between the relay and the gateway. Unless these
> devices are not upgraded to support zero-checksum in UDP for IPv6 the
> traffic may actually become black holed. The most likely is a simple
> firewall that has a rule for IPv6/UDP which doesn't allow zero checksum as
> it is against RFC 2460. Thus all the Multicast data packets will disappear
> on route in the tunnel. There is no mechanism in the AMT protocol to detect
> this and negotiate with the relay so that it will not use the zero-checksum
> for this tunnel.
>
> This must be addressed as I see it. If not the AMT will be so brittle that
> it can't be used in a large number of its intended deployments.
>

You probably have no idea how often this issue was raised and discussed
within the group. The consensus was that there was no single solution to
the problem. The only workable solution I can suggest is to use an IPv4
tunnel to carry the IPv6 packets. The mechanism for deciding whether (or
if) to use an IPv4 tunnel is considered out of scope.


>
> 4) MTU issues
>
> This document total fails to discuss the issues of MTU blackholing. As the
> IP multicast datagrams as well as the encapsulated IGMP/MLD messages can
> with the added tunnel-overhead result in that the sent packet exceeds the
> MTU of the path, these packets could be black holed. This can potentially
> result in very intermittent transport behavior for the tunnel. Thus, some
> discussion of how do handle the MTU issues in this context should be
> introduced.
>

Yep. This could be a problem that has no easy solution. Not only might the
outer packets be fragmented, but the encapsulated packets might be
fragments. Since the protocol obviously has no control over the size of
source packets, it cannot really avoid the potential need to fragment the
tunnel packets. A gateway can only rely on the underlying UDP/IP stack to
provide the necessary reassembly of the tunnel packets. Once the tunnel
packet has been reassembled, the gateway, host IP stack or application may
still need to reassemble the contents of a sequence of tunnel packets.


>
> I am willing to discuss methods here, but I guess several alternatives
> exists and thus which is most appropriate and the level of AMT support for
> them varies I would like the protocol designers to do a first stab at
> resolving this.
>
> Other Issues
> ======================================================================
>
> A) The table of content on page 2 should include more levels of headings.
> Most likely down to 4 is needed to make the TOC usable for finding content
> in the document.
>
> B) The claimed ASM support
> I would like to better understand how one can claim to support ASM when
> one has no up stream path to inject the ASM group participants traffic.
> When one uses ASM one normally does this for a reason and needing the
> possibility to inject packets into the group. These limitations needs to be
> clarified.
>

See above. The protocol is intended for multicast reception. It makes no
assumptions about application layer use of the multicast traffic. It can
only be used to receive ASM IP traffic.


>
>
> C) Section 4:
>
> This section indicates is its figures, the ones in Section 4.1.1 and
> Section 4.1.3.1 that the Router Mode IGMP/MLD functions are outside of AMT.
> Which based on the requirements in Section 5.3.3.4 is not accurate. That
> implementation must be AMT specific to maintain the AMT tunnel to group
> membership handling
>

Yes. The overview was intended to provide a purely informative description
of the protocol. In some cases, protocol details were described in the
abstract to simplify the text.


>
> D) Not all figures have handles that can be used to reference.
>

I never noticed that before. The XML2RFC tool does not appear to generate
figure numbers if the figure is not assigned an anchor tag. Will be fixed.


>
> E) Section 4.1.5.1:
>
>    Similarly, the selection of a unicast Relay address may be source-
>    dependent, as a relay contacted by a gateway to supply multicast
>    traffic must have native multicast connectivity to the traffic source
>
> I find this statement confusing. There is no support in the protocol for
> including the multicast group(s) which the gateway like to get in the
> discovery phase of the protocol.
>
>
You are correct - the protocol does not provide any means for checking
whether a relay can join a group. The intent was to simply state the
obvious. One might encode a group address within an anycast (or unicast)
address used to locate a relay, but such a approach is out of scope for the
protocol.


> F) AMT Gateway in home router.
> One deployment scenario is that the AMT gateway is deployed in the home
> network router to provide access to multicast groups provided by the ISP.
> However, the startup procedures in this deployment is unclear. The text
> appears to indicate that one can both have a gateway implementation that as
> soon as the router boots it starts doing discovery and requests to have
> Queries to send to its internal local network. Other suggestions appears to
> be to wait until some host actually request to join a group. The protocol
> specification appears to do its best to leave very much flexibility and
> thus produce huge variance in the market.
>

Intentional. Existing implementations use both.

>
> G) Figure 3:
> I find no discussion of Membership Updates to rejoin the groups after the
> tunnel has changed its source address as seen by the relay. This I think
> should have some discussion. Yes, it reasonably clear that you will get
> traffic just by sending new membership updates over the new tunnel.
> However, some discussion of the timing between teardown and this membership
> update should be considered. Figure 3 implies it should be sent after the
> teardown, which I think is correct due to the traffic volumes to the NAT
> most likely causing the path change.
>

The basic problem is that a relay has no way of detecting a change in the
source address associated with a gateway. The group discussed several
methods for detecting source address changes to eliminate the need to
reconstruct tunnel state, but these solutions were rejected to minimize
changes to the long-standing definition of the protocol.


>
> H) Figure in Section 4.2.2.2
> Propose that the external side of the NAT should be marked as the one
> having the "e" addresses.
>

Doesn't it already?


>
> I) Seeing the figure in Section 4.2.2.3 I definitely commented on the
> Address Collision issues. It is made somewhat clearer later on this. But,
> maybe an clearer section 4 sub-section to discuss this general issue that
> multiple left side host can have the same address as other behind other
> tunnel-end-points and thus there is need in the Relay to hide this from
> upstream and accept it and use the tunnel context to track the different
> hosts.
>

I don't believe there is any unique "tunnel context" that might be used to
track individual hosts. It would have been nice to have included nonces or
session IDs in the protocol, but the long standing version of the protocol
did not.

>
> J) Section 4.2.2.3:
>
>  To avoid placing an undue burden on the relay platform, the protocol
>    specifically allows zero-valued UDP checksums on the multicast data
>    messages.  This is not an issue in UDP over IPv4 as the UDP checksum
>    field may be set to zero.  However, this is a problem for UDP over
>    IPv6 as that protocol requires a valid, non-zero checksum in UDP
>    datagrams [RFC2460].  Messages sent over IPv6 with a UDP checksum of
>    zero may fail to reach the gateway.  This is a well known issue for
>    UDP-based tunneling protocols that is described
>    [I-D.ietf-6man-udpzero].  A recommended solution is described in
>    [I-D.ietf-6man-udpchecksums].
> I think this needs reformulating and I don't understand what is intended
> with the last sentence.
>
> I'm pretty sure the wording used here was generated as the end result of
group discussion. I think you might trace this in comments on the MBONED
list. Feel free to suggest an alternative.

> K) Section 5.1.1.
> "Destination UDP Port -  The IANA-assigned AMT port number."
>
>
>
> I find it strange that the protocol is mandating that all traffic is sent
>
> to the IANA assigned port. Why can't the protocol not allow more flexible handling
> of the destination port? I find one single thing in the protocol which prevents
> usage of an other relay listener port. That is that the Relay Advertisement would
>
>
>
>
> need a port field in addition to the address.
>
> Limiting AMT to a specific port allows for firewall control over the
protocol. A secondary benefit is that it allows for automatic dissection in
tools such as Wireshark.

That said, there is nothing to prohibit use of another port number in
restricted/private deployments of AMT. I would also not be surprised to see
use of well-known UDP port numbers in an attempt to deploy AMT in a manner
that would allow gateway hosts to circumvent firewall restrictions.

>
> L) Section 5.1.1.4
> A 32-bit random value generated by the gateway and echoed by the
>    relay in a Relay Advertisement message.
>
> Should the above value make it clear that it preferably should be a
> cryptographically random value as defined in RFC 4086?
>
> If I recall, there was no perceived advantage in using a such a value.
While the spec does indicate that the gateway should generate a random
value, an implementation might use a simple sequence number for the
purposes of logging or debugging.

>
> M) There is lack of specification in Section 5.1.1 of what one does if version
>
>
> The intent in 5.1 is to simply describe message formats. All have version
zero, so...

> is different from 0. This is mentioned in Section 5.3.3.1 but not for gateways
> and not all messages types.
>
>  Added version handling text to gateway section.

>
> N) Section 5.1.4.8:
>
>    The Querier's Query Interval Code (QQIC) field in the general query
>    is used by a relay to specify the time offset a gateway should use to
>    schedule a new three-way handshake to refresh the group membership
>    state within the relay (current time + Query Interval).
> In several places the QQIC and QRV are not made clear that this is defined in
> the external references for MLD and IGMP.
>
> Added references. Result seems a little verbose.

>
> O) Section 5:
> When specifying the bit-fields, please indicate the length of each field in the
>
>
>
>
> text. This is an accessibility question. If you have impaired vision interpreting
> the figures field length correctly can be different.
>
> I made a deliberate decision to leave the field lengths out. I based this
decision on the style used in other protocol RFCs in which fixed message
formats were used.

>
> P) Section 5.2.2.4:
>
>
> This section defined what retransmission parameters that one can potentially
>
>
> configure. However, the section fails to define what the max or min values that
> are acceptable are. Wrongly configured retransmission parameters can have significant
> negative impact on the network by causing bursts or unnecessary traffic.
>
> Largely because there was no consensus as to what values to use - I just
decided to allow an implementation or deployed component to use whatever
values seem appropriate.

>
> Q) Section 5.2.3.3:
> The gateway may continue to receive Multicast Data messages long
>    after the gateway sends a Membership Update message that deletes
>    existing group subscriptions.
>
> What is "long" in the above sentence. Are we talking some known number of seconds,
> a TCP MSL, i.e. 2 min?
>
> I'm not sure that it matters how "long" the interval is. The point is that
the gateway may continue to receive data messages. If a membership update
message fails to reach the relay, that period could equal the group timeout
period (120s?).

>
> R) Section 5.2.3.4.3
>    A gateway MAY retransmit a Relay Discovery message if it does not
>    receive a matching Relay Advertisement message within some timeout
>    period.  If the gateway retransmits the message multiple times, the
>    timeout period SHOULD be adjusted to provide an random exponential
>    back-off.  The RECOMMENDED timeout is a random value in the range
>    [initial_timeout, MIN(initial_timeout * 2^retry_count,
>    maximum_timeout)], with a RECOMMENDED initial_timeout of 1 second and
>    a RECOMMENDED maximum_timeout of 120 seconds (which is the
>    recommended minimum NAT mapping timeout described in [RFC4787]).
>
> I wonder if the above exponential backof is really what is desired. As it randomly
> picks the timeout between initial timout value and the 2^retry_count*initial_timeout it
>
>
>
>
> will be both lower biased and also capable of producing timing intervals that doesn't
> grow. If one desire to have random timeout to avoid some clock synchronization effects
> I think an algorithm that is Td = MIN(initial_timeout * 2^retry_count,
>    maximum_timeout) and where the actual timeout is random*Td and where random is a
> random value from the uniform distribution in the interval [0.5,1.5]. Will both ensure
> that the timout between two retransmissions never is less and on average grows with
>
>
>
>
> a factor two.
>
> This section is also not defining a minimal initial timout value, or any method for
> safely determine a more performant value from a safe initial value. To do a RTT
> measurement using the AMT control messages would require some extensions but
>
>
>
>
> could be a good way of deterimining a better initial value than 0.5 seconds which
> would be my recommendation for a default value.
>
> The retransmission behavior with exponential back-off was a pretty late
addition to the document. I'll have to give this more thought, though I
think it would work as-is.

> S) Section 5.2.3.4.4
>
>    If a gateway executes the relay discovery procedure at the start of
>    each membership update cycle and the relay address returned in the
>    latest Relay Advertisement message differs from the address returned
>    in a previous Relay Advertisement message, then the gateway SHOULD
>    send a Teardown message (if supported) to the old relay address,
>    using information from the last Membership Query message received
>    from that relay, as described in Section 5.2.3.7.  This behavior is
>    illustrated in the following diagram.
>
> This text and the figure after it does not appear to be consistent.
> The figure implies a timer that isn't present in the above. The textual
> description appears sensitive to flapping anycast routing. I think
> the figures indication of some higher timeout before redoing Relay discovery
> appears much more robust.
>
> OK. At some point, I had to trade off consistency for legibility. There
are several examples throughout the document where the text and figures
might not be 100% consistent. I have to assume that the document has
provided the context required to interpret the text and figures, otherwise
I could spend months attempting to describe every scenario in the text and
ASCII art. I tried my best.

> T) Section 5.2.3.5.3
> See R)
> Also this text appears redundant to previous text. Maybe generalize this
>
>
>
>
> into its own section being used in general for all messages needing retransmission
>
>
Good idea, but would require more changes than I feel comfortable making at
this point.

>
> U) Section 5.2.3.5.4
>    Querier's Query Interval Code carried by the general-query.  A
>    gateway MAY use a smaller timer duration if required to refresh a NAT
>    mapping that would otherwise timeout.
>
> Maybe the protocol would rather need a NAT keep-alive message to be sent
> from the gateway to the relay. But maybe the Request, Query cycle is light weight
> enough that this works fine.
>
> Yes.

>
>
> V) Section 5.2.3.6.1
>
> o  Insert IGMP or MLD datagrams into a queue for transmission after
>       it receives a Membership Query message.
>
> What assumptions of queue depth exist in the above. Clearly the messages in this
> queue should expire if they become to old.
>
> No assumptions made. I don't even like the idea of queuing IGMP/MLD
messages for eventual delivery. I basically treated the queuing behavior as
out of scope because I didn't want to complicate things any further.

>
> X) Section 5.2.3.7
> Gateway support for the Teardown message is OPTIONAL but RECOMMENDED.
>
>
>
>
> The above is a very strange usage of RFC 2119 keywords. IF you use the synonyms
> then maybe the error of writing it this way is clear.
> Gateway MAY support for the Teardown message but SHOULD.
>
> Will be fixed.

> Y) The usage of retransmission versus repetitions are not always clear.
>
>
>
>
> Some of the messages appears to simply need to be repeated QRV number of times with
> some interval. Others should really be matched with an answer and if not received within
> timeout retransmitted. Can these two cases be made more clear?
>
>
>
>
> Z) Section 5.3.5
>    The hash function RECOMMENDED for use in computing the Response MAC
>    is the MD5 hash digest [RFC1321], though hash functions or keyed-hash
>    functions of greater cryptographic strength may be used.
>
> I think this points to a security vulnerability. I think it needs to be made clear
> that the MAC MUST be keyed. If it is just a digest, then an attacker can calculate the
> MAC and perform an off-path attack.
>
> I don't think it matters whether one computes a hash value that includes a
private secret, or uses a private secret to compute a "keyed" hash. The end
result is the same.

>
> This should be made clear also in Section 6.1 to be a requirement.
>
> AA) Section A.1:
>
> Altough this proposals has its advantages I think it might also illustrate a
>
>
>
> short-coming. First of all 48-bits is quite short for a MAC. I would prefer a
> variable length field.
>
> I agree. Unfortunately, preliminary implementations of the protocol have
been around for so long that group members did not want to change the MAC
size. That said, one negative side-effect of using a larger MAC value would
be increased fragmentation of data message packets.

>
> Secondly, doesn't this actually create more material for an attacker to determine
>
> the key used by the relay?
>
>
>
> That was all I have found.
>
>
Thanks!



>
> Cheers
>
> Magnus
>
>
> _______________________________________________
> MBONED mailing list
> MBONED@ietf.org
> https://www.ietf.org/mailman/listinfo/mboned
>
>


-- 
Greg Bumgardner
Eugene, OR

--0015174be2cc65925604d2d1cbab
Content-Type: text/html; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable

Magnus,<div><br></div><div>I would like to thank you for providing valuable=
 feedback regarding the AMT spec.=A0</div><div><br></div><div>I would first=
 like to state that many of your concerns and criticisms regarding the prot=
ocol were raised during MBONED group discussions. Many alternative solution=
s were identified but ultimately rejected by the group, primarily due to th=
e reluctance of group members to make significant changes to the protocol a=
fter several years of slow development.=A0My comments are inlined below...<=
/div>
<div><br></div><div>Thanks,</div><div><br></div><div>-g.b.</div><div><br></=
div><div><br><div class=3D"gmail_quote">On Wed, Dec 26, 2012 at 5:01 AM, Ma=
gnus Westerlund <span dir=3D"ltr">&lt;<a href=3D"mailto:magnus.westerlund@e=
ricsson.com" target=3D"_blank">magnus.westerlund@ericsson.com</a>&gt;</span=
> wrote:<br>




<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex">




<div>
<div style=3D"direction:ltr;font-size:10pt;font-family:Tahoma"><font face=
=3D"Courier New">Hi,<br>
<br>
Sorry for being late with this IETF last call comments. I will partly blame=
 the ADs requesting this Transport Directorate review a bit late, the other=
 part is all mine and the holidays. Anyway, I do hope you will consider the=
se issues and comments as I believe
 I found some serious ones in addition to a number of clarifications that s=
hould be made.<br>
<br>
<br>
Significant Issues:<br>
<br>
1. Congestion Control<br>
This is clearly a tunnel establishment protocol of something that is IP tra=
ffic. Thus normally the responsibility for congestion control is with the t=
unnelled traffic. However, I would like argue that this does not apply in t=
his case due to the nature of the
 tunnelled traffic, i.e. multicast traffic and secondary due to limitations=
 in the tunnel protocol.
<br>
<br>
Lets start with the second part. This protocol claims to support ASM still =
don&#39;t provide a upstream delivery mechanism, i.e. an ASM receiver is no=
t capable of sending as it should. This prevents several existing mechanism=
 for congestion control that exist in
 protocols supporting multicast. The first is using RTCP for congestion con=
trol in ASM [RFC3550], the second is</font><font face=3D"Courier New"><font=
 size=3D"3">
</font>TCP-Friendly Multicast Congestion Control [RFC4654] that can be used=
 in <font>
the RMT s<font>uite of protocols, and I know <font>has been implemented in =
some NORM implementation. Thus only<font> strictly receiver based
<font>mechanisms, such as </font></font></font></font></font></font><font f=
ace=3D"Courier New"><font><font><font><font><font>Wave and Equation Based R=
ate Control [RFC3738]</font> are available in
 this context. <br></font></font></font></font></font></div></div></blockqu=
ote><div><br></div><div>You are correct, the AMT protocol provides no metho=
d for injecting multicast traffic into a multicast enabled network, and thu=
s does not provide the functionality required for something like ASM RTP/RT=
CP teleconferencing. The protocol is primarily intended for use in receivin=
g SSM traffic. Because the AMT protocol uses IGMP/MLD messages to establish=
 reception/subscription state, the protocol MAY be used to subscribe to ASM=
 broadcasts, but cannot generate them.</div>




<div><br></div><div>I should note that earlier versions of the AMT spec des=
cribed a method for broadcasting SSM via a gateway/relay tunnel, but this c=
apability was removed due to feasibility considerations.</div><div><br>




</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-l=
eft:1px #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-=
size:10pt;font-family:Tahoma"><font face=3D"Courier New"><font><font><font>=
<font>
<br>
<font>Secondly, <font>many multicast <font>usages are in fact deployed
<font>without any congestion control. This based on that the deploy<font>in=
g entity controls
<font>the scope <font>and <font><font>authorization</font> for
<font>requesting multicast d<font>elivery. However, <font>
does <font>restrictions d<font>oes not apply to AMT delivery of mu<font>lti=
cast. I<font>f the gat<font>e<font>way can reach
<font>using unicast the relay it can be delivered the multicast group from =
the domain the relay is attached to. Thus<font>, th<font>is protocol change=
s the deployment restrictions
<font>of multic<font>ast which many non-congestion controlled delivery is b=
ased on. Instead the
<font>non-congestion controlled traffic<font> can</font> now sen<font>t ove=
r an IP/UDP tunnel over Internet</font></font></font></font></font></font><=
/font></font></font></font></font></font></font></font></font></font></font=
></font></font></font></font></font></font></font></font></font></font></fo=
nt></font>
 w<font>here neither r<font>elay nor gateway may have any knowledge about t=
he pat<font>h the traffic may take.=A0</font></font></font></div></div></bl=
ockquote><div><br></div><div>The AMT protocol does not attempt to address t=
he issue of congestion control. The group decided to treat the detection an=
d remediation of congestion (of encapsulated multicast packets) as an appli=
cation-layer responsibility. While we might have attempted to provide some =
level of QoS control on tunnel packets by copying ToS/DS bits from inner to=
 outer headers, this would only occur on a packet-by-packet basis on not on=
 the tunnel as a whole.</div>



<div><br></div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex=
;border-left:1px #ccc solid;padding-left:1ex"><div><div style=3D"direction:=
ltr;font-size:10pt;font-family:Tahoma">=A0</div></div></blockquote><blockqu=
ote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc s=
olid;padding-left:1ex">



<div><div style=3D"direction:ltr;font-size:10pt;font-family:Tahoma"><font><=
font><font>
<br>
<font><font><font>Based on this I would <font>
like to see two <font>changes</font> to this protocol specification. First =
a section discussing the issue of congestion control.</font></font></font><=
/font></font></font></font></div></div></blockquote><div><br></div><div>



I&#39;m not sure what, if anything, I might add to address this. I think th=
e spec makes it pretty clear that the tunneling is provided by best-effort =
delivery of UDP packets (perhaps restricted by relay/router configuration).=
 Perhaps you could suggest some text.=A0</div>



<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex"><div><div style=3D"direction:l=
tr;font-size:10pt;font-family:Tahoma"><font><font><font><font><font><font><=
font><font>Secondly, I think this protocol shou<font>ld have an applicabili=
ty statement limiting its deployment
<font>to restricted <font>environments</font> where the relay and gateway d=
eploy<font>ers can provide
<font>certain resource provisions between the <font>entiti<font>es to avoid=
 the m<font>ulticast traffic affecting other traffic sharing the same bottl=
enecks<font> in ways not allowed by the network
<font>provider.</font></font></font></font></font></font></font></font></fo=
nt></font></font></font></font></font></font></font></font></div></div></bl=
ockquote><div><br></div><div>While I understand your concern, this would es=
sentially negate the entire reason for the protocol, which is to provide mu=
lticast transit to non-multicast connected receivers. However, the AMT spec=
 does not preclude the implementation/deployment of AMT-enabled gateway dev=
ices for which AMT support may be enabled or disabled (or blocked via a fir=
ewall).=A0</div>



<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex">
<div><div style=3D"direction:ltr;font-size:10pt;font-family:Tahoma"><font><=
font><font><br>
</font></font></font><font face=3D"Courier New"><br>
2. Security<br>
<br>
This protocols is frank with it having limited security features and says t=
his is similar to the IGMP and MLD protocols being used. However, I think t=
his is a failure to propoerly consider the threat model. If one uses AMT ov=
er general Internet it will run
 in a network where the one deploying the multicast and the relays no longe=
r control requirements on source address verification or possibilities for =
traffic separation as they can do within the domains where multicast curren=
tly are deployed. The security vulnerabilities
 in IGMP and MLD are much more contained and controllable in a LAN environm=
ent where one has chosen to deploy multicast compared to an Relay exposing =
this to the whole Internet. Once more I think there is only two choices her=
e.<br>





<br>
A) Beef up the security to general Internet threat model, i.e. at a minimal=
 provide a real model for gateway authentication using identities, not only=
 return routability based verifications.
<br>
<br>
B) Limit the applicability of AMT to managed environments and make it clear=
 that the relay will need to limit which gateways are allowed to access the=
 relay based on addressing.
<br>
<br>
Based on the first significant issue with congestion control I expect that =
there is little meaning to do A) unless also one is willing to beef up AMT =
to provide congestion control. Which I think is not according to the design=
 wishes for the protocol designers.<br>



</font></div></div></blockquote><div><br></div><div>The security topic was =
visited and revisited numerous times over the last couple of years. The gro=
up made the=A0decision to limit the protocol to simple source address verif=
ication and leave it at that. An emphasis was placed on protecting relays a=
gainst simple DoS attacks but not gateways.</div>



<div><br></div><div><br></div><blockquote class=3D"gmail_quote" style=3D"ma=
rgin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><div styl=
e=3D"direction:ltr;font-size:10pt;font-family:Tahoma"><font face=3D"Courier=
 New">
3. Use of Zero Checksum <br>
<br>
The AMT specification enables the use of Zero UDP checksum with IPv6, i.e. =
</font>
<br>
<font face=3D"Courier New"><a href=3D"http://datatracker.ietf.org/doc/draft=
-ietf-6man-udpchecksums/" target=3D"_blank">draft-ietf-6man-udpchecksums-06=
</a> and
</font><font face=3D"Courier New"><a href=3D"http://datatracker.ietf.org/do=
c/draft-ietf-6man-udpzero/" target=3D"_blank">draft-ietf-6man-udpzero-08</a=
>. Nothing against this in principal. However, I have noticed that AMT fail=
s to properly address the failure modes of using a zero-checksum.
 AMT is a typical example of a protocol that actually need active verificat=
ion of each tunnel that zero-checksum functions. This as AMT is clearly int=
ended to be deployed with its Gateway part in end-hosts and residential net=
work devices or routers. This means
 the tunnel will pass through both firewalls and NATs on its path between t=
he relay and the gateway. Unless these devices are not upgraded to support =
zero-checksum in UDP for IPv6 the traffic may actually become black holed. =
The most likely is a simple firewall
 that has a rule for IPv6/UDP which doesn&#39;t allow zero checksum as it i=
s against RFC 2460. Thus all the Multicast data packets will disappear on r=
oute in the tunnel. There is no mechanism in the AMT protocol to detect thi=
s and negotiate with the relay so that
 it will not use the zero-checksum for this tunnel. <br>
<br>
This must be addressed as I see it. If not the AMT will be so brittle that =
it can&#39;t be used in a large number of its intended deployments.
<br></font></div></div></blockquote><div><br></div><div>You probably have n=
o idea how often this issue was raised and discussed within the group. The=
=A0consensus was that there was no single solution to the problem. The only=
 workable solution I can suggest is to use an IPv4 tunnel to carry the IPv6=
 packets. The mechanism for deciding whether (or if) to use an IPv4 tunnel =
is considered out of scope.</div>



<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex"><div><div style=3D"direction:l=
tr;font-size:10pt;font-family:Tahoma"><font face=3D"Courier New">
<br>
4) MTU issues<br>
<br>
This document total fails to discuss the issues of MTU blackholing. As the =
IP multicast datagrams as well as the encapsulated IGMP/MLD messages can wi=
th the added tunnel-overhead result in that the sent packet exceeds the MTU=
 of the path, these packets could
 be black holed. This can potentially result in very intermittent transport=
 behavior for the tunnel. Thus, some discussion of how do handle the MTU is=
sues in this context should be introduced.
<br></font></div></div></blockquote><div><br></div><div>Yep. This could be =
a problem that has no easy solution. Not only might the outer packets be fr=
agmented, but the encapsulated packets might be fragments. Since the protoc=
ol obviously has no control over the size of source packets, it cannot real=
ly avoid the potential need to fragment the tunnel packets. A gateway can o=
nly rely on the underlying UDP/IP stack to provide the necessary reassembly=
 of the tunnel packets. Once the tunnel packet has been reassembled, the ga=
teway, host IP stack or application may still need to reassemble the conten=
ts of a sequence of tunnel packets.</div>



<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex"><div><div style=3D"direction:l=
tr;font-size:10pt;font-family:Tahoma"><font face=3D"Courier New">
<br>
I am willing to discuss methods here, but I guess several alternatives exis=
ts and thus which is most appropriate and the level of AMT support for them=
 varies I would like the protocol designers to do a first stab at resolving=
 this.
<br>
<br>
Other Issues<br>
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D<br>
<br>
A) The table of content on page 2 should include more levels of headings. M=
ost likely down to 4 is needed to make the TOC usable for finding content i=
n the document.<br>
<br>
B) </font><font face=3D"Courier New"><font face=3D"Courier New">The claimed=
 ASM support<br>
I would like to better understand how one can claim to support ASM when one=
 has no up stream path to inject the ASM group participants traffic. When o=
ne uses ASM one normally does this for a reason and needing the possibility=
 to inject packets into the group.
 These limitations needs to be clarified. </font></font></div></div></block=
quote><div><br></div><div>See above. The protocol is intended for multicast=
 reception. It makes no assumptions about application layer use of the mult=
icast traffic. It can only be used to receive ASM IP traffic.</div>



<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex"><div><div style=3D"direction:l=
tr;font-size:10pt;font-family:Tahoma"><font face=3D"Courier New"><br>
<br>
C) Section 4:<br>
<br>
This section indicates is its figures, the ones in Section 4.1.1 and Sectio=
n 4.1.3.1 that the Router Mode IGMP/MLD functions are outside of AMT. Which=
 based on the requirements in Section 5.3.3.4 is not accurate. That impleme=
ntation must be AMT specific to
 maintain the AMT tunnel to group membership handling<br></font></div></div=
></blockquote><div><br></div><div>Yes. The overview was intended to provide=
 a purely informative description of the protocol. In some cases, protocol =
details were described in the abstract to simplify the text.</div>



<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex"><div><div style=3D"direction:l=
tr;font-size:10pt;font-family:Tahoma"><font face=3D"Courier New">
<br>
D) Not all figures have handles that can be used to reference.<br></font></=
div></div></blockquote><div><br></div><div>I never noticed that before. The=
 XML2RFC tool does not appear to generate figure numbers if the figure is n=
ot assigned an anchor tag. Will be fixed.</div>



<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex"><div><div style=3D"direction:l=
tr;font-size:10pt;font-family:Tahoma"><font face=3D"Courier New">
<br>
E) Section <a href=3D"http://4.1.5.1" target=3D"_blank">4.1.5.1</a>: <br>
</font><br>
<pre>   Similarly, the selection of a unicast Relay address may be source-
   dependent, as a relay contacted by a gateway to supply multicast
   traffic must have native multicast connectivity to the traffic source
</pre>
<font face=3D"Courier New">I find this statement confusing. There is no sup=
port in the protocol for including the multicast group(s) which the gateway=
 like to get in the discovery phase of the protocol.<br>
<br></font></div></div></blockquote><div><br></div><div>You are correct - t=
he protocol does not provide any means for checking whether a relay can joi=
n a group. The intent was to simply state the obvious. One might encode a g=
roup address within an anycast (or unicast) address used to locate a relay,=
 but such a approach is out of scope for the protocol. =A0</div>



<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex"><div><div style=3D"direction:l=
tr;font-size:10pt;font-family:Tahoma"><font face=3D"Courier New">
F) AMT Gateway in home router.<br>
One deployment scenario is that the AMT gateway is deployed in the home net=
work router to provide access to multicast groups provided by the ISP. Howe=
ver, the startup procedures in this deployment is unclear. The text appears=
 to indicate that one can both have
 a gateway implementation that as soon as the router boots it starts doing =
discovery and requests to have Queries to send to its internal local networ=
k. Other suggestions appears to be to wait until some host actually request=
 to join a group. The protocol specification
 appears to do its best to leave very much flexibility and thus produce hug=
e variance in the market.
<br></font></div></div></blockquote><div><br></div><div>Intentional. Existi=
ng implementations use both.</div><blockquote class=3D"gmail_quote" style=
=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex"><div><di=
v style=3D"direction:ltr;font-size:10pt;font-family:Tahoma">



<font face=3D"Courier New">
<br>
G) Figure 3:<br>
I find no discussion of Membership Updates to rejoin the groups after the t=
unnel has changed its source address as seen by the relay. This I think sho=
uld have some discussion. Yes, it reasonably clear that you will get traffi=
c just by sending new membership
 updates over the new tunnel. However, some discussion of the timing betwee=
n teardown and this membership update should be considered. Figure 3 implie=
s it should be sent after the teardown, which I think is correct due to the=
 traffic volumes to the NAT most
 likely causing the path change. <br></font></div></div></blockquote><div><=
br></div><div>The basic problem is that a relay has no way of detecting a c=
hange in the source address associated with a gateway. The group discussed =
several methods for detecting source address changes to eliminate the need =
to reconstruct tunnel state, but these solutions were rejected to minimize =
changes to the long-standing definition of the protocol.</div>



<div>=A0</div><blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;=
border-left:1px #ccc solid;padding-left:1ex"><div><div style=3D"direction:l=
tr;font-size:10pt;font-family:Tahoma"><font face=3D"Courier New">
<br>
H) Figure in Section 4.2.2.2<br>
Propose that the external side of the NAT should be marked as the one havin=
g the &quot;e&quot; addresses.<br></font></div></div></blockquote><div><br>=
</div><div>Doesn&#39;t it already?</div><div>=A0</div><blockquote class=3D"=
gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-=
left:1ex">



<div><div style=3D"direction:ltr;font-size:10pt;font-family:Tahoma"><font f=
ace=3D"Courier New">
<br>
I) Seeing the figure in Section 4.2.2.3 I definitely commented on the Addre=
ss Collision issues. It is made somewhat clearer later on this. But, maybe =
an clearer section 4 sub-section to discuss this general issue that multipl=
e left side host can have the same
 address as other behind other tunnel-end-points and thus there is need in =
the Relay to hide this from upstream and accept it and use the tunnel conte=
xt to track the different hosts.
<br></font></div></div></blockquote><div><br></div><div>I don&#39;t believe=
 there is any unique &quot;tunnel context&quot; that might be used to track=
 individual hosts. It would have been nice to have included nonces or sessi=
on IDs in the protocol, but the long standing version of the protocol did n=
ot. =A0</div>



<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><font face=3D"Courier New">
<br>
J) Section <a href=3D"http://4.2.2.3" target=3D"_blank">4.2.2.3</a>:<br>
</font><br>
<pre> To avoid placing an undue burden on the relay platform, the protocol
   specifically allows zero-valued UDP checksums on the multicast data
   messages.  This is not an issue in UDP over IPv4 as the UDP checksum
   field may be set to zero.  However, this is a problem for UDP over
   IPv6 as that protocol requires a valid, non-zero checksum in UDP
   datagrams [RFC2460].  Messages sent over IPv6 with a UDP checksum of
   zero may fail to reach the gateway.  This is a well known issue for
   UDP-based tunneling protocols that is described
   [I-D.ietf-6man-udpzero].  A recommended solution is described in
   [I-D.ietf-6man-udpchecksums].
<br>I think this needs reformulating and I don&#39;t understand what is int=
ended <br>with the last sentence. <br><br></pre></div></div></blockquote><d=
iv>I&#39;m pretty sure the wording used here was generated as the end resul=
t of group discussion. I think you might trace this in comments on the MBON=
ED list. Feel free to suggest an alternative.</div>



<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre>K) Section 5.1.1.<br>&quot;Destination UDP Por=
t -  The IANA-assigned AMT port number.&quot;<br>



<br>I find it strange that the protocol is mandating that all traffic is se=
nt<br>
to the IANA assigned port. Why can&#39;t the protocol not allow more flexib=
le handling<br>of the destination port? I find one single thing in the prot=
ocol which prevents <br>usage of an other relay listener port. That is that=
 the Relay Advertisement would <br>




need a port field in addition to the address.<br></pre></div></div></blockq=
uote><div>Limiting AMT to a specific port allows for firewall control over =
the protocol. A secondary benefit is that it allows for automatic dissectio=
n in tools such as Wireshark.</div>



<div><br></div><div>That said, there is nothing to prohibit use of another =
port number in restricted/private deployments of AMT. I would also not be s=
urprised to see use of well-known UDP port numbers in an attempt to deploy =
AMT in a manner that would allow gateway hosts to circumvent firewall restr=
ictions.</div>



<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre><br>L) Section 5.1.1.4<br>A 32-bit random valu=
e generated by the gateway and echoed by the
   relay in a Relay Advertisement message. <br><br>Should the above value m=
ake it clear that it preferably should be a <br>cryptographically random va=
lue as defined in RFC 4086?<br></pre></div></div></blockquote><div>If I rec=
all, there was no perceived advantage in using a such a value. While the sp=
ec does indicate that the gateway should generate a random value, an implem=
entation might use a simple sequence number for the purposes of logging or =
debugging.</div>


<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre><br>M) There is lack of specification in Secti=
on 5.1.1 of what one does if version <br>



</pre></div></div></blockquote><div>The intent in 5.1 is to simply describe=
 message formats. All have version zero, so...</div><blockquote class=3D"gm=
ail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-le=
ft:1ex">


<div><div style=3D"direction:ltr;font-size:10pt;font-family:Tahoma"><pre>is=
 different from 0. This is mentioned in Section 5.3.3.1 but not for gateway=
s <br>and not all messages types. <br></pre></div></div></blockquote><div>


=A0Added version handling text to gateway section.</div><blockquote class=
=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padd=
ing-left:1ex"><div><div style=3D"direction:ltr;font-size:10pt;font-family:T=
ahoma">


<pre><br>N) Section <a href=3D"http://5.1.4.8" target=3D"_blank">5.1.4.8</a=
>:<br><br>   The Querier&#39;s Query Interval Code (QQIC) field in the gene=
ral query
   is used by a relay to specify the time offset a gateway should use to
   schedule a new three-way handshake to refresh the group membership
   state within the relay (current time + Query Interval).
<br>In several places the QQIC and QRV are not made clear that this is defi=
ned in <br>the external references for MLD and IGMP. <br></pre></div></div>=
</blockquote><div>Added references. Result seems a little verbose.</div>


<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre><br>O) Section 5:<br>When specifying the bit-f=
ields, please indicate the length of each field in the <br>




text. This is an accessibility question. If you have impaired vision interp=
reting<br>the figures field length correctly can be different. <br></pre></=
div></div></blockquote><div>I made a deliberate decision to leave the field=
 lengths out. I based this decision on the style used in other protocol RFC=
s in which fixed message formats were used.=A0</div>


<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre><br>P) Section <a href=3D"http://5.2.2.4" targ=
et=3D"_blank">5.2.2.4</a>:<br>


<br>This section defined what retransmission parameters that one can potent=
ially <br>

configure. However, the section fails to define what the max or min values =
that <br>are acceptable are. Wrongly configured retransmission parameters c=
an have significant<br>negative impact on the network by causing bursts or =
unnecessary traffic. <br>


</pre></div></div></blockquote><div>Largely because there was no=A0consensu=
s as to what values to use - I just decided to allow an implementation or d=
eployed component to use whatever values seem appropriate.=A0=A0</div><bloc=
kquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #cc=
c solid;padding-left:1ex">


<div><div style=3D"direction:ltr;font-size:10pt;font-family:Tahoma"><pre><b=
r>Q) Section <a href=3D"http://5.2.3.3" target=3D"_blank">5.2.3.3</a>:<br>T=
he gateway may continue to receive Multicast Data messages long
   after the gateway sends a Membership Update message that deletes
   existing group subscriptions.<br><br>What is &quot;long&quot; in the abo=
ve sentence. Are we talking some known number of seconds, <br>a TCP MSL, i.=
e. 2 min?<br></pre></div></div></blockquote><div>I&#39;m not sure that it m=
atters how &quot;long&quot; the interval is. The point is that the gateway =
may continue to receive data messages. If a membership update message fails=
 to reach the relay, that period could equal the group timeout period (120s=
?).=A0</div>


<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre><br>R) Section 5.2.3.4.3<br>   A gateway MAY r=
etransmit a Relay Discovery message if it does not
   receive a matching Relay Advertisement message within some timeout
   period.  If the gateway retransmits the message multiple times, the
   timeout period SHOULD be adjusted to provide an random exponential
   back-off.  The RECOMMENDED timeout is a random value in the range
   [initial_timeout, MIN(initial_timeout * 2^retry_count,
   maximum_timeout)], with a RECOMMENDED initial_timeout of 1 second and
   a RECOMMENDED maximum_timeout of 120 seconds (which is the
   recommended minimum NAT mapping timeout described in [RFC4787]).<br><br>=
I wonder if the above exponential backof is really what is desired. As it r=
andomly<br>picks the timeout between initial timout value and the 2^retry_c=
ount*initial_timeout it <br>




will be both lower biased and also capable of producing timing intervals th=
at doesn&#39;t <br>grow. If one desire to have random timeout to avoid some=
 clock synchronization effects<br>I think an algorithm that is Td =3D MIN(i=
nitial_timeout * 2^retry_count,
   maximum_timeout) and where the actual timeout is random*Td and where ran=
dom is a <br>random value from the uniform distribution in the interval [0.=
5,1.5]. Will both ensure<br>that the timout between two retransmissions nev=
er is less and on average grows with <br>




a factor two. <br><br>This section is also not defining a minimal initial t=
imout value, or any method for <br>safely determine a more performant value=
 from a safe initial value. To do a RTT <br>measurement using the AMT contr=
ol messages would require some extensions but<br>




could be a good way of deterimining a better initial value than 0.5 seconds=
 which <br>would be my recommendation for a default value. <br><br></pre></=
div></div></blockquote><div>The retransmission behavior with exponential ba=
ck-off was a pretty late addition to the document. I&#39;ll have to give th=
is more thought, though I think it would work as-is.</div>


<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre>S) Section 5.2.3.4.4<br><br>   If a gateway ex=
ecutes the relay discovery procedure at the start of
   each membership update cycle and the relay address returned in the
   latest Relay Advertisement message differs from the address returned
   in a previous Relay Advertisement message, then the gateway SHOULD
   send a Teardown message (if supported) to the old relay address,
   using information from the last Membership Query message received
   from that relay, as described in Section 5.2.3.7.  This behavior is
   illustrated in the following diagram. <br><br>This text and the figure a=
fter it does not appear to be consistent.<br>The figure implies a timer tha=
t isn&#39;t present in the above. The textual <br>description appears sensi=
tive to flapping anycast routing. I think
the figures indication of some higher timeout before redoing Relay discover=
y <br>appears much more robust. <br><br></pre></div></div></blockquote><div=
>OK. At some point, I had to trade off consistency for legibility. There ar=
e several examples throughout the document where the text and figures might=
 not be 100% consistent. I have to assume that the document has provided th=
e context required to interpret the text and figures, otherwise I could spe=
nd months attempting to describe every scenario in the text and ASCII art. =
I tried my best.</div>


<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre>T) Section 5.2.3.5.3<br>See R)<br>Also this te=
xt appears redundant to previous text. Maybe generalize this<br>




into its own section being used in general for all messages needing retrans=
mission<br></pre></div></div></blockquote><div><br></div><div>Good idea, bu=
t would require more changes than I feel comfortable making at this point.=
=A0</div>


<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre><br>U) Section 5.2.3.5.4<br>   Querier&#39;s Q=
uery Interval Code carried by the general-query.  A
   gateway MAY use a smaller timer duration if required to refresh a NAT
   mapping that would otherwise timeout. <br><br>Maybe the protocol would r=
ather need a NAT keep-alive message to be sent<br>from the gateway to the r=
elay. But maybe the Request, Query cycle is light weight <br>enough that th=
is works fine. <br>


</pre></div></div></blockquote><div>Yes.=A0</div><blockquote class=3D"gmail=
_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:=
1ex"><div><div style=3D"direction:ltr;font-size:10pt;font-family:Tahoma"><p=
re>
<br>
V) Section 5.2.3.6.1<br><br>o  Insert IGMP or MLD datagrams into a queue fo=
r transmission after
      it receives a Membership Query message.<br><br>What assumptions of qu=
eue depth exist in the above. Clearly the messages in this <br>queue should=
 expire if they become to old. <br></pre></div></div></blockquote><div>


No assumptions made. I don&#39;t even like the idea of queuing IGMP/MLD mes=
sages for eventual delivery. I basically treated the queuing behavior as ou=
t of scope because I didn&#39;t want to complicate things any further.=A0</=
div>


<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre><br>X) Section 5.2.3.7<br>Gateway support for =
the Teardown message is OPTIONAL but RECOMMENDED.<br>




<br>The above is a very strange usage of RFC 2119 keywords. IF you use the =
synonyms<br>then maybe the error of writing it this way is clear. <br>Gatew=
ay MAY support for the Teardown message but SHOULD. <br><br></pre></div>

</div></blockquote><div>Will be fixed.=A0</div><blockquote class=3D"gmail_q=
uote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1e=
x"><div><div style=3D"direction:ltr;font-size:10pt;font-family:Tahoma"><pre=
>Y) The usage of retransmission versus repetitions are not always clear.<br=
>




Some of the messages appears to simply need to be repeated QRV number of ti=
mes with <br>some interval. Others should really be matched with an answer =
and if not received within<br>timeout retransmitted. Can these two cases be=
 made more clear?<br>




<br>Z) Section 5.3.5<br>   The hash function RECOMMENDED for use in computi=
ng the Response MAC
   is the MD5 hash digest [RFC1321], though hash functions or keyed-hash
   functions of greater cryptographic strength may be used.<br>=A0<br>I thi=
nk this points to a security vulnerability. I think it needs to be made cle=
ar <br>that the MAC MUST be keyed. If it is just a digest, then an attacker=
 can calculate the=20
MAC and perform an off-path attack. <br></pre></div></div></blockquote><div=
>I don&#39;t think it matters whether one computes a hash value that includ=
es a private secret, or uses a private secret to compute a &quot;keyed&quot=
; hash. The end result is the same. =A0</div>

<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre><br>This should be made clear also in Section =
6.1 to be a requirement.<br>

<br>AA) Section A.1:<br><br>Altough this proposals has its advantages I thi=
nk it might also illustrate a <br>


short-coming. First of all 48-bits is quite short for a MAC. I would prefer=
 a <br>variable length field.<br></pre></div></div></blockquote><div>I agre=
e. Unfortunately, preliminary implementations of the protocol have been aro=
und for so long that group members did not want to change the MAC size. Tha=
t said, one negative side-effect of using a larger MAC value would be incre=
ased fragmentation of data message packets.</div>

<blockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1p=
x #ccc solid;padding-left:1ex"><div><div style=3D"direction:ltr;font-size:1=
0pt;font-family:Tahoma"><pre><br>Secondly, doesn&#39;t this actually create=
 more material for an attacker to determine<br>

the key used by the relay?<br>


<br>That was all I have found. <br></pre></div></div></blockquote><div><br>=
</div><div>Thanks!</div><div><br></div><div>=A0</div><blockquote class=3D"g=
mail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-l=
eft:1ex">

<div><div style=3D"direction:ltr;font-size:10pt;font-family:Tahoma"><pre><b=
r>Cheers<span><font color=3D"#888888"><br><br>Magnus<br><br></font></span><=
/pre>
</div>
</div>

<br>_______________________________________________<br>
MBONED mailing list<br>
<a href=3D"mailto:MBONED@ietf.org" target=3D"_blank">MBONED@ietf.org</a><br=
>
<a href=3D"https://www.ietf.org/mailman/listinfo/mboned" target=3D"_blank">=
https://www.ietf.org/mailman/listinfo/mboned</a><br>
<br></blockquote></div><br><br clear=3D"all"><div><br></div>-- <br>Greg Bum=
gardner<br>Eugene, OR<br>
</div>

--0015174be2cc65925604d2d1cbab--

From Martin.Stiemerling@neclab.eu  Wed Jan  9 06:16:10 2013
Return-Path: <Martin.Stiemerling@neclab.eu>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 115ED21F86C8; Wed,  9 Jan 2013 06:16:10 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.349
X-Spam-Level: 
X-Spam-Status: No, score=-103.349 tagged_above=-999 required=5 tests=[AWL=0.250, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id am0OpMVLYwcl; Wed,  9 Jan 2013 06:16:08 -0800 (PST)
Received: from mailer1.neclab.eu (mailer1.neclab.eu [195.37.70.40]) by ietfa.amsl.com (Postfix) with ESMTP id 4288821F85D2; Wed,  9 Jan 2013 06:16:08 -0800 (PST)
Received: from localhost (localhost [127.0.0.1]) by mailer1.neclab.eu (Postfix) with ESMTP id A37EA102CA4; Wed,  9 Jan 2013 15:16:07 +0100 (CET)
X-Virus-Scanned: Amavisd on Debian GNU/Linux (netlab.nec.de)
Received: from mailer1.neclab.eu ([127.0.0.1]) by localhost (atlas-a.office.hd [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id UYXXVKKuA9Pi; Wed,  9 Jan 2013 15:16:07 +0100 (CET)
Received: from ENCELADUS.office.hd (enceladus.office.hd [192.168.24.52]) by mailer1.neclab.eu (Postfix) with ESMTP id 80707102CA2; Wed,  9 Jan 2013 15:15:42 +0100 (CET)
Received: from [10.1.1.190] (10.1.1.190) by skoll.office.hd (192.168.125.11) with Microsoft SMTP Server (TLS) id 14.1.323.3; Wed, 9 Jan 2013 15:15:06 +0100
Message-ID: <50ED7B8D.3010904@neclab.eu>
Date: Wed, 9 Jan 2013 15:15:41 +0100
From: Martin Stiemerling <martin.stiemerling@neclab.eu>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/17.0 Thunderbird/17.0
MIME-Version: 1.0
To: Greg Bumgardner <gbumgard@gmail.com>
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se> <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com>
In-Reply-To: <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com>
Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
Content-Transfer-Encoding: 7bit
X-Originating-IP: [10.1.1.190]
Cc: "mboned@ietf.org" <mboned@ietf.org>, "iesg@ietf.org" <iesg@ietf.org>, "tsv-dir@ietf.org" <tsv-dir@ietf.org>
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 09 Jan 2013 14:16:10 -0000

Hi Greg,

I'm joining the discussion as the DISCUSS holder for most of the issues 
below.

On 01/09/2013 03:26 AM, Greg Bumgardner wrote:
> Magnus,
>
> I would like to thank you for providing valuable feedback regarding the
> AMT spec.
>
> I would first like to state that many of your concerns and criticisms
> regarding the protocol were raised during MBONED group discussions. Many
> alternative solutions were identified but ultimately rejected by the
> group, primarily due to the reluctance of group members to make
> significant changes to the protocol after several years of slow
> development. My comments are inlined below...

Can you give some insights why the protocol development took so long?

Anyhow, I can understand the reluctance of changing parts of a protocol 
after such a long time. But there are a lot of rough edges in the 
protocol that need attention before moving it forward on the Standards 
Track.

>
> Thanks,
>
> -g.b.
>
>
> On Wed, Dec 26, 2012 at 5:01 AM, Magnus Westerlund
> <magnus.westerlund@ericsson.com <mailto:magnus.westerlund@ericsson.com>>
> wrote:
>
>     Hi,
>
>     Sorry for being late with this IETF last call comments. I will
>     partly blame the ADs requesting this Transport Directorate review a
>     bit late, the other part is all mine and the holidays. Anyway, I do
>     hope you will consider these issues and comments as I believe I
>     found some serious ones in addition to a number of clarifications
>     that should be made.
>
>
>     Significant Issues:
>
>     1. Congestion Control
>     This is clearly a tunnel establishment protocol of something that is
>     IP traffic. Thus normally the responsibility for congestion control
>     is with the tunnelled traffic. However, I would like argue that this
>     does not apply in this case due to the nature of the tunnelled
>     traffic, i.e. multicast traffic and secondary due to limitations in
>     the tunnel protocol.
>
>     Lets start with the second part. This protocol claims to support ASM
>     still don't provide a upstream delivery mechanism, i.e. an ASM
>     receiver is not capable of sending as it should. This prevents
>     several existing mechanism for congestion control that exist in
>     protocols supporting multicast. The first is using RTCP for
>     congestion control in ASM [RFC3550], the second isTCP-Friendly
>     Multicast Congestion Control [RFC4654] that can be used in the RMT
>     suite of protocols, and I know has been implemented in some NORM
>     implementation. Thus onlystrictly receiver based mechanisms, such as
>     Wave and Equation Based Rate Control [RFC3738] are available in this
>     context.
>
>
> You are correct, the AMT protocol provides no method for injecting
> multicast traffic into a multicast enabled network, and thus does not
> provide the functionality required for something like ASM RTP/RTCP
> teleconferencing. The protocol is primarily intended for use in
> receiving SSM traffic. Because the AMT protocol uses IGMP/MLD messages
> to establish reception/subscription state, the protocol MAY be used to
> subscribe to ASM broadcasts, but cannot generate them.
>
> I should note that earlier versions of the AMT spec described a method
> for broadcasting SSM via a gateway/relay tunnel, but this capability was
> removed due to feasibility considerations.

This means that the draft should state that it does not support ASM, but 
only a subset of it. Further, it cuts away the feedback over RTCP and 
RFC 4654.

>
>
>     Secondly, many multicast usages are in fact deployed without any
>     congestion control. This based on that the deploying entity controls
>     the scope and authorization for requesting multicast delivery.
>     However, does restrictions does not apply to AMT delivery of
>     multicast. If the gateway can reach using unicast the relay it can
>     be delivered the multicast group from the domain the relay is
>     attached to. Thus, this protocol changes the deployment restrictions
>     of multicast which many non-congestion controlled delivery is based
>     on. Instead the non-congestion controlled trafficcan now sent over
>     an IP/UDP tunnel over Internet where neither relay nor gateway may
>     have any knowledge about the path the traffic may take.
>
>
> The AMT protocol does not attempt to address the issue of congestion
> control. The group decided to treat the detection and remediation of
> congestion (of encapsulated multicast packets) as an application-layer
> responsibility. While we might have attempted to provide some level of
> QoS control on tunnel packets by copying ToS/DS bits from inner to outer
> headers, this would only occur on a packet-by-packet basis on not on the
> tunnel as a whole.

What's the group's view on how the application layer is handling 
congestion control in this case when using AMT?

>
>
>     Based on this I would like to see two changes to this protocol
>     specification. First a section discussing the issue of congestion
>     control.
>
>
> I'm not sure what, if anything, I might add to address this. I think the
> spec makes it pretty clear that the tunneling is provided by best-effort
> delivery of UDP packets (perhaps restricted by relay/router
> configuration). Perhaps you could suggest some text.

There are multiple issues here:
- there is no feedback loop between relay and gateway so that any 
information about the tunneled multicast traffic can be sent back from 
the gateway to the relay. This feedback loop can be used for congestion 
control, but also to figure out if the non-zero IPv6/UDP traffic gets 
black holed somewhere.
- the tunneled traffic can exceed the capacity of the network path 
between relay and gateway. This needs to be monitored and acted upon any 
congestion.
- since a single relay can feed many gateways, even tunneled multicast 
flows from the same relay can compete for network path capacity. I.e., 
if the tunneled traffic shares a network path. In the worst case, an 
added multicast flow can lead to congestion for all other tunneled 
multicast flows.
- any tunneled flow can impact other traffic from other hosts in the 
Internet.

>
>     Secondly, I think this protocol should have an applicability
>     statement limiting its deployment to restricted environments where
>     the relay and gateway deployers can provide certain resource
>     provisions between the entities to avoid the multicast traffic
>     affecting other traffic sharing the same bottlenecksin ways not
>     allowed by the network provider.
>
>
> While I understand your concern, this would essentially negate the
> entire reason for the protocol, which is to provide multicast transit to
> non-multicast connected receivers. However, the AMT spec does not
> preclude the implementation/deployment of AMT-enabled gateway devices
> for which AMT support may be enabled or disabled (or blocked via a
> firewall).

The point is that it is not safe to operate in the public Internet with 
the current protocol. Either with fix the issues in the protocol or we 
limit the applicability.

>
>
>
>     2. Security
>
>     This protocols is frank with it having limited security features and
>     says this is similar to the IGMP and MLD protocols being used.
>     However, I think this is a failure to propoerly consider the threat
>     model. If one uses AMT over general Internet it will run in a
>     network where the one deploying the multicast and the relays no
>     longer control requirements on source address verification or
>     possibilities for traffic separation as they can do within the
>     domains where multicast currently are deployed. The security
>     vulnerabilities in IGMP and MLD are much more contained and
>     controllable in a LAN environment where one has chosen to deploy
>     multicast compared to an Relay exposing this to the whole Internet.
>     Once more I think there is only two choices here.
>
>     A) Beef up the security to general Internet threat model, i.e. at a
>     minimal provide a real model for gateway authentication using
>     identities, not only return routability based verifications.
>
>     B) Limit the applicability of AMT to managed environments and make
>     it clear that the relay will need to limit which gateways are
>     allowed to access the relay based on addressing.
>
>     Based on the first significant issue with congestion control I
>     expect that there is little meaning to do A) unless also one is
>     willing to beef up AMT to provide congestion control. Which I think
>     is not according to the design wishes for the protocol designers.
>
>
> The security topic was visited and revisited numerous times over the
> last couple of years. The group made the decision to limit the protocol
> to simple source address verification and leave it at that. An emphasis
> was placed on protecting relays against simple DoS attacks but not gateways.

I can see the point, but I assume that most people are just looking to 
closed environments, e.g., multicast in access networks, such as xDSL or 
3GPP. However, AMT assumes to run across the public Internet and threats 
are extremely different there. Magnus made a good proposal that is 
relatively easy to implement.

>
>
>     3. Use of Zero Checksum
>
>     The AMT specification enables the use of Zero UDP checksum with
>     IPv6, i.e.
>     draft-ietf-6man-udpchecksums-06
>     <http://datatracker.ietf.org/doc/draft-ietf-6man-udpchecksums/> and
>     draft-ietf-6man-udpzero-08
>     <http://datatracker.ietf.org/doc/draft-ietf-6man-udpzero/>. Nothing
>     against this in principal. However, I have noticed that AMT fails to
>     properly address the failure modes of using a zero-checksum. AMT is
>     a typical example of a protocol that actually need active
>     verification of each tunnel that zero-checksum functions. This as
>     AMT is clearly intended to be deployed with its Gateway part in
>     end-hosts and residential network devices or routers. This means the
>     tunnel will pass through both firewalls and NATs on its path between
>     the relay and the gateway. Unless these devices are not upgraded to
>     support zero-checksum in UDP for IPv6 the traffic may actually
>     become black holed. The most likely is a simple firewall that has a
>     rule for IPv6/UDP which doesn't allow zero checksum as it is against
>     RFC 2460. Thus all the Multicast data packets will disappear on
>     route in the tunnel. There is no mechanism in the AMT protocol to
>     detect this and negotiate with the relay so that it will not use the
>     zero-checksum for this tunnel.
>
>     This must be addressed as I see it. If not the AMT will be so
>     brittle that it can't be used in a large number of its intended
>     deployments.
>
>
> You probably have no idea how often this issue was raised and discussed
> within the group. The consensus was that there was no single solution to
> the problem. The only workable solution I can suggest is to use an IPv4
> tunnel to carry the IPv6 packets. The mechanism for deciding whether (or
> if) to use an IPv4 tunnel is considered out of scope.

There is a solution, as mentioned above: feedback loop. If the relay 
says "sending you x packets" and the gateway sees zero, there is an 
issue. In the easiest case, a log file entry is generate saying that the 
tunneled traffic couldn't reach this particular gateway.

>
>
>     4) MTU issues
>
>     This document total fails to discuss the issues of MTU blackholing.
>     As the IP multicast datagrams as well as the encapsulated IGMP/MLD
>     messages can with the added tunnel-overhead result in that the sent
>     packet exceeds the MTU of the path, these packets could be black
>     holed. This can potentially result in very intermittent transport
>     behavior for the tunnel. Thus, some discussion of how do handle the
>     MTU issues in this context should be introduced.
>
>
> Yep. This could be a problem that has no easy solution. Not only might
> the outer packets be fragmented, but the encapsulated packets might be
> fragments. Since the protocol obviously has no control over the size of
> source packets, it cannot really avoid the potential need to fragment
> the tunnel packets. A gateway can only rely on the underlying UDP/IP
> stack to provide the necessary reassembly of the tunnel packets. Once
> the tunnel packet has been reassembled, the gateway, host IP stack or
> application may still need to reassemble the contents of a sequence of
> tunnel packets.

In the case the WG doesn't have any solution, which I couldn't believe, 
it is worth to document the issues. One operational solution can be to 
limit the MTU in the native multicast network, if this part and the 
relay are under the same administrative control. However, this would 
require that the AMT relay logs any issue with the MTU accross the tunnel.
Further, Magnus point is about that the tunneled packets can get 
fragmented, less about the inner parts of the tunnel.

Addressing the MTU issue will increase the viability of this a lot.

>
>
>     I am willing to discuss methods here, but I guess several
>     alternatives exists and thus which is most appropriate and the level
>     of AMT support for them varies I would like the protocol designers
>     to do a first stab at resolving this.
>
>     Other Issues
>     ======================================================================
>
>     A) The table of content on page 2 should include more levels of
>     headings. Most likely down to 4 is needed to make the TOC usable for
>     finding content in the document.
>
>     B) The claimed ASM support
>     I would like to better understand how one can claim to support ASM
>     when one has no up stream path to inject the ASM group participants
>     traffic. When one uses ASM one normally does this for a reason and
>     needing the possibility to inject packets into the group. These
>     limitations needs to be clarified.
>
>
> See above. The protocol is intended for multicast reception. It makes no
> assumptions about application layer use of the multicast traffic. It can
> only be used to receive ASM IP traffic.

Document it, please. Otherwise, the draft is incomplete.

Thank you,

   Martin
-- 
IETF Transport Area Director

martin.stiemerling@neclab.eu

NEC Laboratories Europe - Network Research Division NEC Europe Limited
Registered Office: NEC House, 1 Victoria Road, London W3 6BL
Registered in England 283

From gjshep@gmail.com  Thu Jan 10 08:35:36 2013
Return-Path: <gjshep@gmail.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6E2C621F8945; Thu, 10 Jan 2013 08:35:36 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.499
X-Spam-Level: 
X-Spam-Status: No, score=-103.499 tagged_above=-999 required=5 tests=[AWL=0.100, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 6TYorQIcgEmW; Thu, 10 Jan 2013 08:35:35 -0800 (PST)
Received: from mail-ob0-f182.google.com (mail-ob0-f182.google.com [209.85.214.182]) by ietfa.amsl.com (Postfix) with ESMTP id 9C0EE21F846B; Thu, 10 Jan 2013 08:35:34 -0800 (PST)
Received: by mail-ob0-f182.google.com with SMTP id 16so755923obc.27 for <multiple recipients>; Thu, 10 Jan 2013 08:35:34 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:reply-to:in-reply-to:references:date:message-id :subject:from:to:cc:content-type; bh=azf1icUTtwvuFwCMPC/EzEkKFNoI3YZfjSyeKc/ZPHc=; b=wOq9Lput0Q0Mg00p2G/N/fCrCUtGxxvDFTz6brlUacg6n6oRMXWKCsORD5izwTgE6o p42KLZ/nQrODfAw/fqyNwHv7m+BSiLm2iuh20xqKfijaoqzdDkS6cdoMgi7DXEq4UWKS KU+lpTtPGcM5ZsNhGcEOLF/4iizFWiE7yNjRz3uNeJwmeEiUidae54jF7gmm0BSA2I8l 47ZnGhgDDmKNTxvVSAiav0TFOdr55r2jb1effSCmlmt7Ps2wJZbGbDoxpz61QLuX+ooU mK++aQHMklmxRe2zH18TrTdcZ/JjXkEG+ApwR01bEAOrznzwefQKZK1xnM5bbO5cAS4i 4xMQ==
MIME-Version: 1.0
Received: by 10.60.31.130 with SMTP id a2mr43240164oei.95.1357835734019; Thu, 10 Jan 2013 08:35:34 -0800 (PST)
Received: by 10.76.105.107 with HTTP; Thu, 10 Jan 2013 08:35:33 -0800 (PST)
In-Reply-To: <50ED7B8D.3010904@neclab.eu>
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se> <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com> <50ED7B8D.3010904@neclab.eu>
Date: Thu, 10 Jan 2013 08:35:33 -0800
Message-ID: <CABFReBoCTN-KP=LHrVy+Bg4y6g4aezUH-2rWzvZ0GVmAtejERA@mail.gmail.com>
From: Greg Shepherd <gjshep@gmail.com>
To: Martin Stiemerling <martin.stiemerling@neclab.eu>
Content-Type: text/plain; charset=ISO-8859-1
Cc: "mboned@ietf.org" <mboned@ietf.org>, Greg Bumgardner <gbumgard@gmail.com>, "iesg@ietf.org" <iesg@ietf.org>, "tsv-dir@ietf.org" <tsv-dir@ietf.org>
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: gjshep@gmail.com
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 10 Jan 2013 16:35:36 -0000

On Wed, Jan 9, 2013 at 6:15 AM, Martin Stiemerling
<martin.stiemerling@neclab.eu> wrote:
> Hi Greg,
>
> I'm joining the discussion as the DISCUSS holder for most of the issues
> below.
>
>
> On 01/09/2013 03:26 AM, Greg Bumgardner wrote:
>>
>> Magnus,
>>
>> I would like to thank you for providing valuable feedback regarding the
>> AMT spec.
>>
>> I would first like to state that many of your concerns and criticisms
>> regarding the protocol were raised during MBONED group discussions. Many
>> alternative solutions were identified but ultimately rejected by the
>> group, primarily due to the reluctance of group members to make
>> significant changes to the protocol after several years of slow
>> development. My comments are inlined below...
>
>
> Can you give some insights why the protocol development took so long?

Yes, I can, though I'm not sure how relevant the incubation period is
to its adoption. And since I've breaking multicast networks for 15+
years now, asking a question like this may get you more than you were
expecting :)

The need for multicast transition services is as old as multicast
itself. Deering's first work - DVMRP - had to assume there was no
'native' path and so incorporated encapsulation as an integral part of
the protocol. I believe DVMRP may be the first use of overlay
encapsulation in the IETF, but I welcome you to dig for your self and
back with me. It's the earliest I've been able to find.

Once we moved to PIM, encapsulation for receiver delivery was thrown
out and we all pushed to "go native". The reality was - and still is -
that those most wanting end-to-end native multicast (content owners)
had no way to motivate network operators to enable multicast services.
Deployment experience with PIM revealed to many the need for
encapsulation overlay mechanism, and from this several drafts were
introduced.

In 2003 we held a BOF at IETF57 with the intention of agreeing on a
list of requirements and consolidating the efforts into a single
draft. We selected the AMT draft as being the closest fit, though
significant changes still need to be addressed. You see the list of
drafts and read the BOF minutes here:

http://www.ietf.org/meeting/57/session/mlm.txt

Authors came and went.. Progress was slow, until ATT picked up the
torch, demonstrated running code, and outlined their deployment plans.
Greg Bumgardner also wrote a java Gateway implementation and took over
as primary author. Finally we had solid industry support and a
qualified and motivated set of contributors.

Last year we finally reached WG consensus, along with at least three
shipping implementations of both the Relay and the Gateway , and at
least one active and growing commercial deployment of a
IPMulticast/AMT service. I doubt many drafts in the editors' queue can
claim that much support. :)

> Anyhow, I can understand the reluctance of changing parts of a protocol
> after such a long time. But there are a lot of rough edges in the protocol
> that need attention before moving it forward on the Standards Track.

Time is irrelevant to any reluctance to change - commercial
implementations and deployments are, however quite relevant.

Cheers,
Greg

>>
>> Thanks,
>>
>> -g.b.
>>
>>
>> On Wed, Dec 26, 2012 at 5:01 AM, Magnus Westerlund
>> <magnus.westerlund@ericsson.com <mailto:magnus.westerlund@ericsson.com>>
>>
>> wrote:
>>
>>     Hi,
>>
>>     Sorry for being late with this IETF last call comments. I will
>>     partly blame the ADs requesting this Transport Directorate review a
>>     bit late, the other part is all mine and the holidays. Anyway, I do
>>     hope you will consider these issues and comments as I believe I
>>     found some serious ones in addition to a number of clarifications
>>     that should be made.
>>
>>
>>     Significant Issues:
>>
>>     1. Congestion Control
>>     This is clearly a tunnel establishment protocol of something that is
>>     IP traffic. Thus normally the responsibility for congestion control
>>     is with the tunnelled traffic. However, I would like argue that this
>>     does not apply in this case due to the nature of the tunnelled
>>     traffic, i.e. multicast traffic and secondary due to limitations in
>>     the tunnel protocol.
>>
>>     Lets start with the second part. This protocol claims to support ASM
>>     still don't provide a upstream delivery mechanism, i.e. an ASM
>>     receiver is not capable of sending as it should. This prevents
>>     several existing mechanism for congestion control that exist in
>>     protocols supporting multicast. The first is using RTCP for
>>     congestion control in ASM [RFC3550], the second isTCP-Friendly
>>     Multicast Congestion Control [RFC4654] that can be used in the RMT
>>     suite of protocols, and I know has been implemented in some NORM
>>     implementation. Thus onlystrictly receiver based mechanisms, such as
>>
>>     Wave and Equation Based Rate Control [RFC3738] are available in this
>>     context.
>>
>>
>> You are correct, the AMT protocol provides no method for injecting
>> multicast traffic into a multicast enabled network, and thus does not
>> provide the functionality required for something like ASM RTP/RTCP
>> teleconferencing. The protocol is primarily intended for use in
>> receiving SSM traffic. Because the AMT protocol uses IGMP/MLD messages
>> to establish reception/subscription state, the protocol MAY be used to
>> subscribe to ASM broadcasts, but cannot generate them.
>>
>> I should note that earlier versions of the AMT spec described a method
>> for broadcasting SSM via a gateway/relay tunnel, but this capability was
>> removed due to feasibility considerations.
>
>
> This means that the draft should state that it does not support ASM, but
> only a subset of it. Further, it cuts away the feedback over RTCP and RFC
> 4654.
>
>>
>>
>>     Secondly, many multicast usages are in fact deployed without any
>>     congestion control. This based on that the deploying entity controls
>>     the scope and authorization for requesting multicast delivery.
>>     However, does restrictions does not apply to AMT delivery of
>>     multicast. If the gateway can reach using unicast the relay it can
>>     be delivered the multicast group from the domain the relay is
>>     attached to. Thus, this protocol changes the deployment restrictions
>>     of multicast which many non-congestion controlled delivery is based
>>     on. Instead the non-congestion controlled trafficcan now sent over
>>
>>     an IP/UDP tunnel over Internet where neither relay nor gateway may
>>     have any knowledge about the path the traffic may take.
>>
>>
>> The AMT protocol does not attempt to address the issue of congestion
>> control. The group decided to treat the detection and remediation of
>> congestion (of encapsulated multicast packets) as an application-layer
>> responsibility. While we might have attempted to provide some level of
>> QoS control on tunnel packets by copying ToS/DS bits from inner to outer
>> headers, this would only occur on a packet-by-packet basis on not on the
>> tunnel as a whole.
>
>
> What's the group's view on how the application layer is handling congestion
> control in this case when using AMT?
>
>
>>
>>
>>     Based on this I would like to see two changes to this protocol
>>     specification. First a section discussing the issue of congestion
>>     control.
>>
>>
>> I'm not sure what, if anything, I might add to address this. I think the
>> spec makes it pretty clear that the tunneling is provided by best-effort
>> delivery of UDP packets (perhaps restricted by relay/router
>> configuration). Perhaps you could suggest some text.
>
>
> There are multiple issues here:
> - there is no feedback loop between relay and gateway so that any
> information about the tunneled multicast traffic can be sent back from the
> gateway to the relay. This feedback loop can be used for congestion control,
> but also to figure out if the non-zero IPv6/UDP traffic gets black holed
> somewhere.
> - the tunneled traffic can exceed the capacity of the network path between
> relay and gateway. This needs to be monitored and acted upon any congestion.
> - since a single relay can feed many gateways, even tunneled multicast flows
> from the same relay can compete for network path capacity. I.e., if the
> tunneled traffic shares a network path. In the worst case, an added
> multicast flow can lead to congestion for all other tunneled multicast
> flows.
> - any tunneled flow can impact other traffic from other hosts in the
> Internet.
>
>>
>>     Secondly, I think this protocol should have an applicability
>>     statement limiting its deployment to restricted environments where
>>     the relay and gateway deployers can provide certain resource
>>     provisions between the entities to avoid the multicast traffic
>>     affecting other traffic sharing the same bottlenecksin ways not
>>
>>     allowed by the network provider.
>>
>>
>> While I understand your concern, this would essentially negate the
>> entire reason for the protocol, which is to provide multicast transit to
>> non-multicast connected receivers. However, the AMT spec does not
>> preclude the implementation/deployment of AMT-enabled gateway devices
>> for which AMT support may be enabled or disabled (or blocked via a
>> firewall).
>
>
> The point is that it is not safe to operate in the public Internet with the
> current protocol. Either with fix the issues in the protocol or we limit the
> applicability.
>
>
>>
>>
>>
>>     2. Security
>>
>>     This protocols is frank with it having limited security features and
>>     says this is similar to the IGMP and MLD protocols being used.
>>     However, I think this is a failure to propoerly consider the threat
>>     model. If one uses AMT over general Internet it will run in a
>>     network where the one deploying the multicast and the relays no
>>     longer control requirements on source address verification or
>>     possibilities for traffic separation as they can do within the
>>     domains where multicast currently are deployed. The security
>>     vulnerabilities in IGMP and MLD are much more contained and
>>     controllable in a LAN environment where one has chosen to deploy
>>     multicast compared to an Relay exposing this to the whole Internet.
>>     Once more I think there is only two choices here.
>>
>>     A) Beef up the security to general Internet threat model, i.e. at a
>>     minimal provide a real model for gateway authentication using
>>     identities, not only return routability based verifications.
>>
>>     B) Limit the applicability of AMT to managed environments and make
>>     it clear that the relay will need to limit which gateways are
>>     allowed to access the relay based on addressing.
>>
>>     Based on the first significant issue with congestion control I
>>     expect that there is little meaning to do A) unless also one is
>>     willing to beef up AMT to provide congestion control. Which I think
>>     is not according to the design wishes for the protocol designers.
>>
>>
>> The security topic was visited and revisited numerous times over the
>> last couple of years. The group made the decision to limit the protocol
>> to simple source address verification and leave it at that. An emphasis
>> was placed on protecting relays against simple DoS attacks but not
>> gateways.
>
>
> I can see the point, but I assume that most people are just looking to
> closed environments, e.g., multicast in access networks, such as xDSL or
> 3GPP. However, AMT assumes to run across the public Internet and threats are
> extremely different there. Magnus made a good proposal that is relatively
> easy to implement.
>
>>
>>
>>     3. Use of Zero Checksum
>>
>>     The AMT specification enables the use of Zero UDP checksum with
>>     IPv6, i.e.
>>     draft-ietf-6man-udpchecksums-06
>>     <http://datatracker.ietf.org/doc/draft-ietf-6man-udpchecksums/> and
>>     draft-ietf-6man-udpzero-08
>>     <http://datatracker.ietf.org/doc/draft-ietf-6man-udpzero/>. Nothing
>>
>>     against this in principal. However, I have noticed that AMT fails to
>>     properly address the failure modes of using a zero-checksum. AMT is
>>     a typical example of a protocol that actually need active
>>     verification of each tunnel that zero-checksum functions. This as
>>     AMT is clearly intended to be deployed with its Gateway part in
>>     end-hosts and residential network devices or routers. This means the
>>     tunnel will pass through both firewalls and NATs on its path between
>>     the relay and the gateway. Unless these devices are not upgraded to
>>     support zero-checksum in UDP for IPv6 the traffic may actually
>>     become black holed. The most likely is a simple firewall that has a
>>     rule for IPv6/UDP which doesn't allow zero checksum as it is against
>>     RFC 2460. Thus all the Multicast data packets will disappear on
>>     route in the tunnel. There is no mechanism in the AMT protocol to
>>     detect this and negotiate with the relay so that it will not use the
>>     zero-checksum for this tunnel.
>>
>>     This must be addressed as I see it. If not the AMT will be so
>>     brittle that it can't be used in a large number of its intended
>>     deployments.
>>
>>
>> You probably have no idea how often this issue was raised and discussed
>> within the group. The consensus was that there was no single solution to
>> the problem. The only workable solution I can suggest is to use an IPv4
>> tunnel to carry the IPv6 packets. The mechanism for deciding whether (or
>> if) to use an IPv4 tunnel is considered out of scope.
>
>
> There is a solution, as mentioned above: feedback loop. If the relay says
> "sending you x packets" and the gateway sees zero, there is an issue. In the
> easiest case, a log file entry is generate saying that the tunneled traffic
> couldn't reach this particular gateway.
>
>
>>
>>
>>     4) MTU issues
>>
>>     This document total fails to discuss the issues of MTU blackholing.
>>     As the IP multicast datagrams as well as the encapsulated IGMP/MLD
>>     messages can with the added tunnel-overhead result in that the sent
>>     packet exceeds the MTU of the path, these packets could be black
>>     holed. This can potentially result in very intermittent transport
>>     behavior for the tunnel. Thus, some discussion of how do handle the
>>     MTU issues in this context should be introduced.
>>
>>
>> Yep. This could be a problem that has no easy solution. Not only might
>> the outer packets be fragmented, but the encapsulated packets might be
>> fragments. Since the protocol obviously has no control over the size of
>> source packets, it cannot really avoid the potential need to fragment
>> the tunnel packets. A gateway can only rely on the underlying UDP/IP
>> stack to provide the necessary reassembly of the tunnel packets. Once
>> the tunnel packet has been reassembled, the gateway, host IP stack or
>> application may still need to reassemble the contents of a sequence of
>> tunnel packets.
>
>
> In the case the WG doesn't have any solution, which I couldn't believe, it
> is worth to document the issues. One operational solution can be to limit
> the MTU in the native multicast network, if this part and the relay are
> under the same administrative control. However, this would require that the
> AMT relay logs any issue with the MTU accross the tunnel.
> Further, Magnus point is about that the tunneled packets can get fragmented,
> less about the inner parts of the tunnel.
>
> Addressing the MTU issue will increase the viability of this a lot.
>
>
>>
>>
>>     I am willing to discuss methods here, but I guess several
>>     alternatives exists and thus which is most appropriate and the level
>>     of AMT support for them varies I would like the protocol designers
>>     to do a first stab at resolving this.
>>
>>     Other Issues
>>     ======================================================================
>>
>>     A) The table of content on page 2 should include more levels of
>>     headings. Most likely down to 4 is needed to make the TOC usable for
>>     finding content in the document.
>>
>>     B) The claimed ASM support
>>     I would like to better understand how one can claim to support ASM
>>     when one has no up stream path to inject the ASM group participants
>>     traffic. When one uses ASM one normally does this for a reason and
>>     needing the possibility to inject packets into the group. These
>>     limitations needs to be clarified.
>>
>>
>> See above. The protocol is intended for multicast reception. It makes no
>> assumptions about application layer use of the multicast traffic. It can
>> only be used to receive ASM IP traffic.
>
>
> Document it, please. Otherwise, the draft is incomplete.
>
> Thank you,
>
>   Martin
> --
> IETF Transport Area Director
>
> martin.stiemerling@neclab.eu
>
> NEC Laboratories Europe - Network Research Division NEC Europe Limited
> Registered Office: NEC House, 1 Victoria Road, London W3 6BL
> Registered in England 283
>
> _______________________________________________
> MBONED mailing list
> MBONED@ietf.org
> https://www.ietf.org/mailman/listinfo/mboned

From magnus.westerlund@ericsson.com  Wed Jan 16 00:54:02 2013
Return-Path: <magnus.westerlund@ericsson.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9EBBD21F86C2; Wed, 16 Jan 2013 00:54:02 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -106.049
X-Spam-Level: 
X-Spam-Status: No, score=-106.049 tagged_above=-999 required=5 tests=[AWL=0.200, BAYES_00=-2.599, HELO_EQ_SE=0.35, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id r6IkPjceW48c; Wed, 16 Jan 2013 00:54:01 -0800 (PST)
Received: from mailgw2.ericsson.se (mailgw2.ericsson.se [193.180.251.37]) by ietfa.amsl.com (Postfix) with ESMTP id D78B021F8691; Wed, 16 Jan 2013 00:53:59 -0800 (PST)
X-AuditID: c1b4fb25-b7f366d000004d10-45-50f66aa649b5
Received: from esessmw0247.eemea.ericsson.se (Unknown_Domain [153.88.253.124]) by mailgw2.ericsson.se (Symantec Mail Security) with SMTP id 1C.BE.19728.6AA66F05; Wed, 16 Jan 2013 09:53:59 +0100 (CET)
Received: from [127.0.0.1] (153.88.115.8) by esessmw0247.eemea.ericsson.se (153.88.115.94) with Microsoft SMTP Server id 8.3.279.1; Wed, 16 Jan 2013 09:53:58 +0100
Message-ID: <50F66AA5.6050705@ericsson.com>
Date: Wed, 16 Jan 2013 09:53:57 +0100
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1; rv:17.0) Gecko/20130107 Thunderbird/17.0.2
MIME-Version: 1.0
To: gjshep@gmail.com
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se> <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com> <50ED7B8D.3010904@neclab.eu> <CABFReBoCTN-KP=LHrVy+Bg4y6g4aezUH-2rWzvZ0GVmAtejERA@mail.gmail.com>
In-Reply-To: <CABFReBoCTN-KP=LHrVy+Bg4y6g4aezUH-2rWzvZ0GVmAtejERA@mail.gmail.com>
X-Enigmail-Version: 1.5
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFlrFLMWRmVeSWpSXmKPExsUyM+Jvje7yrG8BBk16Fru/NzBbbJmTZzHj z0Rmi46+q2wWU1pOM1p8u3KD3YHNY+esu+weS5b8ZPKYcewlewBzFJdNSmpOZllqkb5dAlfG mq8GBSunMlb0/Wlnb2D8XtrFyMkhIWAi8eTqNSYIW0ziwr31bF2MXBxCAicZJW42XmGHcJYz Shxqn8kIUsUroC3xbNNrdhCbRUBVYsmUNcwgNpuAhcTNH41sILaoQLDEhoOrmCDqBSVOznzC AmKLCIhIXJ8zmxVkKLPABUaJUw+OgznCAo2MEv2NIJNA1n1ilDh3ZRdYO6dAoMTrJ5uBVnMA HSguseYNB0iYWUBPYsrVFkYIW16ieetssCuEgK5raOpgncAoNAvJ8llIWmYhaVnAyLyKkT03 MTMnvdxoEyMwvA9u+a26g/HOOZFDjNIcLErivOGuFwKEBNITS1KzU1MLUovii0pzUosPMTJx cEo1MG4RXb3Mi19q+bK0X1pxHA0CB6barpz8Y9OmX03S1w873Dn48nQYF/caxd8XVv1tirNN cnRTmP4746W1QmNhNcvqLcosL2Yt9ovlvPOKU/vo6dg7j291cPG4OJUI1uQulfhXlu8zd0U2 /3yPrFiNtW+OsRZvXb1JZueG1bdjD1mGeHaUTetZzajEUpyRaKjFXFScCACFql08PQIAAA==
Cc: "mboned@ietf.org" <mboned@ietf.org>, Greg Bumgardner <gbumgard@gmail.com>, "tsv-dir@ietf.org" <tsv-dir@ietf.org>, "iesg@ietf.org" <iesg@ietf.org>, Martin Stiemerling <martin.stiemerling@neclab.eu>
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Jan 2013 08:54:02 -0000

Hi,

I think the major question here, are anyone really objecting to
restricting the applicability of AMT to managed environments, i.e. where
the full path between relay and client are in domains that are
explicitly aware that they will have AMT traffic and approve of it. The
comments appears to indicate that this is where it is deployed. If one
makes such an applicability statement most of the significant comments
can be more easily resolved.

Cheers

Magnus

On 2013-01-10 17:35, Greg Shepherd wrote:
> On Wed, Jan 9, 2013 at 6:15 AM, Martin Stiemerling
> <martin.stiemerling@neclab.eu> wrote:
>> Hi Greg,
>>
>> I'm joining the discussion as the DISCUSS holder for most of the issues
>> below.
>>
>>
>> On 01/09/2013 03:26 AM, Greg Bumgardner wrote:
>>>
>>> Magnus,
>>>
>>> I would like to thank you for providing valuable feedback regarding the
>>> AMT spec.
>>>
>>> I would first like to state that many of your concerns and criticisms
>>> regarding the protocol were raised during MBONED group discussions. Many
>>> alternative solutions were identified but ultimately rejected by the
>>> group, primarily due to the reluctance of group members to make
>>> significant changes to the protocol after several years of slow
>>> development. My comments are inlined below...
>>
>>
>> Can you give some insights why the protocol development took so long?
> 
> Yes, I can, though I'm not sure how relevant the incubation period is
> to its adoption. And since I've breaking multicast networks for 15+
> years now, asking a question like this may get you more than you were
> expecting :)
> 
> The need for multicast transition services is as old as multicast
> itself. Deering's first work - DVMRP - had to assume there was no
> 'native' path and so incorporated encapsulation as an integral part of
> the protocol. I believe DVMRP may be the first use of overlay
> encapsulation in the IETF, but I welcome you to dig for your self and
> back with me. It's the earliest I've been able to find.
> 
> Once we moved to PIM, encapsulation for receiver delivery was thrown
> out and we all pushed to "go native". The reality was - and still is -
> that those most wanting end-to-end native multicast (content owners)
> had no way to motivate network operators to enable multicast services.
> Deployment experience with PIM revealed to many the need for
> encapsulation overlay mechanism, and from this several drafts were
> introduced.
> 
> In 2003 we held a BOF at IETF57 with the intention of agreeing on a
> list of requirements and consolidating the efforts into a single
> draft. We selected the AMT draft as being the closest fit, though
> significant changes still need to be addressed. You see the list of
> drafts and read the BOF minutes here:
> 
> http://www.ietf.org/meeting/57/session/mlm.txt
> 
> Authors came and went.. Progress was slow, until ATT picked up the
> torch, demonstrated running code, and outlined their deployment plans.
> Greg Bumgardner also wrote a java Gateway implementation and took over
> as primary author. Finally we had solid industry support and a
> qualified and motivated set of contributors.
> 
> Last year we finally reached WG consensus, along with at least three
> shipping implementations of both the Relay and the Gateway , and at
> least one active and growing commercial deployment of a
> IPMulticast/AMT service. I doubt many drafts in the editors' queue can
> claim that much support. :)
> 
>> Anyhow, I can understand the reluctance of changing parts of a protocol
>> after such a long time. But there are a lot of rough edges in the protocol
>> that need attention before moving it forward on the Standards Track.
> 
> Time is irrelevant to any reluctance to change - commercial
> implementations and deployments are, however quite relevant.
> 
> Cheers,
> Greg
> 
>>>
>>> Thanks,
>>>
>>> -g.b.
>>>
>>>
>>> On Wed, Dec 26, 2012 at 5:01 AM, Magnus Westerlund
>>> <magnus.westerlund@ericsson.com <mailto:magnus.westerlund@ericsson.com>>
>>>
>>> wrote:
>>>
>>>     Hi,
>>>
>>>     Sorry for being late with this IETF last call comments. I will
>>>     partly blame the ADs requesting this Transport Directorate review a
>>>     bit late, the other part is all mine and the holidays. Anyway, I do
>>>     hope you will consider these issues and comments as I believe I
>>>     found some serious ones in addition to a number of clarifications
>>>     that should be made.
>>>
>>>
>>>     Significant Issues:
>>>
>>>     1. Congestion Control
>>>     This is clearly a tunnel establishment protocol of something that is
>>>     IP traffic. Thus normally the responsibility for congestion control
>>>     is with the tunnelled traffic. However, I would like argue that this
>>>     does not apply in this case due to the nature of the tunnelled
>>>     traffic, i.e. multicast traffic and secondary due to limitations in
>>>     the tunnel protocol.
>>>
>>>     Lets start with the second part. This protocol claims to support ASM
>>>     still don't provide a upstream delivery mechanism, i.e. an ASM
>>>     receiver is not capable of sending as it should. This prevents
>>>     several existing mechanism for congestion control that exist in
>>>     protocols supporting multicast. The first is using RTCP for
>>>     congestion control in ASM [RFC3550], the second isTCP-Friendly
>>>     Multicast Congestion Control [RFC4654] that can be used in the RMT
>>>     suite of protocols, and I know has been implemented in some NORM
>>>     implementation. Thus onlystrictly receiver based mechanisms, such as
>>>
>>>     Wave and Equation Based Rate Control [RFC3738] are available in this
>>>     context.
>>>
>>>
>>> You are correct, the AMT protocol provides no method for injecting
>>> multicast traffic into a multicast enabled network, and thus does not
>>> provide the functionality required for something like ASM RTP/RTCP
>>> teleconferencing. The protocol is primarily intended for use in
>>> receiving SSM traffic. Because the AMT protocol uses IGMP/MLD messages
>>> to establish reception/subscription state, the protocol MAY be used to
>>> subscribe to ASM broadcasts, but cannot generate them.
>>>
>>> I should note that earlier versions of the AMT spec described a method
>>> for broadcasting SSM via a gateway/relay tunnel, but this capability was
>>> removed due to feasibility considerations.
>>
>>
>> This means that the draft should state that it does not support ASM, but
>> only a subset of it. Further, it cuts away the feedback over RTCP and RFC
>> 4654.
>>
>>>
>>>
>>>     Secondly, many multicast usages are in fact deployed without any
>>>     congestion control. This based on that the deploying entity controls
>>>     the scope and authorization for requesting multicast delivery.
>>>     However, does restrictions does not apply to AMT delivery of
>>>     multicast. If the gateway can reach using unicast the relay it can
>>>     be delivered the multicast group from the domain the relay is
>>>     attached to. Thus, this protocol changes the deployment restrictions
>>>     of multicast which many non-congestion controlled delivery is based
>>>     on. Instead the non-congestion controlled trafficcan now sent over
>>>
>>>     an IP/UDP tunnel over Internet where neither relay nor gateway may
>>>     have any knowledge about the path the traffic may take.
>>>
>>>
>>> The AMT protocol does not attempt to address the issue of congestion
>>> control. The group decided to treat the detection and remediation of
>>> congestion (of encapsulated multicast packets) as an application-layer
>>> responsibility. While we might have attempted to provide some level of
>>> QoS control on tunnel packets by copying ToS/DS bits from inner to outer
>>> headers, this would only occur on a packet-by-packet basis on not on the
>>> tunnel as a whole.
>>
>>
>> What's the group's view on how the application layer is handling congestion
>> control in this case when using AMT?
>>
>>
>>>
>>>
>>>     Based on this I would like to see two changes to this protocol
>>>     specification. First a section discussing the issue of congestion
>>>     control.
>>>
>>>
>>> I'm not sure what, if anything, I might add to address this. I think the
>>> spec makes it pretty clear that the tunneling is provided by best-effort
>>> delivery of UDP packets (perhaps restricted by relay/router
>>> configuration). Perhaps you could suggest some text.
>>
>>
>> There are multiple issues here:
>> - there is no feedback loop between relay and gateway so that any
>> information about the tunneled multicast traffic can be sent back from the
>> gateway to the relay. This feedback loop can be used for congestion control,
>> but also to figure out if the non-zero IPv6/UDP traffic gets black holed
>> somewhere.
>> - the tunneled traffic can exceed the capacity of the network path between
>> relay and gateway. This needs to be monitored and acted upon any congestion.
>> - since a single relay can feed many gateways, even tunneled multicast flows
>> from the same relay can compete for network path capacity. I.e., if the
>> tunneled traffic shares a network path. In the worst case, an added
>> multicast flow can lead to congestion for all other tunneled multicast
>> flows.
>> - any tunneled flow can impact other traffic from other hosts in the
>> Internet.
>>
>>>
>>>     Secondly, I think this protocol should have an applicability
>>>     statement limiting its deployment to restricted environments where
>>>     the relay and gateway deployers can provide certain resource
>>>     provisions between the entities to avoid the multicast traffic
>>>     affecting other traffic sharing the same bottlenecksin ways not
>>>
>>>     allowed by the network provider.
>>>
>>>
>>> While I understand your concern, this would essentially negate the
>>> entire reason for the protocol, which is to provide multicast transit to
>>> non-multicast connected receivers. However, the AMT spec does not
>>> preclude the implementation/deployment of AMT-enabled gateway devices
>>> for which AMT support may be enabled or disabled (or blocked via a
>>> firewall).
>>
>>
>> The point is that it is not safe to operate in the public Internet with the
>> current protocol. Either with fix the issues in the protocol or we limit the
>> applicability.
>>
>>
>>>
>>>
>>>
>>>     2. Security
>>>
>>>     This protocols is frank with it having limited security features and
>>>     says this is similar to the IGMP and MLD protocols being used.
>>>     However, I think this is a failure to propoerly consider the threat
>>>     model. If one uses AMT over general Internet it will run in a
>>>     network where the one deploying the multicast and the relays no
>>>     longer control requirements on source address verification or
>>>     possibilities for traffic separation as they can do within the
>>>     domains where multicast currently are deployed. The security
>>>     vulnerabilities in IGMP and MLD are much more contained and
>>>     controllable in a LAN environment where one has chosen to deploy
>>>     multicast compared to an Relay exposing this to the whole Internet.
>>>     Once more I think there is only two choices here.
>>>
>>>     A) Beef up the security to general Internet threat model, i.e. at a
>>>     minimal provide a real model for gateway authentication using
>>>     identities, not only return routability based verifications.
>>>
>>>     B) Limit the applicability of AMT to managed environments and make
>>>     it clear that the relay will need to limit which gateways are
>>>     allowed to access the relay based on addressing.
>>>
>>>     Based on the first significant issue with congestion control I
>>>     expect that there is little meaning to do A) unless also one is
>>>     willing to beef up AMT to provide congestion control. Which I think
>>>     is not according to the design wishes for the protocol designers.
>>>
>>>
>>> The security topic was visited and revisited numerous times over the
>>> last couple of years. The group made the decision to limit the protocol
>>> to simple source address verification and leave it at that. An emphasis
>>> was placed on protecting relays against simple DoS attacks but not
>>> gateways.
>>
>>
>> I can see the point, but I assume that most people are just looking to
>> closed environments, e.g., multicast in access networks, such as xDSL or
>> 3GPP. However, AMT assumes to run across the public Internet and threats are
>> extremely different there. Magnus made a good proposal that is relatively
>> easy to implement.
>>
>>>
>>>
>>>     3. Use of Zero Checksum
>>>
>>>     The AMT specification enables the use of Zero UDP checksum with
>>>     IPv6, i.e.
>>>     draft-ietf-6man-udpchecksums-06
>>>     <http://datatracker.ietf.org/doc/draft-ietf-6man-udpchecksums/> and
>>>     draft-ietf-6man-udpzero-08
>>>     <http://datatracker.ietf.org/doc/draft-ietf-6man-udpzero/>. Nothing
>>>
>>>     against this in principal. However, I have noticed that AMT fails to
>>>     properly address the failure modes of using a zero-checksum. AMT is
>>>     a typical example of a protocol that actually need active
>>>     verification of each tunnel that zero-checksum functions. This as
>>>     AMT is clearly intended to be deployed with its Gateway part in
>>>     end-hosts and residential network devices or routers. This means the
>>>     tunnel will pass through both firewalls and NATs on its path between
>>>     the relay and the gateway. Unless these devices are not upgraded to
>>>     support zero-checksum in UDP for IPv6 the traffic may actually
>>>     become black holed. The most likely is a simple firewall that has a
>>>     rule for IPv6/UDP which doesn't allow zero checksum as it is against
>>>     RFC 2460. Thus all the Multicast data packets will disappear on
>>>     route in the tunnel. There is no mechanism in the AMT protocol to
>>>     detect this and negotiate with the relay so that it will not use the
>>>     zero-checksum for this tunnel.
>>>
>>>     This must be addressed as I see it. If not the AMT will be so
>>>     brittle that it can't be used in a large number of its intended
>>>     deployments.
>>>
>>>
>>> You probably have no idea how often this issue was raised and discussed
>>> within the group. The consensus was that there was no single solution to
>>> the problem. The only workable solution I can suggest is to use an IPv4
>>> tunnel to carry the IPv6 packets. The mechanism for deciding whether (or
>>> if) to use an IPv4 tunnel is considered out of scope.
>>
>>
>> There is a solution, as mentioned above: feedback loop. If the relay says
>> "sending you x packets" and the gateway sees zero, there is an issue. In the
>> easiest case, a log file entry is generate saying that the tunneled traffic
>> couldn't reach this particular gateway.
>>
>>
>>>
>>>
>>>     4) MTU issues
>>>
>>>     This document total fails to discuss the issues of MTU blackholing.
>>>     As the IP multicast datagrams as well as the encapsulated IGMP/MLD
>>>     messages can with the added tunnel-overhead result in that the sent
>>>     packet exceeds the MTU of the path, these packets could be black
>>>     holed. This can potentially result in very intermittent transport
>>>     behavior for the tunnel. Thus, some discussion of how do handle the
>>>     MTU issues in this context should be introduced.
>>>
>>>
>>> Yep. This could be a problem that has no easy solution. Not only might
>>> the outer packets be fragmented, but the encapsulated packets might be
>>> fragments. Since the protocol obviously has no control over the size of
>>> source packets, it cannot really avoid the potential need to fragment
>>> the tunnel packets. A gateway can only rely on the underlying UDP/IP
>>> stack to provide the necessary reassembly of the tunnel packets. Once
>>> the tunnel packet has been reassembled, the gateway, host IP stack or
>>> application may still need to reassemble the contents of a sequence of
>>> tunnel packets.
>>
>>
>> In the case the WG doesn't have any solution, which I couldn't believe, it
>> is worth to document the issues. One operational solution can be to limit
>> the MTU in the native multicast network, if this part and the relay are
>> under the same administrative control. However, this would require that the
>> AMT relay logs any issue with the MTU accross the tunnel.
>> Further, Magnus point is about that the tunneled packets can get fragmented,
>> less about the inner parts of the tunnel.
>>
>> Addressing the MTU issue will increase the viability of this a lot.
>>
>>
>>>
>>>
>>>     I am willing to discuss methods here, but I guess several
>>>     alternatives exists and thus which is most appropriate and the level
>>>     of AMT support for them varies I would like the protocol designers
>>>     to do a first stab at resolving this.
>>>
>>>     Other Issues
>>>     ======================================================================
>>>
>>>     A) The table of content on page 2 should include more levels of
>>>     headings. Most likely down to 4 is needed to make the TOC usable for
>>>     finding content in the document.
>>>
>>>     B) The claimed ASM support
>>>     I would like to better understand how one can claim to support ASM
>>>     when one has no up stream path to inject the ASM group participants
>>>     traffic. When one uses ASM one normally does this for a reason and
>>>     needing the possibility to inject packets into the group. These
>>>     limitations needs to be clarified.
>>>
>>>
>>> See above. The protocol is intended for multicast reception. It makes no
>>> assumptions about application layer use of the multicast traffic. It can
>>> only be used to receive ASM IP traffic.
>>
>>
>> Document it, please. Otherwise, the draft is incomplete.
>>
>> Thank you,
>>
>>   Martin
>> --
>> IETF Transport Area Director
>>
>> martin.stiemerling@neclab.eu
>>
>> NEC Laboratories Europe - Network Research Division NEC Europe Limited
>> Registered Office: NEC House, 1 Victoria Road, London W3 6BL
>> Registered in England 283
>>
>> _______________________________________________
>> MBONED mailing list
>> MBONED@ietf.org
>> https://www.ietf.org/mailman/listinfo/mboned
> 


-- 

Magnus Westerlund

----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM
----------------------------------------------------------------------
Ericsson AB                | Phone  +46 10 7148287
Färögatan 6                | Mobile +46 73 0949079
SE-164 80 Stockholm, Sweden| mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


From gjshep@gmail.com  Wed Jan 16 06:04:13 2013
Return-Path: <gjshep@gmail.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 6071C21F8928; Wed, 16 Jan 2013 06:04:13 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.524
X-Spam-Level: 
X-Spam-Status: No, score=-103.524 tagged_above=-999 required=5 tests=[AWL=0.075, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kRteFVpjiV3H; Wed, 16 Jan 2013 06:04:11 -0800 (PST)
Received: from mail-wi0-f173.google.com (mail-wi0-f173.google.com [209.85.212.173]) by ietfa.amsl.com (Postfix) with ESMTP id EA18421F88F0; Wed, 16 Jan 2013 06:04:10 -0800 (PST)
Received: by mail-wi0-f173.google.com with SMTP id hn17so3479063wib.0 for <multiple recipients>; Wed, 16 Jan 2013 06:04:08 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:reply-to:in-reply-to:references:date :message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=r5WkSq2MUjfgZszA8vUNXDS+f9BVjC6AaVk0VEcVG0M=; b=yK3bxKVbbS4aKs3TsDqCtMQvI1nz2+PJJbv5OCPV6h6KIn4cBG2mWJe10BQQVClK3X Z9Jhg7QKiE2P6x3mutKbnOdi2S/KoH2KKlURj0a+Qw8iQyF6L8iKjEndgxk9bY0U8xza kuhzLqI0XqZnwt+jNGSXH+Ehokq5SziIqFgIJOxIyS4J0xZfhU5gQf2dy/qSEz8UuOKR tZofZ0jiA9k9+zewfi6TnzvBQZXW6qWRFUYA6wpnNKmS0BWksi9LGXCRiBooZ6rOXD7s t0Ti+NtYsaVwcGMonpWRSDUBGayHkXX5qkBhtXWB6OrqRerXHqCe9rs5CAvUhW0Tsy9g YUYw==
MIME-Version: 1.0
X-Received: by 10.180.8.130 with SMTP id r2mr2404804wia.28.1358345047987; Wed, 16 Jan 2013 06:04:07 -0800 (PST)
Received: by 10.180.97.66 with HTTP; Wed, 16 Jan 2013 06:04:07 -0800 (PST)
In-Reply-To: <50F66AA5.6050705@ericsson.com>
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se> <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com> <50ED7B8D.3010904@neclab.eu> <CABFReBoCTN-KP=LHrVy+Bg4y6g4aezUH-2rWzvZ0GVmAtejERA@mail.gmail.com> <50F66AA5.6050705@ericsson.com>
Date: Wed, 16 Jan 2013 06:04:07 -0800
Message-ID: <CABFReBpiT3=LJbS8_Mh3kigAqsMUAZP9ECB9LaiO99gL9Tj-xQ@mail.gmail.com>
From: Greg Shepherd <gjshep@gmail.com>
To: Magnus Westerlund <magnus.westerlund@ericsson.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: "mboned@ietf.org" <mboned@ietf.org>, Greg Bumgardner <gbumgard@gmail.com>, "tsv-dir@ietf.org" <tsv-dir@ietf.org>, "iesg@ietf.org" <iesg@ietf.org>, Martin Stiemerling <martin.stiemerling@neclab.eu>
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: gjshep@gmail.com
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Jan 2013 14:04:13 -0000

Why would you require such restriction? If your concern is about
support for RTCP we already have RFC 5760 to address this. It's an
application layer problem.

Greg

On Wed, Jan 16, 2013 at 12:53 AM, Magnus Westerlund
<magnus.westerlund@ericsson.com> wrote:
> Hi,
>
> I think the major question here, are anyone really objecting to
> restricting the applicability of AMT to managed environments, i.e. where
> the full path between relay and client are in domains that are
> explicitly aware that they will have AMT traffic and approve of it. The
> comments appears to indicate that this is where it is deployed. If one
> makes such an applicability statement most of the significant comments
> can be more easily resolved.
>
> Cheers
>
> Magnus
>
> On 2013-01-10 17:35, Greg Shepherd wrote:
>> On Wed, Jan 9, 2013 at 6:15 AM, Martin Stiemerling
>> <martin.stiemerling@neclab.eu> wrote:
>>> Hi Greg,
>>>
>>> I'm joining the discussion as the DISCUSS holder for most of the issues
>>> below.
>>>
>>>
>>> On 01/09/2013 03:26 AM, Greg Bumgardner wrote:
>>>>
>>>> Magnus,
>>>>
>>>> I would like to thank you for providing valuable feedback regarding th=
e
>>>> AMT spec.
>>>>
>>>> I would first like to state that many of your concerns and criticisms
>>>> regarding the protocol were raised during MBONED group discussions. Ma=
ny
>>>> alternative solutions were identified but ultimately rejected by the
>>>> group, primarily due to the reluctance of group members to make
>>>> significant changes to the protocol after several years of slow
>>>> development. My comments are inlined below...
>>>
>>>
>>> Can you give some insights why the protocol development took so long?
>>
>> Yes, I can, though I'm not sure how relevant the incubation period is
>> to its adoption. And since I've breaking multicast networks for 15+
>> years now, asking a question like this may get you more than you were
>> expecting :)
>>
>> The need for multicast transition services is as old as multicast
>> itself. Deering's first work - DVMRP - had to assume there was no
>> 'native' path and so incorporated encapsulation as an integral part of
>> the protocol. I believe DVMRP may be the first use of overlay
>> encapsulation in the IETF, but I welcome you to dig for your self and
>> back with me. It's the earliest I've been able to find.
>>
>> Once we moved to PIM, encapsulation for receiver delivery was thrown
>> out and we all pushed to "go native". The reality was - and still is -
>> that those most wanting end-to-end native multicast (content owners)
>> had no way to motivate network operators to enable multicast services.
>> Deployment experience with PIM revealed to many the need for
>> encapsulation overlay mechanism, and from this several drafts were
>> introduced.
>>
>> In 2003 we held a BOF at IETF57 with the intention of agreeing on a
>> list of requirements and consolidating the efforts into a single
>> draft. We selected the AMT draft as being the closest fit, though
>> significant changes still need to be addressed. You see the list of
>> drafts and read the BOF minutes here:
>>
>> http://www.ietf.org/meeting/57/session/mlm.txt
>>
>> Authors came and went.. Progress was slow, until ATT picked up the
>> torch, demonstrated running code, and outlined their deployment plans.
>> Greg Bumgardner also wrote a java Gateway implementation and took over
>> as primary author. Finally we had solid industry support and a
>> qualified and motivated set of contributors.
>>
>> Last year we finally reached WG consensus, along with at least three
>> shipping implementations of both the Relay and the Gateway , and at
>> least one active and growing commercial deployment of a
>> IPMulticast/AMT service. I doubt many drafts in the editors' queue can
>> claim that much support. :)
>>
>>> Anyhow, I can understand the reluctance of changing parts of a protocol
>>> after such a long time. But there are a lot of rough edges in the proto=
col
>>> that need attention before moving it forward on the Standards Track.
>>
>> Time is irrelevant to any reluctance to change - commercial
>> implementations and deployments are, however quite relevant.
>>
>> Cheers,
>> Greg
>>
>>>>
>>>> Thanks,
>>>>
>>>> -g.b.
>>>>
>>>>
>>>> On Wed, Dec 26, 2012 at 5:01 AM, Magnus Westerlund
>>>> <magnus.westerlund@ericsson.com <mailto:magnus.westerlund@ericsson.com=
>>
>>>>
>>>> wrote:
>>>>
>>>>     Hi,
>>>>
>>>>     Sorry for being late with this IETF last call comments. I will
>>>>     partly blame the ADs requesting this Transport Directorate review =
a
>>>>     bit late, the other part is all mine and the holidays. Anyway, I d=
o
>>>>     hope you will consider these issues and comments as I believe I
>>>>     found some serious ones in addition to a number of clarifications
>>>>     that should be made.
>>>>
>>>>
>>>>     Significant Issues:
>>>>
>>>>     1. Congestion Control
>>>>     This is clearly a tunnel establishment protocol of something that =
is
>>>>     IP traffic. Thus normally the responsibility for congestion contro=
l
>>>>     is with the tunnelled traffic. However, I would like argue that th=
is
>>>>     does not apply in this case due to the nature of the tunnelled
>>>>     traffic, i.e. multicast traffic and secondary due to limitations i=
n
>>>>     the tunnel protocol.
>>>>
>>>>     Lets start with the second part. This protocol claims to support A=
SM
>>>>     still don't provide a upstream delivery mechanism, i.e. an ASM
>>>>     receiver is not capable of sending as it should. This prevents
>>>>     several existing mechanism for congestion control that exist in
>>>>     protocols supporting multicast. The first is using RTCP for
>>>>     congestion control in ASM [RFC3550], the second isTCP-Friendly
>>>>     Multicast Congestion Control [RFC4654] that can be used in the RMT
>>>>     suite of protocols, and I know has been implemented in some NORM
>>>>     implementation. Thus onlystrictly receiver based mechanisms, such =
as
>>>>
>>>>     Wave and Equation Based Rate Control [RFC3738] are available in th=
is
>>>>     context.
>>>>
>>>>
>>>> You are correct, the AMT protocol provides no method for injecting
>>>> multicast traffic into a multicast enabled network, and thus does not
>>>> provide the functionality required for something like ASM RTP/RTCP
>>>> teleconferencing. The protocol is primarily intended for use in
>>>> receiving SSM traffic. Because the AMT protocol uses IGMP/MLD messages
>>>> to establish reception/subscription state, the protocol MAY be used to
>>>> subscribe to ASM broadcasts, but cannot generate them.
>>>>
>>>> I should note that earlier versions of the AMT spec described a method
>>>> for broadcasting SSM via a gateway/relay tunnel, but this capability w=
as
>>>> removed due to feasibility considerations.
>>>
>>>
>>> This means that the draft should state that it does not support ASM, bu=
t
>>> only a subset of it. Further, it cuts away the feedback over RTCP and R=
FC
>>> 4654.
>>>
>>>>
>>>>
>>>>     Secondly, many multicast usages are in fact deployed without any
>>>>     congestion control. This based on that the deploying entity contro=
ls
>>>>     the scope and authorization for requesting multicast delivery.
>>>>     However, does restrictions does not apply to AMT delivery of
>>>>     multicast. If the gateway can reach using unicast the relay it can
>>>>     be delivered the multicast group from the domain the relay is
>>>>     attached to. Thus, this protocol changes the deployment restrictio=
ns
>>>>     of multicast which many non-congestion controlled delivery is base=
d
>>>>     on. Instead the non-congestion controlled trafficcan now sent over
>>>>
>>>>     an IP/UDP tunnel over Internet where neither relay nor gateway may
>>>>     have any knowledge about the path the traffic may take.
>>>>
>>>>
>>>> The AMT protocol does not attempt to address the issue of congestion
>>>> control. The group decided to treat the detection and remediation of
>>>> congestion (of encapsulated multicast packets) as an application-layer
>>>> responsibility. While we might have attempted to provide some level of
>>>> QoS control on tunnel packets by copying ToS/DS bits from inner to out=
er
>>>> headers, this would only occur on a packet-by-packet basis on not on t=
he
>>>> tunnel as a whole.
>>>
>>>
>>> What's the group's view on how the application layer is handling conges=
tion
>>> control in this case when using AMT?
>>>
>>>
>>>>
>>>>
>>>>     Based on this I would like to see two changes to this protocol
>>>>     specification. First a section discussing the issue of congestion
>>>>     control.
>>>>
>>>>
>>>> I'm not sure what, if anything, I might add to address this. I think t=
he
>>>> spec makes it pretty clear that the tunneling is provided by best-effo=
rt
>>>> delivery of UDP packets (perhaps restricted by relay/router
>>>> configuration). Perhaps you could suggest some text.
>>>
>>>
>>> There are multiple issues here:
>>> - there is no feedback loop between relay and gateway so that any
>>> information about the tunneled multicast traffic can be sent back from =
the
>>> gateway to the relay. This feedback loop can be used for congestion con=
trol,
>>> but also to figure out if the non-zero IPv6/UDP traffic gets black hole=
d
>>> somewhere.
>>> - the tunneled traffic can exceed the capacity of the network path betw=
een
>>> relay and gateway. This needs to be monitored and acted upon any conges=
tion.
>>> - since a single relay can feed many gateways, even tunneled multicast =
flows
>>> from the same relay can compete for network path capacity. I.e., if the
>>> tunneled traffic shares a network path. In the worst case, an added
>>> multicast flow can lead to congestion for all other tunneled multicast
>>> flows.
>>> - any tunneled flow can impact other traffic from other hosts in the
>>> Internet.
>>>
>>>>
>>>>     Secondly, I think this protocol should have an applicability
>>>>     statement limiting its deployment to restricted environments where
>>>>     the relay and gateway deployers can provide certain resource
>>>>     provisions between the entities to avoid the multicast traffic
>>>>     affecting other traffic sharing the same bottlenecksin ways not
>>>>
>>>>     allowed by the network provider.
>>>>
>>>>
>>>> While I understand your concern, this would essentially negate the
>>>> entire reason for the protocol, which is to provide multicast transit =
to
>>>> non-multicast connected receivers. However, the AMT spec does not
>>>> preclude the implementation/deployment of AMT-enabled gateway devices
>>>> for which AMT support may be enabled or disabled (or blocked via a
>>>> firewall).
>>>
>>>
>>> The point is that it is not safe to operate in the public Internet with=
 the
>>> current protocol. Either with fix the issues in the protocol or we limi=
t the
>>> applicability.
>>>
>>>
>>>>
>>>>
>>>>
>>>>     2. Security
>>>>
>>>>     This protocols is frank with it having limited security features a=
nd
>>>>     says this is similar to the IGMP and MLD protocols being used.
>>>>     However, I think this is a failure to propoerly consider the threa=
t
>>>>     model. If one uses AMT over general Internet it will run in a
>>>>     network where the one deploying the multicast and the relays no
>>>>     longer control requirements on source address verification or
>>>>     possibilities for traffic separation as they can do within the
>>>>     domains where multicast currently are deployed. The security
>>>>     vulnerabilities in IGMP and MLD are much more contained and
>>>>     controllable in a LAN environment where one has chosen to deploy
>>>>     multicast compared to an Relay exposing this to the whole Internet=
.
>>>>     Once more I think there is only two choices here.
>>>>
>>>>     A) Beef up the security to general Internet threat model, i.e. at =
a
>>>>     minimal provide a real model for gateway authentication using
>>>>     identities, not only return routability based verifications.
>>>>
>>>>     B) Limit the applicability of AMT to managed environments and make
>>>>     it clear that the relay will need to limit which gateways are
>>>>     allowed to access the relay based on addressing.
>>>>
>>>>     Based on the first significant issue with congestion control I
>>>>     expect that there is little meaning to do A) unless also one is
>>>>     willing to beef up AMT to provide congestion control. Which I thin=
k
>>>>     is not according to the design wishes for the protocol designers.
>>>>
>>>>
>>>> The security topic was visited and revisited numerous times over the
>>>> last couple of years. The group made the decision to limit the protoco=
l
>>>> to simple source address verification and leave it at that. An emphasi=
s
>>>> was placed on protecting relays against simple DoS attacks but not
>>>> gateways.
>>>
>>>
>>> I can see the point, but I assume that most people are just looking to
>>> closed environments, e.g., multicast in access networks, such as xDSL o=
r
>>> 3GPP. However, AMT assumes to run across the public Internet and threat=
s are
>>> extremely different there. Magnus made a good proposal that is relative=
ly
>>> easy to implement.
>>>
>>>>
>>>>
>>>>     3. Use of Zero Checksum
>>>>
>>>>     The AMT specification enables the use of Zero UDP checksum with
>>>>     IPv6, i.e.
>>>>     draft-ietf-6man-udpchecksums-06
>>>>     <http://datatracker.ietf.org/doc/draft-ietf-6man-udpchecksums/> an=
d
>>>>     draft-ietf-6man-udpzero-08
>>>>     <http://datatracker.ietf.org/doc/draft-ietf-6man-udpzero/>. Nothin=
g
>>>>
>>>>     against this in principal. However, I have noticed that AMT fails =
to
>>>>     properly address the failure modes of using a zero-checksum. AMT i=
s
>>>>     a typical example of a protocol that actually need active
>>>>     verification of each tunnel that zero-checksum functions. This as
>>>>     AMT is clearly intended to be deployed with its Gateway part in
>>>>     end-hosts and residential network devices or routers. This means t=
he
>>>>     tunnel will pass through both firewalls and NATs on its path betwe=
en
>>>>     the relay and the gateway. Unless these devices are not upgraded t=
o
>>>>     support zero-checksum in UDP for IPv6 the traffic may actually
>>>>     become black holed. The most likely is a simple firewall that has =
a
>>>>     rule for IPv6/UDP which doesn't allow zero checksum as it is again=
st
>>>>     RFC 2460. Thus all the Multicast data packets will disappear on
>>>>     route in the tunnel. There is no mechanism in the AMT protocol to
>>>>     detect this and negotiate with the relay so that it will not use t=
he
>>>>     zero-checksum for this tunnel.
>>>>
>>>>     This must be addressed as I see it. If not the AMT will be so
>>>>     brittle that it can't be used in a large number of its intended
>>>>     deployments.
>>>>
>>>>
>>>> You probably have no idea how often this issue was raised and discusse=
d
>>>> within the group. The consensus was that there was no single solution =
to
>>>> the problem. The only workable solution I can suggest is to use an IPv=
4
>>>> tunnel to carry the IPv6 packets. The mechanism for deciding whether (=
or
>>>> if) to use an IPv4 tunnel is considered out of scope.
>>>
>>>
>>> There is a solution, as mentioned above: feedback loop. If the relay sa=
ys
>>> "sending you x packets" and the gateway sees zero, there is an issue. I=
n the
>>> easiest case, a log file entry is generate saying that the tunneled tra=
ffic
>>> couldn't reach this particular gateway.
>>>
>>>
>>>>
>>>>
>>>>     4) MTU issues
>>>>
>>>>     This document total fails to discuss the issues of MTU blackholing=
.
>>>>     As the IP multicast datagrams as well as the encapsulated IGMP/MLD
>>>>     messages can with the added tunnel-overhead result in that the sen=
t
>>>>     packet exceeds the MTU of the path, these packets could be black
>>>>     holed. This can potentially result in very intermittent transport
>>>>     behavior for the tunnel. Thus, some discussion of how do handle th=
e
>>>>     MTU issues in this context should be introduced.
>>>>
>>>>
>>>> Yep. This could be a problem that has no easy solution. Not only might
>>>> the outer packets be fragmented, but the encapsulated packets might be
>>>> fragments. Since the protocol obviously has no control over the size o=
f
>>>> source packets, it cannot really avoid the potential need to fragment
>>>> the tunnel packets. A gateway can only rely on the underlying UDP/IP
>>>> stack to provide the necessary reassembly of the tunnel packets. Once
>>>> the tunnel packet has been reassembled, the gateway, host IP stack or
>>>> application may still need to reassemble the contents of a sequence of
>>>> tunnel packets.
>>>
>>>
>>> In the case the WG doesn't have any solution, which I couldn't believe,=
 it
>>> is worth to document the issues. One operational solution can be to lim=
it
>>> the MTU in the native multicast network, if this part and the relay are
>>> under the same administrative control. However, this would require that=
 the
>>> AMT relay logs any issue with the MTU accross the tunnel.
>>> Further, Magnus point is about that the tunneled packets can get fragme=
nted,
>>> less about the inner parts of the tunnel.
>>>
>>> Addressing the MTU issue will increase the viability of this a lot.
>>>
>>>
>>>>
>>>>
>>>>     I am willing to discuss methods here, but I guess several
>>>>     alternatives exists and thus which is most appropriate and the lev=
el
>>>>     of AMT support for them varies I would like the protocol designers
>>>>     to do a first stab at resolving this.
>>>>
>>>>     Other Issues
>>>>     =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>>>>
>>>>     A) The table of content on page 2 should include more levels of
>>>>     headings. Most likely down to 4 is needed to make the TOC usable f=
or
>>>>     finding content in the document.
>>>>
>>>>     B) The claimed ASM support
>>>>     I would like to better understand how one can claim to support ASM
>>>>     when one has no up stream path to inject the ASM group participant=
s
>>>>     traffic. When one uses ASM one normally does this for a reason and
>>>>     needing the possibility to inject packets into the group. These
>>>>     limitations needs to be clarified.
>>>>
>>>>
>>>> See above. The protocol is intended for multicast reception. It makes =
no
>>>> assumptions about application layer use of the multicast traffic. It c=
an
>>>> only be used to receive ASM IP traffic.
>>>
>>>
>>> Document it, please. Otherwise, the draft is incomplete.
>>>
>>> Thank you,
>>>
>>>   Martin
>>> --
>>> IETF Transport Area Director
>>>
>>> martin.stiemerling@neclab.eu
>>>
>>> NEC Laboratories Europe - Network Research Division NEC Europe Limited
>>> Registered Office: NEC House, 1 Victoria Road, London W3 6BL
>>> Registered in England 283
>>>
>>> _______________________________________________
>>> MBONED mailing list
>>> MBONED@ietf.org
>>> https://www.ietf.org/mailman/listinfo/mboned
>>
>
>
> --
>
> Magnus Westerlund
>
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM
> ----------------------------------------------------------------------
> Ericsson AB                | Phone  +46 10 7148287
> F=E4r=F6gatan 6                | Mobile +46 73 0949079
> SE-164 80 Stockholm, Sweden| mailto: magnus.westerlund@ericsson.com
> ----------------------------------------------------------------------
>

From gjshep@gmail.com  Wed Jan 16 07:19:22 2013
Return-Path: <gjshep@gmail.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id CFE6321F8938 for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 07:19:22 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -102.6
X-Spam-Level: 
X-Spam-Status: No, score=-102.6 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, NO_RELAYS=-0.001, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id M5pqSoOMNv+i for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 07:19:22 -0800 (PST)
Received: from mail-wg0-x229.google.com (mail-wg0-x229.google.com [IPv6:2a00:1450:400c:c00::229]) by ietfa.amsl.com (Postfix) with ESMTP id 1A72721F850B for <mboned@ietf.org>; Wed, 16 Jan 2013 07:19:21 -0800 (PST)
Received: by mail-wg0-f41.google.com with SMTP id ds1so2753055wgb.4 for <mboned@ietf.org>; Wed, 16 Jan 2013 07:19:20 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:reply-to:in-reply-to:references:date :message-id:subject:from:to:content-type:content-transfer-encoding; bh=5dgKpoyTTy4sxkMcQn5whN1ulpeMigJ/vIthyLnDhTQ=; b=v5gTDG4CheKP4jasF7t4WrFvOxS5U8LDxVvEKPiwES+dbcp3KlqUS/iViXDi/8wLTs VhTFUgiDYpVzttt+00hv1tNH1I7zpnsw8K4vZFprepbfYtzSz6t/geRdjDYa2fcepy0I cFSxI4bH3ZcLUxC4vsAeC2Y7oWh8PYPavwJaB5uN4IZ1o+EmO1hy7inbx9z8L6VLf1Vf Mg2OrqJaHWLKXMfPfIpbrsC2OGqlv/07GpHUF18y8z5PKqcawcM+D9pDSLWHZ/Cv32s3 buRvJ4OmAgei0QX7D20MucTQdDLh4Jc7GiREMHJcDvh/uw8s8K8Ta/8oPWhl9TvwLl6p fDeQ==
MIME-Version: 1.0
X-Received: by 10.194.19.10 with SMTP id a10mr2847660wje.45.1358349560700; Wed, 16 Jan 2013 07:19:20 -0800 (PST)
Received: by 10.180.97.66 with HTTP; Wed, 16 Jan 2013 07:19:20 -0800 (PST)
In-Reply-To: <D4D47BCFFE5A004F95D707546AC0D7E91F633F7E@SACEXCMBX01-PRD.hq.netapp.com>
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se> <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com> <50ED7B8D.3010904@neclab.eu> <CABFReBoCTN-KP=LHrVy+Bg4y6g4aezUH-2rWzvZ0GVmAtejERA@mail.gmail.com> <50F66AA5.6050705@ericsson.com> <CABFReBpiT3=LJbS8_Mh3kigAqsMUAZP9ECB9LaiO99gL9Tj-xQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633B44@SACEXCMBX01-PRD.hq.netapp.com> <CABFReBqixqSCYhQ6fP=81rjWw_eYePfgW2FsFfu+Xm55sz+DsQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633DDA@SACEXCMBX01-PRD.hq.netapp.com> <CABFReBpyt4gaRF5k19wf8NBP6FGOJjS8OewheXvUNXbt5OQTnQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633F7E@SACEXCMBX01-PRD.hq.netapp.com>
Date: Wed, 16 Jan 2013 07:19:20 -0800
Message-ID: <CABFReBoGr8DiwcJtJy9qNoSjhCt+1Xr12LT5+z4nJSreVxGKtg@mail.gmail.com>
From: Greg Shepherd <gjshep@gmail.com>
To: "Eggert, Lars" <lars@netapp.com>, mboned@ietf.org
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: gjshep@gmail.com
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Jan 2013 15:19:22 -0000

On Wed, Jan 16, 2013 at 7:11 AM, Eggert, Lars <lars@netapp.com> wrote:
> Hi,
>
> On Jan 16, 2013, at 16:04, Greg Shepherd <gjshep@gmail.com> wrote:
>> So some MUST have congestion control text in the spec would be
>> sufficient?
>
> Or an applicability statement that limits AMT to managed environments, as=
 Magnus has pointed out.

That is far from the target deployment. I think an applicability
statement requiring the use of an application-layer CG mechanism
should be enough. It would carry no less weight than a 'managed env'
statement and would actually be more helpful.

>> How would you suggest detecting this without overloading
>> the protocol?
>
> I don't know what you mean with "overloading the protocol." Obviously, yo=
u'll need to have some mechanism that detects loss and some mechanism that =
tracks the bandwidth use of the tunneled flow. It's going to need some prot=
ocol machinery. And, I guess TSVWG would really want to review that algorit=
hm in detail.

AMT is architected to minimize the load on the Relay, so that current
router equipment can provide Relay services with minimum impact.
Having to monitor every flow to detect CG would be a ridiculous amount
of overloading and render the protocol useless.

> Lars

From magnus.westerlund@ericsson.com  Wed Jan 16 08:52:30 2013
Return-Path: <magnus.westerlund@ericsson.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id F336721F8A6C; Wed, 16 Jan 2013 08:52:29 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -106.099
X-Spam-Level: 
X-Spam-Status: No, score=-106.099 tagged_above=-999 required=5 tests=[AWL=0.150, BAYES_00=-2.599, HELO_EQ_SE=0.35, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 1b6yd1lirzil; Wed, 16 Jan 2013 08:52:29 -0800 (PST)
Received: from mailgw7.ericsson.se (mailgw7.ericsson.se [193.180.251.48]) by ietfa.amsl.com (Postfix) with ESMTP id 5E5B621F8919; Wed, 16 Jan 2013 08:52:28 -0800 (PST)
X-AuditID: c1b4fb30-b7f0d6d000007e61-67-50f6dacaedf8
Received: from esessmw0247.eemea.ericsson.se (Unknown_Domain [153.88.253.124]) by mailgw7.ericsson.se (Symantec Mail Security) with SMTP id 42.A5.32353.ACAD6F05; Wed, 16 Jan 2013 17:52:27 +0100 (CET)
Received: from [127.0.0.1] (153.88.115.8) by esessmw0247.eemea.ericsson.se (153.88.115.94) with Microsoft SMTP Server id 8.3.279.1; Wed, 16 Jan 2013 17:52:26 +0100
Message-ID: <50F6DACA.3010006@ericsson.com>
Date: Wed, 16 Jan 2013 17:52:26 +0100
From: Magnus Westerlund <magnus.westerlund@ericsson.com>
User-Agent: Mozilla/5.0 (Windows NT 6.1; rv:17.0) Gecko/20130107 Thunderbird/17.0.2
MIME-Version: 1.0
To: gjshep@gmail.com
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se> <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com> <50ED7B8D.3010904@neclab.eu> <CABFReBoCTN-KP=LHrVy+Bg4y6g4aezUH-2rWzvZ0GVmAtejERA@mail.gmail.com> <50F66AA5.6050705@ericsson.com> <CABFReBpiT3=LJbS8_Mh3kigAqsMUAZP9ECB9LaiO99gL9Tj-xQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633B44@SACEXCMBX01-PRD.hq.netapp.com> <CABFReBqixqSCYhQ6fP=81rjWw_eYePfgW2FsFfu+Xm55sz+DsQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633DDA@SACEXCMBX01-PRD.hq.netapp.com> <CABFReBpyt4gaRF5k19wf8NBP6FGOJjS8OewheXvUNXbt5OQTnQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633F7E@SACEXCMBX01-PRD.hq.netapp.com> <CABFReBoGr8DiwcJtJy9qNoSjhCt+1Xr12LT5+z4nJSreVxGKtg@mail.gmail.com>
In-Reply-To: <CABFReBoGr8DiwcJtJy9qNoSjhCt+1Xr12LT5+z4nJSreVxGKtg@mail.gmail.com>
X-Enigmail-Version: 1.5
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 8bit
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFprHLMWRmVeSWpSXmKPExsUyM+Jvje7pW98CDI59ZbTYMifP4sXrHhaL KS2nGS2+XbnB7sDisXPWXXaPJUt+MnnM+PSFLYA5issmJTUnsyy1SN8ugStj9op+1oJr4hVn +3MbGOcJdzFyckgImEh8ebaSGcIWk7hwbz1bFyMXh5DASUaJi2eusUM4yxkl1h44zAhSxSug LXF8/jU2EJtFQFVi6csbLCA2m4CFxM0fjWBxUYFgiQ0HVzFB1AtKnJz5BKxGREBE4vqc2axd jBwczAIhEuc3KIPMFxZoZJTob1zDDLGskU3iR0cTO0gDp0CgxM8VGxhBGiQExCXWvOEACTML 6ElMudrCCGHLSzRvnQ32gRDQbQ1NHawTGIVmIVk9C0nLLCQtCxiZVzGy5yZm5qSXm29iBIbx wS2/DXYwbrovdohRmoNFSZw33PVCgJBAemJJanZqakFqUXxRaU5q8SFGJg5OqQbG1i+HlVcw WlZw+zatts1ikMo9KWRbtaREv7XU+MHzj4dXpEt9fcCkktUf8cnY02l9z9eZdveuipv2rXhX JzP1VfTcoD+SdktP/p82WXH9L3fpRa8ZJRVmWdnelJ317eoq91eK5/Ys63iYdHn1ndfpL9bX ++r5WqlI3VK86qL5om3Xgh1czy8uVGIpzkg01GIuKk4EAHwt290xAgAA
Cc: mboned@ietf.org, "Eggert, Lars" <lars@netapp.com>, TSV Dir <tsv-dir@ietf.org>
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Jan 2013 16:52:30 -0000

On 2013-01-16 16:19, Greg Shepherd wrote:
> On Wed, Jan 16, 2013 at 7:11 AM, Eggert, Lars <lars@netapp.com>
> wrote:
>> Hi,
>> 
>> On Jan 16, 2013, at 16:04, Greg Shepherd <gjshep@gmail.com> wrote:
>>> So some MUST have congestion control text in the spec would be 
>>> sufficient?
>> 
>> Or an applicability statement that limits AMT to managed
>> environments, as Magnus has pointed out.
> 
> That is far from the target deployment. I think an applicability 
> statement requiring the use of an application-layer CG mechanism 
> should be enough. It would carry no less weight than a 'managed env' 
> statement and would actually be more helpful.

Yes, the restriction can also be on the application layer to require
them to congestion controlled. But frankly how many of the multicast
deployed applications actually are congestion controlled? IPTV using
MPEG transport streams over UDP is definitely not congestion controlled.

Sure, they should be congestion controlled, however IPTV is deployed in
managed networks where the operator deploying the IPTV has the full
control and responsibility for the path. With AMT I can start breaking
this by having the AMT client be located far from the network the relay
belongs to.

With AMT not supporting ASM, the solutions for congestion control that
can function are the receiver driven ones and the ones that uses its own
unicast feedback to deal with SSM.


> 
>>> How would you suggest detecting this without overloading the
>>> protocol?
>> 
>> I don't know what you mean with "overloading the protocol."
>> Obviously, you'll need to have some mechanism that detects loss and
>> some mechanism that tracks the bandwidth use of the tunneled flow.
>> It's going to need some protocol machinery. And, I guess TSVWG
>> would really want to review that algorithm in detail.
> 
> AMT is architected to minimize the load on the Relay, so that
> current router equipment can provide Relay services with minimum
> impact. Having to monitor every flow to detect CG would be a
> ridiculous amount of overloading and render the protocol useless.

Might be so. But AMT is creating the issue of moving likely
non-congestion controlled traffic into the general Internet without
making any attempt to address that aspect. That is not acceptable in a
standards track protocol. Not even in an experimental one that hasn't
have limited applicability. Especially as we are talking about
applications that commonly include multi mbps packet streams such as HD
TV etc.

Cheers

Magnus Westerlund

----------------------------------------------------------------------
Multimedia Technologies, Ericsson Research EAB/TVM
----------------------------------------------------------------------
Ericsson AB                | Phone  +46 10 7148287
Färögatan 6                | Mobile +46 73 0949079
SE-164 80 Stockholm, Sweden| mailto: magnus.westerlund@ericsson.com
----------------------------------------------------------------------


From lars@netapp.com  Wed Jan 16 06:49:18 2013
Return-Path: <lars@netapp.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5529721F8923; Wed, 16 Jan 2013 06:49:18 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -10.486
X-Spam-Level: 
X-Spam-Status: No, score=-10.486 tagged_above=-999 required=5 tests=[AWL=0.113, BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id gElkjvqQV9-u; Wed, 16 Jan 2013 06:49:17 -0800 (PST)
Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id DC3E921F876E; Wed, 16 Jan 2013 06:49:17 -0800 (PST)
X-IronPort-AV: E=Sophos;i="4.84,479,1355126400";  d="scan'208";a="8326177"
Received: from smtp2.corp.netapp.com ([10.57.159.114]) by mx12-out.netapp.com with ESMTP; 16 Jan 2013 06:49:17 -0800
Received: from vmwexceht05-prd.hq.netapp.com (vmwexceht05-prd.hq.netapp.com [10.106.77.35]) by smtp2.corp.netapp.com (8.13.1/8.13.1/NTAP-1.6) with ESMTP id r0GEnG7T019373; Wed, 16 Jan 2013 06:49:16 -0800 (PST)
Received: from SACEXCMBX01-PRD.hq.netapp.com ([169.254.2.51]) by vmwexceht05-prd.hq.netapp.com ([10.106.77.35]) with mapi id 14.02.0328.009; Wed, 16 Jan 2013 06:49:16 -0800
From: "Eggert, Lars" <lars@netapp.com>
To: "<gjshep@gmail.com>" <gjshep@gmail.com>
Thread-Topic: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
Thread-Index: AQHN7nPmrGMcaFMY+U6WG/9TZUlzTZhDSjmAgAjtBYCAAFapgIAADJmA
Date: Wed, 16 Jan 2013 14:49:15 +0000
Message-ID: <D4D47BCFFE5A004F95D707546AC0D7E91F633B44@SACEXCMBX01-PRD.hq.netapp.com>
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se> <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com> <50ED7B8D.3010904@neclab.eu> <CABFReBoCTN-KP=LHrVy+Bg4y6g4aezUH-2rWzvZ0GVmAtejERA@mail.gmail.com> <50F66AA5.6050705@ericsson.com> <CABFReBpiT3=LJbS8_Mh3kigAqsMUAZP9ECB9LaiO99gL9Tj-xQ@mail.gmail.com>
In-Reply-To: <CABFReBpiT3=LJbS8_Mh3kigAqsMUAZP9ECB9LaiO99gL9Tj-xQ@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.106.53.51]
Content-Type: text/plain; charset="iso-8859-1"
Content-ID: <5BCBAE531BE3194EB8AB2D0F28449D50@tahoe.netapp.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Mailman-Approved-At: Wed, 16 Jan 2013 09:09:39 -0800
Cc: "mboned@ietf.org" <mboned@ietf.org>, Greg Bumgardner <gbumgard@gmail.com>, "iesg@ietf.org" <iesg@ietf.org>, "tsv-dir@ietf.org" <tsv-dir@ietf.org>
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt>	(Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Jan 2013 14:49:18 -0000

On Jan 16, 2013, at 15:04, Greg Shepherd <gjshep@gmail.com> wrote:
> Why would you require such restriction? If your concern is about
> support for RTCP we already have RFC 5760 to address this. It's an
> application layer problem.

AMT is a tunneling scheme that can cause non-congestion controlled traffic =
to be tunneled over the wider Internet. Internet flows need to implement a =
congestion control mechanism (RFC2914). Because AMT can be used transparent=
ly, i.e., without the application knowing, this makes it AMT's problem and =
not the application's. (In other words, without AMT, that traffic would not=
 hit the wider Internet.)

The only scenario where it's OK to run non-congestion-controlled flows over=
 the wider Internet is if the path capacity has been provisioned accordingl=
y (i.e., in a managed environment).

Lars=

From lars@netapp.com  Wed Jan 16 07:25:01 2013
Return-Path: <lars@netapp.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 80E0B21F8A2F for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 07:25:01 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -10.49
X-Spam-Level: 
X-Spam-Status: No, score=-10.49 tagged_above=-999 required=5 tests=[AWL=0.109,  BAYES_00=-2.599, RCVD_IN_DNSWL_HI=-8]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 0YjusibIUMK0 for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 07:25:01 -0800 (PST)
Received: from mx12.netapp.com (mx12.netapp.com [216.240.18.77]) by ietfa.amsl.com (Postfix) with ESMTP id 1ECF121F8A25 for <mboned@ietf.org>; Wed, 16 Jan 2013 07:25:01 -0800 (PST)
X-IronPort-AV: E=Sophos;i="4.84,479,1355126400";  d="scan'208";a="8336551"
Received: from smtp1.corp.netapp.com ([10.57.156.124]) by mx12-out.netapp.com with ESMTP; 16 Jan 2013 07:25:00 -0800
Received: from vmwexceht02-prd.hq.netapp.com (vmwexceht02-prd.hq.netapp.com [10.106.76.240]) by smtp1.corp.netapp.com (8.13.1/8.13.1/NTAP-1.6) with ESMTP id r0GFP04T010680; Wed, 16 Jan 2013 07:25:00 -0800 (PST)
Received: from SACEXCMBX01-PRD.hq.netapp.com ([169.254.2.51]) by vmwexceht02-prd.hq.netapp.com ([10.106.76.240]) with mapi id 14.02.0328.009; Wed, 16 Jan 2013 07:25:00 -0800
From: "Eggert, Lars" <lars@netapp.com>
To: "<gjshep@gmail.com>" <gjshep@gmail.com>
Thread-Topic: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
Thread-Index: AQHN8/l5bAy8oCiRhkSmyD2C9FZdzZhMkuqAgAAAtgCAAAIcgIAAAiMAgAABkoA=
Date: Wed, 16 Jan 2013 15:24:59 +0000
Message-ID: <D4D47BCFFE5A004F95D707546AC0D7E91F634186@SACEXCMBX01-PRD.hq.netapp.com>
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se> <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com> <50ED7B8D.3010904@neclab.eu> <CABFReBoCTN-KP=LHrVy+Bg4y6g4aezUH-2rWzvZ0GVmAtejERA@mail.gmail.com> <50F66AA5.6050705@ericsson.com> <CABFReBpiT3=LJbS8_Mh3kigAqsMUAZP9ECB9LaiO99gL9Tj-xQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633B44@SACEXCMBX01-PRD.hq.netapp.com> <CABFReBqixqSCYhQ6fP=81rjWw_eYePfgW2FsFfu+Xm55sz+DsQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633DDA@SACEXCMBX01-PRD.hq.netapp.com> <CABFReBpyt4gaRF5k19wf8NBP6FGOJjS8OewheXvUNXbt5OQTnQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633F7E@SACEXCMBX01-PRD.hq.netapp.com> <CABFReBoGr8DiwcJtJy9qNoSjhCt+1Xr12LT5+z4nJSreVxGKtg@mail.gmail.com>
In-Reply-To: <CABFReBoGr8DiwcJtJy9qNoSjhCt+1Xr12LT5+z4nJSreVxGKtg@mail.gmail.com>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
x-originating-ip: [10.106.53.51]
Content-Type: text/plain; charset="iso-8859-1"
Content-ID: <20005999DD54384A91CAE962FD445C8B@tahoe.netapp.com>
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-Mailman-Approved-At: Wed, 16 Jan 2013 09:09:39 -0800
Cc: "<mboned@ietf.org>" <mboned@ietf.org>
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Jan 2013 15:25:01 -0000

Hi,

On Jan 16, 2013, at 16:19, Greg Shepherd <gjshep@gmail.com>
 wrote:
>=20
> That is far from the target deployment. I think an applicability
> statement requiring the use of an application-layer CG mechanism
> should be enough. It would carry no less weight than a 'managed env'
> statement and would actually be more helpful.

is there a multicast CC algorithm that can be pointed to as an existence pr=
oof?

> AMT is architected to minimize the load on the Relay, so that current
> router equipment can provide Relay services with minimum impact.
> Having to monitor every flow to detect CG would be a ridiculous amount
> of overloading and render the protocol useless.

So maybe the requirements that have been placed on AMT cannot be met under =
the principles that we all have come to consensus on for the Internet. In t=
hat case, those principles win. RFC2914 predates (even) the first AMT draft=
, so this shouldn't really be news.

Lars=

From gjshep@gmail.com  Wed Jan 16 12:23:01 2013
Return-Path: <gjshep@gmail.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id B5BDF21F8812; Wed, 16 Jan 2013 12:23:00 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.1
X-Spam-Level: 
X-Spam-Status: No, score=-103.1 tagged_above=-999 required=5 tests=[AWL=0.500,  BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id 3P-aCbi4sVxH; Wed, 16 Jan 2013 12:22:59 -0800 (PST)
Received: from mail-wg0-f52.google.com (mail-wg0-f52.google.com [74.125.82.52]) by ietfa.amsl.com (Postfix) with ESMTP id 5E69621F872C; Wed, 16 Jan 2013 12:22:57 -0800 (PST)
Received: by mail-wg0-f52.google.com with SMTP id 12so1118429wgh.31 for <multiple recipients>; Wed, 16 Jan 2013 12:22:52 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:reply-to:in-reply-to:references:date :message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=ctLidP1Lgz8LmJCVwKage9gtXJdx7MwVo71z4vFpXU4=; b=qSExTviPqFIOeNylproHJ9/f8C3G3iuh3jZcoCtzfaW820IYoNJ3wnALph5AohsRv2 wTGDE7W1HeZgfXdMrIIS8GqQMBzIfIxsiWsSwyFM7oeJj8BYBXDT87o90/9iG2APO+If X0J5NQXBo1oBxUdkpoJ8oh6Zula2nmRXl+GpenD6MMavkMTxnjj4/GtIMar80U4i6HJk a/+BnxPJPVAt5z+S2LRN6FxtmEXfIc8IPx8FScCNH30lr/oVBBeQTQAoU7hHKGGwBjkb yfqtm3qhtgiH72o49dHqIKbeiGcH06uinHTOkfXjaOIxveznDW2aHkP4wdgz8H0PWQdw qeuQ==
MIME-Version: 1.0
X-Received: by 10.180.77.68 with SMTP id q4mr4434375wiw.10.1358367772721; Wed, 16 Jan 2013 12:22:52 -0800 (PST)
Received: by 10.180.97.66 with HTTP; Wed, 16 Jan 2013 12:22:52 -0800 (PST)
In-Reply-To: <50F6DACA.3010006@ericsson.com>
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se> <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com> <50ED7B8D.3010904@neclab.eu> <CABFReBoCTN-KP=LHrVy+Bg4y6g4aezUH-2rWzvZ0GVmAtejERA@mail.gmail.com> <50F66AA5.6050705@ericsson.com> <CABFReBpiT3=LJbS8_Mh3kigAqsMUAZP9ECB9LaiO99gL9Tj-xQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633B44@SACEXCMBX01-PRD.hq.netapp.com> <CABFReBqixqSCYhQ6fP=81rjWw_eYePfgW2FsFfu+Xm55sz+DsQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633DDA@SACEXCMBX01-PRD.hq.netapp.com> <CABFReBpyt4gaRF5k19wf8NBP6FGOJjS8OewheXvUNXbt5OQTnQ@mail.gmail.com> <D4D47BCFFE5A004F95D707546AC0D7E91F633F7E@SACEXCMBX01-PRD.hq.netapp.com> <CABFReBoGr8DiwcJtJy9qNoSjhCt+1Xr12LT5+z4nJSreVxGKtg@mail.gmail.com> <50F6DACA.3010006@ericsson.com>
Date: Wed, 16 Jan 2013 12:22:52 -0800
Message-ID: <CABFReBoUzOYrU-3iwtMN01KY=5GZn5qeC6fNf2iM5o6Xa0cj_Q@mail.gmail.com>
From: Greg Shepherd <gjshep@gmail.com>
To: Magnus Westerlund <magnus.westerlund@ericsson.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: mboned@ietf.org, "Eggert, Lars" <lars@netapp.com>, TSV Dir <tsv-dir@ietf.org>
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: gjshep@gmail.com
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Jan 2013 20:23:01 -0000

On Wed, Jan 16, 2013 at 8:52 AM, Magnus Westerlund
<magnus.westerlund@ericsson.com> wrote:
> On 2013-01-16 16:19, Greg Shepherd wrote:
>> On Wed, Jan 16, 2013 at 7:11 AM, Eggert, Lars <lars@netapp.com>
>> wrote:
>>> Hi,
>>>
>>> On Jan 16, 2013, at 16:04, Greg Shepherd <gjshep@gmail.com> wrote:
>>>> So some MUST have congestion control text in the spec would be
>>>> sufficient?
>>>
>>> Or an applicability statement that limits AMT to managed
>>> environments, as Magnus has pointed out.
>>
>> That is far from the target deployment. I think an applicability
>> statement requiring the use of an application-layer CG mechanism
>> should be enough. It would carry no less weight than a 'managed env'
>> statement and would actually be more helpful.
>
> Yes, the restriction can also be on the application layer to require
> them to congestion controlled.

Excellent.

> But frankly how many of the multicast
> deployed applications actually are congestion controlled? IPTV using
> MPEG transport streams over UDP is definitely not congestion controlled.

Is the absolute count of solutions relevant? I can get you a count if
that is deemed necessary. With the expanding deployment of AMT and the
above CG requirement I'd expect many more solutions to develop.

> Sure, they should be congestion controlled, however IPTV is deployed in
> managed networks where the operator deploying the IPTV has the full
> control and responsibility for the path. With AMT I can start breaking
> this by having the AMT client be located far from the network the relay
> belongs to.

..and IPTV is within a managed env where there is also end-to-end
network control so mcast discontinuity is less of an issue, and AMT is
not required. If IPTV is to expand beyond the managed env and adopt
AMT to address mcast discontinuity then the content originator will
need to adopt an application layer CG mechanism as required by the
spec.

> With AMT not supporting ASM,..

AMT does not support sourcing from the Gateway - it does support ASM
joins. Just a nit here, but I understand your point.

>..the solutions for congestion control that
> can function are the receiver driven ones and the ones that uses its own
> unicast feedback to deal with SSM.

Agreed. And they exist. Specifying the need to use one is an excellent
suggestion, thanks.

Greg

>>
>>>> How would you suggest detecting this without overloading the
>>>> protocol?
>>>
>>> I don't know what you mean with "overloading the protocol."
>>> Obviously, you'll need to have some mechanism that detects loss and
>>> some mechanism that tracks the bandwidth use of the tunneled flow.
>>> It's going to need some protocol machinery. And, I guess TSVWG
>>> would really want to review that algorithm in detail.
>>
>> AMT is architected to minimize the load on the Relay, so that
>> current router equipment can provide Relay services with minimum
>> impact. Having to monitor every flow to detect CG would be a
>> ridiculous amount of overloading and render the protocol useless.
>
> Might be so. But AMT is creating the issue of moving likely
> non-congestion controlled traffic into the general Internet without
> making any attempt to address that aspect. That is not acceptable in a
> standards track protocol. Not even in an experimental one that hasn't
> have limited applicability. Especially as we are talking about
> applications that commonly include multi mbps packet streams such as HD
> TV etc.
>
> Cheers
>
> Magnus Westerlund
>
> ----------------------------------------------------------------------
> Multimedia Technologies, Ericsson Research EAB/TVM
> ----------------------------------------------------------------------
> Ericsson AB                | Phone  +46 10 7148287
> F=E4r=F6gatan 6                | Mobile +46 73 0949079
> SE-164 80 Stockholm, Sweden| mailto: magnus.westerlund@ericsson.com
> ----------------------------------------------------------------------
>

From lenny@juniper.net  Wed Jan 16 12:37:19 2013
Return-Path: <lenny@juniper.net>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 775A011E80BF for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 12:37:19 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -106.599
X-Spam-Level: 
X-Spam-Status: No, score=-106.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id awYBHk8wTCNP for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 12:37:14 -0800 (PST)
Received: from exprod7og118.obsmtp.com (exprod7og118.obsmtp.com [64.18.2.8]) by ietfa.amsl.com (Postfix) with ESMTP id B58DA11E80AE for <mboned@ietf.org>; Wed, 16 Jan 2013 12:37:14 -0800 (PST)
Received: from P-EMHUB03-HQ.jnpr.net ([66.129.224.36]) (using TLSv1) by exprod7ob118.postini.com ([64.18.6.12]) with SMTP ID DSNKUPcPekoXsANVq3dp8MJNaST41RB3uWw3@postini.com; Wed, 16 Jan 2013 12:37:14 PST
Received: from magenta.juniper.net (172.17.27.123) by P-EMHUB03-HQ.jnpr.net (172.24.192.33) with Microsoft SMTP Server (TLS) id 8.3.213.0; Wed, 16 Jan 2013 12:36:28 -0800
Received: from eng-mail01.juniper.net (eng-mail01.juniper.net [172.17.28.114]) by magenta.juniper.net (8.11.3/8.11.3) with ESMTP id r0GKaR308777	for <mboned@ietf.org>; Wed, 16 Jan 2013 12:36:27 -0800 (PST)	(envelope-from lenny@juniper.net)
Received: by eng-mail01.juniper.net (Postfix, from userid 1709)	id 7B0AF1144E;  Wed, 16 Jan 2013 12:36:27 -0800 (PST)
Received: from localhost (localhost [127.0.0.1])	by eng-mail01.juniper.net (Postfix) with ESMTP id 695441141B	for <mboned@ietf.org>; Wed, 16 Jan 2013 12:36:27 -0800 (PST)
Date: Wed, 16 Jan 2013 12:36:27 -0800
From: Leonard Giuliano <lenny@juniper.net>
To: MBONED WG <mboned@ietf.org>
Message-ID: <20130116123147.A35594@eng-mail01.juniper.net>
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"; format=flowed
Subject: [MBONED] Adoption of draft-mcbride-armd-mcast-overview
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Jan 2013 20:37:19 -0000

We'd like to see if there is consensus from the WG to adopt 
draft-mcbride-armd-mcast-overview as an MBoned document.  In Atlanta, this 
draft was presented and 5 were in favor, none opposed adopting.  Please 
speak up if you support or oppose adopting this draft as a WG document.

The latest version can be found here: 
https://datatracker.ietf.org/doc/draft-mcbride-armd-mcast-overview/


From william.atwood@concordia.ca  Wed Jan 16 15:02:21 2013
Return-Path: <william.atwood@concordia.ca>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 9E1EB11E80A6 for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 15:02:21 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.166
X-Spam-Level: 
X-Spam-Status: No, score=-6.166 tagged_above=-999 required=5 tests=[AWL=0.433,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id xkAXfBeN6GgQ for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 15:02:20 -0800 (PST)
Received: from oldperseverance.encs.concordia.ca (oldperseverance.encs.concordia.ca [132.205.96.94]) by ietfa.amsl.com (Postfix) with ESMTP id 8565511E809A for <mboned@ietf.org>; Wed, 16 Jan 2013 15:02:20 -0800 (PST)
Received: from [IPv6:::1] (bill@poise.encs.concordia.ca [132.205.2.209]) by oldperseverance.encs.concordia.ca (envelope-from william.atwood@concordia.ca) (8.13.7/8.13.7) with ESMTP id r0GN2IWY030494 for <mboned@ietf.org>; Wed, 16 Jan 2013 18:02:18 -0500
Message-ID: <50F7317A.3070708@concordia.ca>
Date: Wed, 16 Jan 2013 18:02:18 -0500
From: John William Atwood <william.atwood@concordia.ca>
Organization: Concordia University, Montreal
User-Agent: Mozilla/5.0 (Windows NT 6.0; rv:17.0) Gecko/20130107 Thunderbird/17.0.2
MIME-Version: 1.0
To: mboned@ietf.org
References: <20130116123147.A35594@eng-mail01.juniper.net>
In-Reply-To: <20130116123147.A35594@eng-mail01.juniper.net>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
X-Scanned-By: MIMEDefang 2.58 on oldperseverance.encs.concordia.ca at 2013/01/16 18:02:18 EST
Subject: Re: [MBONED] Adoption of draft-mcbride-armd-mcast-overview
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Jan 2013 23:02:21 -0000

I am in favor of adopting this draft as a WG document in mboned.

  Bill

On 1/16/2013 3:36 PM, Leonard Giuliano wrote:
> 
> We'd like to see if there is consensus from the WG to adopt
> draft-mcbride-armd-mcast-overview as an MBoned document.  In Atlanta,
> this draft was presented and 5 were in favor, none opposed adopting. 
> Please speak up if you support or oppose adopting this draft as a WG
> document.
> 
> The latest version can be found here:
> https://datatracker.ietf.org/doc/draft-mcbride-armd-mcast-overview/
> 
> _______________________________________________
> MBONED mailing list
> MBONED@ietf.org
> https://www.ietf.org/mailman/listinfo/mboned

-- 
Dr. J.W. Atwood, Eng.             tel:   +1 (514) 848-2424 x3046
Distinguished Professor Emeritus  fax:   +1 (514) 848-2830
Department of Computer Science
   and Software Engineering
Concordia University EV 3.185     email:william.atwood@concordia.ca
1455 de Maisonneuve Blvd. West    http://users.encs.concordia.ca/~bill
Montreal, Quebec Canada H3G 1M8

From Tina.Tsou.Zouting@huawei.com  Wed Jan 16 15:07:29 2013
Return-Path: <Tina.Tsou.Zouting@huawei.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id BC3CA11E80ED for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 15:07:29 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.598
X-Spam-Level: 
X-Spam-Status: No, score=-6.598 tagged_above=-999 required=5 tests=[AWL=0.001,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id OBLd6XDh-HDT for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 15:07:28 -0800 (PST)
Received: from lhrrgout.huawei.com (lhrrgout.huawei.com [194.213.3.17]) by ietfa.amsl.com (Postfix) with ESMTP id 0ABD611E80D9 for <mboned@ietf.org>; Wed, 16 Jan 2013 15:07:27 -0800 (PST)
Received: from 172.18.7.190 (EHLO lhreml203-edg.china.huawei.com) ([172.18.7.190]) by lhrrg01-dlp.huawei.com (MOS 4.3.5-GA FastPath queued) with ESMTP id AOV73788; Wed, 16 Jan 2013 23:07:17 +0000 (GMT)
Received: from LHREML406-HUB.china.huawei.com (10.201.5.243) by lhreml203-edg.huawei.com (172.18.7.221) with Microsoft SMTP Server (TLS) id 14.1.323.3; Wed, 16 Jan 2013 23:07:04 +0000
Received: from DFWEML405-HUB.china.huawei.com (10.193.5.102) by lhreml406-hub.china.huawei.com (10.201.5.243) with Microsoft SMTP Server (TLS) id 14.1.323.3; Wed, 16 Jan 2013 23:07:16 +0000
Received: from DFWEML513-MBS.china.huawei.com ([169.254.4.4]) by dfweml405-hub.china.huawei.com ([10.193.5.102]) with mapi id 14.01.0323.003; Wed, 16 Jan 2013 15:07:12 -0800
From: Tina TSOU <Tina.Tsou.Zouting@huawei.com>
To: John William Atwood <william.atwood@concordia.ca>
Thread-Topic: [MBONED] Adoption of draft-mcbride-armd-mcast-overview
Thread-Index: AQHN9ClfnBEebxR1vUC+iLO38c3i85hNGNoA//97QY8=
Date: Wed, 16 Jan 2013 23:07:11 +0000
Message-ID: <F882C240-DEC4-463C-B5DD-E453BA2637FF@huawei.com>
References: <20130116123147.A35594@eng-mail01.juniper.net>, <50F7317A.3070708@concordia.ca>
In-Reply-To: <50F7317A.3070708@concordia.ca>
Accept-Language: en-US, zh-CN
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-CFilter-Loop: Reflected
Cc: "mboned@ietf.org" <mboned@ietf.org>
Subject: Re: [MBONED] Adoption of draft-mcbride-armd-mcast-overview
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Jan 2013 23:07:30 -0000

Dear all,
Support the adoption.

Thank you,
Tina

On Jan 16, 2013, at 3:02 PM, "John William Atwood" <william.atwood@concordi=
a.ca> wrote:

> I am in favor of adopting this draft as a WG document in mboned.
>=20
>  Bill
>=20
> On 1/16/2013 3:36 PM, Leonard Giuliano wrote:
>>=20
>> We'd like to see if there is consensus from the WG to adopt
>> draft-mcbride-armd-mcast-overview as an MBoned document.  In Atlanta,
>> this draft was presented and 5 were in favor, none opposed adopting.=20
>> Please speak up if you support or oppose adopting this draft as a WG
>> document.
>>=20
>> The latest version can be found here:
>> https://datatracker.ietf.org/doc/draft-mcbride-armd-mcast-overview/
>>=20
>> _______________________________________________
>> MBONED mailing list
>> MBONED@ietf.org
>> https://www.ietf.org/mailman/listinfo/mboned
>=20
> --=20
> Dr. J.W. Atwood, Eng.             tel:   +1 (514) 848-2424 x3046
> Distinguished Professor Emeritus  fax:   +1 (514) 848-2830
> Department of Computer Science
>   and Software Engineering
> Concordia University EV 3.185     email:william.atwood@concordia.ca
> 1455 de Maisonneuve Blvd. West    http://users.encs.concordia.ca/~bill
> Montreal, Quebec Canada H3G 1M8
> _______________________________________________
> MBONED mailing list
> MBONED@ietf.org
> https://www.ietf.org/mailman/listinfo/mboned

From albert.e.manfredi@boeing.com  Wed Jan 16 15:58:33 2013
Return-Path: <albert.e.manfredi@boeing.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 8D4BD21F8746 for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 15:58:33 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id rpbrxIiOi9KD for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 15:58:33 -0800 (PST)
Received: from slb-mbsout-01.boeing.com (slb-mbsout-01.boeing.com [130.76.64.128]) by ietfa.amsl.com (Postfix) with ESMTP id F137721F8735 for <mboned@ietf.org>; Wed, 16 Jan 2013 15:58:32 -0800 (PST)
Received: from slb-mbsout-01.boeing.com (localhost.localdomain [127.0.0.1]) by slb-mbsout-01.boeing.com (8.14.4/8.14.4/DOWNSTREAM_MBSOUT) with ESMTP id r0GNwWQa009491 for <mboned@ietf.org>; Wed, 16 Jan 2013 15:58:32 -0800
Received: from XCH-MWHT-02.mw.nos.boeing.com (xch-mwht-02.mw.nos.boeing.com [134.57.113.20]) by slb-mbsout-01.boeing.com (8.14.4/8.14.4/UPSTREAM_MBSOUT) with ESMTP id r0GNwVsm009481 (version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=OK); Wed, 16 Jan 2013 15:58:31 -0800
Received: from XCH-MW-08V.mw.nos.boeing.com ([134.57.118.180]) by XCH-MWHT-02.mw.nos.boeing.com ([134.57.113.20]) with mapi; Wed, 16 Jan 2013 17:58:30 -0600
From: "Manfredi, Albert E" <albert.e.manfredi@boeing.com>
To: Leonard Giuliano <lenny@juniper.net>, MBONED WG <mboned@ietf.org>
Date: Wed, 16 Jan 2013 17:58:29 -0600
Thread-Topic: [MBONED] Adoption of draft-mcbride-armd-mcast-overview
Thread-Index: Ac30KUvfmhRn1VV8SFmvKTXGf60pgQAG9goA
Message-ID: <B0147C3DD45E42478038FC347CCB65FE02C136A65E@XCH-MW-08V.mw.nos.boeing.com>
References: <20130116123147.A35594@eng-mail01.juniper.net>
In-Reply-To: <20130116123147.A35594@eng-mail01.juniper.net>
Accept-Language: en-US
Content-Language: en-US
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: en-US
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-TM-AS-MML: No
Subject: Re: [MBONED] Adoption of draft-mcbride-armd-mcast-overview
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Wed, 16 Jan 2013 23:58:33 -0000

Looks good to me too.

I'd actually prefer to see a periodic update of the current state of IP mul=
ticasting, just to know what everyone else is up to, what solutions are fal=
ling by the wayside, things like that.

Bert

-----Original Message-----
From: mboned-bounces@ietf.org [mailto:mboned-bounces@ietf.org] On Behalf Of=
 Leonard Giuliano
Sent: Wednesday, January 16, 2013 3:36 PM
To: MBONED WG
Subject: [MBONED] Adoption of draft-mcbride-armd-mcast-overview


We'd like to see if there is consensus from the WG to adopt=20
draft-mcbride-armd-mcast-overview as an MBoned document.  In Atlanta, this=
=20
draft was presented and 5 were in favor, none opposed adopting.  Please=20
speak up if you support or oppose adopting this draft as a WG document.

The latest version can be found here:=20
https://datatracker.ietf.org/doc/draft-mcbride-armd-mcast-overview/

_______________________________________________
MBONED mailing list
MBONED@ietf.org
https://www.ietf.org/mailman/listinfo/mboned

From christian.jacquenet@orange.com  Wed Jan 16 23:55:23 2013
Return-Path: <christian.jacquenet@orange.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AF4FF21F8A0F for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 23:55:23 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.248
X-Spam-Level: 
X-Spam-Status: No, score=-2.248 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, HELO_EQ_FR=0.35, UNPARSEABLE_RELAY=0.001]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id Bjd-jDPhTNh3 for <mboned@ietfa.amsl.com>; Wed, 16 Jan 2013 23:55:23 -0800 (PST)
Received: from relais-inet.francetelecom.com (relais-ias92.francetelecom.com [193.251.215.92]) by ietfa.amsl.com (Postfix) with ESMTP id D9F7F21F89DC for <mboned@ietf.org>; Wed, 16 Jan 2013 23:55:22 -0800 (PST)
Received: from omfedm08.si.francetelecom.fr (unknown [xx.xx.xx.4]) by omfedm10.si.francetelecom.fr (ESMTP service) with ESMTP id 982952649BE; Thu, 17 Jan 2013 08:55:21 +0100 (CET)
Received: from PUEXCH41.nanterre.francetelecom.fr (unknown [10.101.44.30]) by omfedm08.si.francetelecom.fr (ESMTP service) with ESMTP id 6DEAC238080; Thu, 17 Jan 2013 08:55:21 +0100 (CET)
Received: from PUEXCB1C.nanterre.francetelecom.fr ([10.101.44.9]) by PUEXCH41.nanterre.francetelecom.fr ([10.101.44.30]) with mapi; Thu, 17 Jan 2013 08:55:21 +0100
From: <christian.jacquenet@orange.com>
To: Leonard Giuliano <lenny@juniper.net>, MBONED WG <mboned@ietf.org>
Date: Thu, 17 Jan 2013 08:55:20 +0100
Thread-Topic: [MBONED] Adoption of draft-mcbride-armd-mcast-overview
Thread-Index: Ac30KUnhdLnK7Cf8R7eCmxNMutRalQAXqOSQ
Message-ID: <19378_1358409321_50F7AE69_19378_539_1_983A1D8DA0DA5F4EB747BF34CBEE5CD15A68BEE2D6@PUEXCB1C.nanterre.francetelecom.fr>
References: <20130116123147.A35594@eng-mail01.juniper.net>
In-Reply-To: <20130116123147.A35594@eng-mail01.juniper.net>
Accept-Language: fr-FR
Content-Language: fr-FR
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
acceptlanguage: fr-FR
Content-Type: text/plain; charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-PMX-Version: 5.6.1.2065439, Antispam-Engine: 2.7.2.376379, Antispam-Data: 2012.10.24.110314
Subject: Re: [MBONED] Adoption of draft-mcbride-armd-mcast-overview
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 17 Jan 2013 07:55:23 -0000

Hello WG,

I support the adoption of this draft as a mboned WG document.

Cheers,

Christian.=20

-----Message d'origine-----
De : mboned-bounces@ietf.org [mailto:mboned-bounces@ietf.org] De la part de=
 Leonard Giuliano
Envoy=E9 : mercredi 16 janvier 2013 21:36
=C0 : MBONED WG
Objet : [MBONED] Adoption of draft-mcbride-armd-mcast-overview


We'd like to see if there is consensus from the WG to adopt draft-mcbride-a=
rmd-mcast-overview as an MBoned document.  In Atlanta, this draft was prese=
nted and 5 were in favor, none opposed adopting.  Please speak up if you su=
pport or oppose adopting this draft as a WG document.

The latest version can be found here:=20
https://datatracker.ietf.org/doc/draft-mcbride-armd-mcast-overview/

_______________________________________________
MBONED mailing list
MBONED@ietf.org
https://www.ietf.org/mailman/listinfo/mboned

___________________________________________________________________________=
______________________________________________

Ce message et ses pieces jointes peuvent contenir des informations confiden=
tielles ou privilegiees et ne doivent donc
pas etre diffuses, exploites ou copies sans autorisation. Si vous avez recu=
 ce message par erreur, veuillez le signaler
a l'expediteur et le detruire ainsi que les pieces jointes. Les messages el=
ectroniques etant susceptibles d'alteration,
France Telecom - Orange decline toute responsabilite si ce message a ete al=
tere, deforme ou falsifie. Merci.

This message and its attachments may contain confidential or privileged inf=
ormation that may be protected by law;
they should not be distributed, used or copied without authorisation.
If you have received this email in error, please notify the sender and dele=
te this message and its attachments.
As emails may be altered, France Telecom - Orange is not liable for message=
s that have been modified, changed or falsified.
Thank you.


From wmaton@ottix.net  Thu Jan 17 02:39:21 2013
Return-Path: <wmaton@ottix.net>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id 5964921F88EE for <mboned@ietfa.amsl.com>; Thu, 17 Jan 2013 02:39:21 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -2.599
X-Spam-Level: 
X-Spam-Status: No, score=-2.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id kaM4AObk0Mwv for <mboned@ietfa.amsl.com>; Thu, 17 Jan 2013 02:39:21 -0800 (PST)
Received: from iskra.ottix.net (iskra.ottix.net [IPv6:2001:410:90ff::2]) by ietfa.amsl.com (Postfix) with ESMTP id C39D221F88E3 for <mboned@ietf.org>; Thu, 17 Jan 2013 02:39:20 -0800 (PST)
Received: from iskra.ottix.net (localhost [127.0.0.1]) by iskra.ottix.net (8.14.5/8.14.5) with ESMTP id r0HAdIS0004816 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NO) for <mboned@ietf.org>; Thu, 17 Jan 2013 05:39:18 -0500
Received: from localhost (wmaton@localhost) by iskra.ottix.net (8.14.5/8.14.3/Submit) with ESMTP id r0HAdHM1004813 for <mboned@ietf.org>; Thu, 17 Jan 2013 05:39:17 -0500
Date: Thu, 17 Jan 2013 05:39:17 -0500 (EST)
From: "William F. M. Sotomayor" <wmaton@ottix.net>
To: MBONED WG <mboned@ietf.org>
In-Reply-To: <20130116123147.A35594@eng-mail01.juniper.net>
Message-ID: <Pine.LNX.4.64.1301170538400.26347@iskra.ottix.net>
References: <20130116123147.A35594@eng-mail01.juniper.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII; format=flowed
Subject: Re: [MBONED] Adoption of draft-mcbride-armd-mcast-overview
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 17 Jan 2013 10:39:21 -0000

I would support the adoption of this document.  It promises to be an 
interesting study.

On Wed, 16 Jan 2013, Leonard Giuliano wrote:

>
> We'd like to see if there is consensus from the WG to adopt 
> draft-mcbride-armd-mcast-overview as an MBoned document.  In Atlanta, this 
> draft was presented and 5 were in favor, none opposed adopting.  Please speak 
> up if you support or oppose adopting this draft as a WG document.
>
> The latest version can be found here: 
> https://datatracker.ietf.org/doc/draft-mcbride-armd-mcast-overview/
>
> _______________________________________________
> MBONED mailing list
> MBONED@ietf.org
> https://www.ietf.org/mailman/listinfo/mboned
>

wfms

From diego@tid.es  Thu Jan 17 02:57:49 2013
Return-Path: <diego@tid.es>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id D5F9421F8745 for <mboned@ietfa.amsl.com>; Thu, 17 Jan 2013 02:57:49 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -6.413
X-Spam-Level: 
X-Spam-Status: No, score=-6.413 tagged_above=-999 required=5 tests=[AWL=0.186,  BAYES_00=-2.599, RCVD_IN_DNSWL_MED=-4]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id BIlT84bvSdUK for <mboned@ietfa.amsl.com>; Thu, 17 Jan 2013 02:57:49 -0800 (PST)
Received: from tidos.tid.es (tidos.tid.es [195.235.93.44]) by ietfa.amsl.com (Postfix) with ESMTP id B62B321F869E for <mboned@ietf.org>; Thu, 17 Jan 2013 02:57:48 -0800 (PST)
Received: from sbrightmailg01.hi.inet (sbrightmailg01.hi.inet [10.95.64.104]) by tid.hi.inet (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTP id <0MGR00MTYNSADB@tid.hi.inet> for mboned@ietf.org; Thu, 17 Jan 2013 11:57:46 +0100 (MET)
Received: from tid (tid.hi.inet [10.95.64.10])	by sbrightmailg01.hi.inet (Symantec Messaging Gateway) with SMTP id 14.DF.03184.A29D7F05; Thu, 17 Jan 2013 11:57:46 +0100 (CET)
Received: from correo.tid.es (mailhost.hi.inet [10.95.64.100]) by tid.hi.inet (iPlanet Messaging Server 5.2 HotFix 2.14 (built Aug 8 2006)) with ESMTPS id <0MGR00MTVNSADB@tid.hi.inet> for mboned@ietf.org; Thu, 17 Jan 2013 11:57:46 +0100 (MET)
Received: from EX10-MB2-MAD.hi.inet ([169.254.2.165]) by ex10-htcas4-mad.hi.inet ([::1]) with mapi id 14.02.0328.009; Thu, 17 Jan 2013 11:57:46 +0100
Date: Thu, 17 Jan 2013 10:57:45 +0000
From: "Diego R. Lopez" <diego@tid.es>
In-reply-to: <Pine.LNX.4.64.1301170538400.26347@iskra.ottix.net>
X-Originating-IP: [10.95.64.115]
To: MBONED WG <mboned@ietf.org>
Message-id: <E6D8B95470ED0845B3376F61DCAB1A0468CFB4C0@EX10-MB2-MAD.hi.inet>
Content-id: <A972DA7B1EC7574C9EE34C3C845C60F0@hi.inet>
MIME-version: 1.0
Content-type: text/plain; charset=utf-8
Content-language: en-US
Content-transfer-encoding: base64
Accept-Language: en-US
Thread-topic: [MBONED] Adoption of draft-mcbride-armd-mcast-overview
Thread-index: AQHN9ClL6f6lseMkPEK3mm1Ig/c5V5hNRLaAgAAFJ4A=
X-AuditID: 0a5f4068-b7fc06d000000c70-67-50f7d92ac7b4
X-MS-Has-Attach: 
X-MS-TNEF-Correlator: 
X-Brightmail-Tracker: H4sIAAAAAAAAA+NgFprPKsWRmVeSWpSXmKPExsXCFe/Apat183uAwfYOU4spLacZHRg9liz5 yRTAGMVlk5Kak1mWWqRvl8CVMfvaLKaCHRwVn9Z2MjcwzuHoYuTkkBAwkfi4ejcThC0mceHe erYuRi4OIYENjBLvbx1ihnB+Mkr8fbOfCcKZySjRtHgykMPBwSKgKjHzQyZINxuQ+aj5NzuI LSzgLPHpyhUWEJtTwEFi16V7rBAbFCT+nHsMFhcRkJdo7e4Fq+cV8JaYfeYU2BXMAmYSa+cs ZYOIC0r8mHyPBWQVs4C6xJQpuRAl4hLNrTdZIGxFiWmLGhhBbEagB76fWsMEMd5F4t2hP1Cr rCQWvV3LBnGCgMSSPeeZIWxRiZeP/4GdJiRQILFm+xG2CYzis5BcMQvJFbMQrpiF5IpZSK5Y wMi6ilGsOKkoMz2jJDcxMyfdwFAvI1MvMy+1ZBMjJLYydjAu36lyiFGAg1GJh/fA2c8BQqyJ ZcWVuYcYJTmYlER5w69/DxDiS8pPqcxILM6ILyrNSS0+xCjBwawkwqtwDSjHm5JYWZValA+T kuHgUJLgDbwBlBIsSk1PrUjLzAEmEJg0EwcnSDsPULsRSA1vcUFibnFmOkT+FKOklDjvG5Cd AiCJjNI8uN5XjOJARwrzCoK08QBTHVzXK6CBTEADRS6CDSxJREhJNTBW7m/qXnVt3p+opVtm 5S6s9t+/6YhL8K6JujpaEWaOq56X/FmuM2FGup6AldHt9QsPfHfu+7WKs+z7ga78wnVrLbW7 FbZUsU1genI42cmCXWm5mbPIwxuay5u89FSe9Nvpf4+Vz5uwz0agQbsx/2eJpX3d2pepm/tz tzRqvSwsrrtSxhNiqKLEUpyRaKjFXFScCAA4ZHeDMgMAAA==
References: <20130116123147.A35594@eng-mail01.juniper.net> <Pine.LNX.4.64.1301170538400.26347@iskra.ottix.net>
Subject: Re: [MBONED] Adoption of draft-mcbride-armd-mcast-overview
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 17 Jan 2013 10:57:50 -0000
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==

From Martin.Stiemerling@neclab.eu  Thu Jan 17 07:17:25 2013
Return-Path: <Martin.Stiemerling@neclab.eu>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id C7D4221F8588; Thu, 17 Jan 2013 07:17:25 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.566
X-Spam-Level: 
X-Spam-Status: No, score=-103.566 tagged_above=-999 required=5 tests=[AWL=0.033, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id VrB+4acRut-8; Thu, 17 Jan 2013 07:17:24 -0800 (PST)
Received: from mailer1.neclab.eu (mailer1.neclab.eu [195.37.70.40]) by ietfa.amsl.com (Postfix) with ESMTP id BAC1C21F8581; Thu, 17 Jan 2013 07:17:23 -0800 (PST)
Received: from localhost (localhost [127.0.0.1]) by mailer1.neclab.eu (Postfix) with ESMTP id 28E82102DE1; Thu, 17 Jan 2013 16:17:23 +0100 (CET)
X-Virus-Scanned: Amavisd on Debian GNU/Linux (netlab.nec.de)
Received: from mailer1.neclab.eu ([127.0.0.1]) by localhost (atlas-a.office.hd [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id KsQEQYsbeQFJ; Thu, 17 Jan 2013 16:17:23 +0100 (CET)
Received: from ENCELADUS.office.hd (enceladus.office.hd [192.168.24.52]) by mailer1.neclab.eu (Postfix) with ESMTP id 06556102DDF; Thu, 17 Jan 2013 16:16:53 +0100 (CET)
Received: from [10.7.0.105] (10.7.0.105) by skoll.office.hd (192.168.125.11) with Microsoft SMTP Server (TLS) id 14.1.323.3; Thu, 17 Jan 2013 16:15:49 +0100
Message-ID: <50F815C8.102@neclab.eu>
Date: Thu, 17 Jan 2013 16:16:24 +0100
From: Martin Stiemerling <martin.stiemerling@neclab.eu>
User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:17.0) Gecko/20130106 Thunderbird/17.0.2
MIME-Version: 1.0
To: <gjshep@gmail.com>
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se> <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com> <50ED7B8D.3010904@neclab.eu> <CABFReBoCTN-KP=LHrVy+Bg4y6g4aezUH-2rWzvZ0GVmAtejERA@mail.gmail.com> <50F66AA5.6050705@ericsson.com> <CABFReBpiT3=LJbS8_Mh3kigAqsMUAZP9ECB9LaiO99gL9Tj-xQ@mail.gmail.com>
In-Reply-To: <CABFReBpiT3=LJbS8_Mh3kigAqsMUAZP9ECB9LaiO99gL9Tj-xQ@mail.gmail.com>
Content-Type: text/plain; charset="ISO-8859-1"; format=flowed
Content-Transfer-Encoding: 8bit
X-Originating-IP: [10.7.0.105]
Cc: "mboned@ietf.org" <mboned@ietf.org>, Greg Bumgardner <gbumgard@gmail.com>, "iesg@ietf.org" <iesg@ietf.org>, "tsv-dir@ietf.org" <tsv-dir@ietf.org>
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 17 Jan 2013 15:17:25 -0000

On 01/16/2013 03:04 PM, Greg Shepherd wrote:
> Why would you require such restriction? If your concern is about
> support for RTCP we already have RFC 5760 to address this. It's an
> application layer problem.

RFC 5760 has some nice security considerations, e.g., possible attacks 
on the Receiver-to-Distribution-Source Communication. It might be also 
good to document how AMT in conjunction with RFC 5760 is handled in 
actual deployments. E.g., AMT can introduce new issues, as the multicast 
source is located in a firewall protected network that blocks RTCP 
traffic coming from receivers somewhere else in the Internet.

AMT should discuss and document the changes in the deployment model as 
described in RFC 5760.

   Martin

>
> Greg
>
> On Wed, Jan 16, 2013 at 12:53 AM, Magnus Westerlund
> <magnus.westerlund@ericsson.com> wrote:
>> Hi,
>>
>> I think the major question here, are anyone really objecting to
>> restricting the applicability of AMT to managed environments, i.e. where
>> the full path between relay and client are in domains that are
>> explicitly aware that they will have AMT traffic and approve of it. The
>> comments appears to indicate that this is where it is deployed. If one
>> makes such an applicability statement most of the significant comments
>> can be more easily resolved.
>>
>> Cheers
>>
>> Magnus
>>
>> On 2013-01-10 17:35, Greg Shepherd wrote:
>>> On Wed, Jan 9, 2013 at 6:15 AM, Martin Stiemerling
>>> <martin.stiemerling@neclab.eu> wrote:
>>>> Hi Greg,
>>>>
>>>> I'm joining the discussion as the DISCUSS holder for most of the issues
>>>> below.
>>>>
>>>>
>>>> On 01/09/2013 03:26 AM, Greg Bumgardner wrote:
>>>>>
>>>>> Magnus,
>>>>>
>>>>> I would like to thank you for providing valuable feedback regarding the
>>>>> AMT spec.
>>>>>
>>>>> I would first like to state that many of your concerns and criticisms
>>>>> regarding the protocol were raised during MBONED group discussions. Many
>>>>> alternative solutions were identified but ultimately rejected by the
>>>>> group, primarily due to the reluctance of group members to make
>>>>> significant changes to the protocol after several years of slow
>>>>> development. My comments are inlined below...
>>>>
>>>>
>>>> Can you give some insights why the protocol development took so long?
>>>
>>> Yes, I can, though I'm not sure how relevant the incubation period is
>>> to its adoption. And since I've breaking multicast networks for 15+
>>> years now, asking a question like this may get you more than you were
>>> expecting :)
>>>
>>> The need for multicast transition services is as old as multicast
>>> itself. Deering's first work - DVMRP - had to assume there was no
>>> 'native' path and so incorporated encapsulation as an integral part of
>>> the protocol. I believe DVMRP may be the first use of overlay
>>> encapsulation in the IETF, but I welcome you to dig for your self and
>>> back with me. It's the earliest I've been able to find.
>>>
>>> Once we moved to PIM, encapsulation for receiver delivery was thrown
>>> out and we all pushed to "go native". The reality was - and still is -
>>> that those most wanting end-to-end native multicast (content owners)
>>> had no way to motivate network operators to enable multicast services.
>>> Deployment experience with PIM revealed to many the need for
>>> encapsulation overlay mechanism, and from this several drafts were
>>> introduced.
>>>
>>> In 2003 we held a BOF at IETF57 with the intention of agreeing on a
>>> list of requirements and consolidating the efforts into a single
>>> draft. We selected the AMT draft as being the closest fit, though
>>> significant changes still need to be addressed. You see the list of
>>> drafts and read the BOF minutes here:
>>>
>>> http://www.ietf.org/meeting/57/session/mlm.txt
>>>
>>> Authors came and went.. Progress was slow, until ATT picked up the
>>> torch, demonstrated running code, and outlined their deployment plans.
>>> Greg Bumgardner also wrote a java Gateway implementation and took over
>>> as primary author. Finally we had solid industry support and a
>>> qualified and motivated set of contributors.
>>>
>>> Last year we finally reached WG consensus, along with at least three
>>> shipping implementations of both the Relay and the Gateway , and at
>>> least one active and growing commercial deployment of a
>>> IPMulticast/AMT service. I doubt many drafts in the editors' queue can
>>> claim that much support. :)
>>>
>>>> Anyhow, I can understand the reluctance of changing parts of a protocol
>>>> after such a long time. But there are a lot of rough edges in the protocol
>>>> that need attention before moving it forward on the Standards Track.
>>>
>>> Time is irrelevant to any reluctance to change - commercial
>>> implementations and deployments are, however quite relevant.
>>>
>>> Cheers,
>>> Greg
>>>
>>>>>
>>>>> Thanks,
>>>>>
>>>>> -g.b.
>>>>>
>>>>>
>>>>> On Wed, Dec 26, 2012 at 5:01 AM, Magnus Westerlund
>>>>> <magnus.westerlund@ericsson.com <mailto:magnus.westerlund@ericsson.com>>
>>>>>
>>>>> wrote:
>>>>>
>>>>>      Hi,
>>>>>
>>>>>      Sorry for being late with this IETF last call comments. I will
>>>>>      partly blame the ADs requesting this Transport Directorate review a
>>>>>      bit late, the other part is all mine and the holidays. Anyway, I do
>>>>>      hope you will consider these issues and comments as I believe I
>>>>>      found some serious ones in addition to a number of clarifications
>>>>>      that should be made.
>>>>>
>>>>>
>>>>>      Significant Issues:
>>>>>
>>>>>      1. Congestion Control
>>>>>      This is clearly a tunnel establishment protocol of something that is
>>>>>      IP traffic. Thus normally the responsibility for congestion control
>>>>>      is with the tunnelled traffic. However, I would like argue that this
>>>>>      does not apply in this case due to the nature of the tunnelled
>>>>>      traffic, i.e. multicast traffic and secondary due to limitations in
>>>>>      the tunnel protocol.
>>>>>
>>>>>      Lets start with the second part. This protocol claims to support ASM
>>>>>      still don't provide a upstream delivery mechanism, i.e. an ASM
>>>>>      receiver is not capable of sending as it should. This prevents
>>>>>      several existing mechanism for congestion control that exist in
>>>>>      protocols supporting multicast. The first is using RTCP for
>>>>>      congestion control in ASM [RFC3550], the second isTCP-Friendly
>>>>>      Multicast Congestion Control [RFC4654] that can be used in the RMT
>>>>>      suite of protocols, and I know has been implemented in some NORM
>>>>>      implementation. Thus onlystrictly receiver based mechanisms, such as
>>>>>
>>>>>      Wave and Equation Based Rate Control [RFC3738] are available in this
>>>>>      context.
>>>>>
>>>>>
>>>>> You are correct, the AMT protocol provides no method for injecting
>>>>> multicast traffic into a multicast enabled network, and thus does not
>>>>> provide the functionality required for something like ASM RTP/RTCP
>>>>> teleconferencing. The protocol is primarily intended for use in
>>>>> receiving SSM traffic. Because the AMT protocol uses IGMP/MLD messages
>>>>> to establish reception/subscription state, the protocol MAY be used to
>>>>> subscribe to ASM broadcasts, but cannot generate them.
>>>>>
>>>>> I should note that earlier versions of the AMT spec described a method
>>>>> for broadcasting SSM via a gateway/relay tunnel, but this capability was
>>>>> removed due to feasibility considerations.
>>>>
>>>>
>>>> This means that the draft should state that it does not support ASM, but
>>>> only a subset of it. Further, it cuts away the feedback over RTCP and RFC
>>>> 4654.
>>>>
>>>>>
>>>>>
>>>>>      Secondly, many multicast usages are in fact deployed without any
>>>>>      congestion control. This based on that the deploying entity controls
>>>>>      the scope and authorization for requesting multicast delivery.
>>>>>      However, does restrictions does not apply to AMT delivery of
>>>>>      multicast. If the gateway can reach using unicast the relay it can
>>>>>      be delivered the multicast group from the domain the relay is
>>>>>      attached to. Thus, this protocol changes the deployment restrictions
>>>>>      of multicast which many non-congestion controlled delivery is based
>>>>>      on. Instead the non-congestion controlled trafficcan now sent over
>>>>>
>>>>>      an IP/UDP tunnel over Internet where neither relay nor gateway may
>>>>>      have any knowledge about the path the traffic may take.
>>>>>
>>>>>
>>>>> The AMT protocol does not attempt to address the issue of congestion
>>>>> control. The group decided to treat the detection and remediation of
>>>>> congestion (of encapsulated multicast packets) as an application-layer
>>>>> responsibility. While we might have attempted to provide some level of
>>>>> QoS control on tunnel packets by copying ToS/DS bits from inner to outer
>>>>> headers, this would only occur on a packet-by-packet basis on not on the
>>>>> tunnel as a whole.
>>>>
>>>>
>>>> What's the group's view on how the application layer is handling congestion
>>>> control in this case when using AMT?
>>>>
>>>>
>>>>>
>>>>>
>>>>>      Based on this I would like to see two changes to this protocol
>>>>>      specification. First a section discussing the issue of congestion
>>>>>      control.
>>>>>
>>>>>
>>>>> I'm not sure what, if anything, I might add to address this. I think the
>>>>> spec makes it pretty clear that the tunneling is provided by best-effort
>>>>> delivery of UDP packets (perhaps restricted by relay/router
>>>>> configuration). Perhaps you could suggest some text.
>>>>
>>>>
>>>> There are multiple issues here:
>>>> - there is no feedback loop between relay and gateway so that any
>>>> information about the tunneled multicast traffic can be sent back from the
>>>> gateway to the relay. This feedback loop can be used for congestion control,
>>>> but also to figure out if the non-zero IPv6/UDP traffic gets black holed
>>>> somewhere.
>>>> - the tunneled traffic can exceed the capacity of the network path between
>>>> relay and gateway. This needs to be monitored and acted upon any congestion.
>>>> - since a single relay can feed many gateways, even tunneled multicast flows
>>>> from the same relay can compete for network path capacity. I.e., if the
>>>> tunneled traffic shares a network path. In the worst case, an added
>>>> multicast flow can lead to congestion for all other tunneled multicast
>>>> flows.
>>>> - any tunneled flow can impact other traffic from other hosts in the
>>>> Internet.
>>>>
>>>>>
>>>>>      Secondly, I think this protocol should have an applicability
>>>>>      statement limiting its deployment to restricted environments where
>>>>>      the relay and gateway deployers can provide certain resource
>>>>>      provisions between the entities to avoid the multicast traffic
>>>>>      affecting other traffic sharing the same bottlenecksin ways not
>>>>>
>>>>>      allowed by the network provider.
>>>>>
>>>>>
>>>>> While I understand your concern, this would essentially negate the
>>>>> entire reason for the protocol, which is to provide multicast transit to
>>>>> non-multicast connected receivers. However, the AMT spec does not
>>>>> preclude the implementation/deployment of AMT-enabled gateway devices
>>>>> for which AMT support may be enabled or disabled (or blocked via a
>>>>> firewall).
>>>>
>>>>
>>>> The point is that it is not safe to operate in the public Internet with the
>>>> current protocol. Either with fix the issues in the protocol or we limit the
>>>> applicability.
>>>>
>>>>
>>>>>
>>>>>
>>>>>
>>>>>      2. Security
>>>>>
>>>>>      This protocols is frank with it having limited security features and
>>>>>      says this is similar to the IGMP and MLD protocols being used.
>>>>>      However, I think this is a failure to propoerly consider the threat
>>>>>      model. If one uses AMT over general Internet it will run in a
>>>>>      network where the one deploying the multicast and the relays no
>>>>>      longer control requirements on source address verification or
>>>>>      possibilities for traffic separation as they can do within the
>>>>>      domains where multicast currently are deployed. The security
>>>>>      vulnerabilities in IGMP and MLD are much more contained and
>>>>>      controllable in a LAN environment where one has chosen to deploy
>>>>>      multicast compared to an Relay exposing this to the whole Internet.
>>>>>      Once more I think there is only two choices here.
>>>>>
>>>>>      A) Beef up the security to general Internet threat model, i.e. at a
>>>>>      minimal provide a real model for gateway authentication using
>>>>>      identities, not only return routability based verifications.
>>>>>
>>>>>      B) Limit the applicability of AMT to managed environments and make
>>>>>      it clear that the relay will need to limit which gateways are
>>>>>      allowed to access the relay based on addressing.
>>>>>
>>>>>      Based on the first significant issue with congestion control I
>>>>>      expect that there is little meaning to do A) unless also one is
>>>>>      willing to beef up AMT to provide congestion control. Which I think
>>>>>      is not according to the design wishes for the protocol designers.
>>>>>
>>>>>
>>>>> The security topic was visited and revisited numerous times over the
>>>>> last couple of years. The group made the decision to limit the protocol
>>>>> to simple source address verification and leave it at that. An emphasis
>>>>> was placed on protecting relays against simple DoS attacks but not
>>>>> gateways.
>>>>
>>>>
>>>> I can see the point, but I assume that most people are just looking to
>>>> closed environments, e.g., multicast in access networks, such as xDSL or
>>>> 3GPP. However, AMT assumes to run across the public Internet and threats are
>>>> extremely different there. Magnus made a good proposal that is relatively
>>>> easy to implement.
>>>>
>>>>>
>>>>>
>>>>>      3. Use of Zero Checksum
>>>>>
>>>>>      The AMT specification enables the use of Zero UDP checksum with
>>>>>      IPv6, i.e.
>>>>>      draft-ietf-6man-udpchecksums-06
>>>>>      <http://datatracker.ietf.org/doc/draft-ietf-6man-udpchecksums/> and
>>>>>      draft-ietf-6man-udpzero-08
>>>>>      <http://datatracker.ietf.org/doc/draft-ietf-6man-udpzero/>. Nothing
>>>>>
>>>>>      against this in principal. However, I have noticed that AMT fails to
>>>>>      properly address the failure modes of using a zero-checksum. AMT is
>>>>>      a typical example of a protocol that actually need active
>>>>>      verification of each tunnel that zero-checksum functions. This as
>>>>>      AMT is clearly intended to be deployed with its Gateway part in
>>>>>      end-hosts and residential network devices or routers. This means the
>>>>>      tunnel will pass through both firewalls and NATs on its path between
>>>>>      the relay and the gateway. Unless these devices are not upgraded to
>>>>>      support zero-checksum in UDP for IPv6 the traffic may actually
>>>>>      become black holed. The most likely is a simple firewall that has a
>>>>>      rule for IPv6/UDP which doesn't allow zero checksum as it is against
>>>>>      RFC 2460. Thus all the Multicast data packets will disappear on
>>>>>      route in the tunnel. There is no mechanism in the AMT protocol to
>>>>>      detect this and negotiate with the relay so that it will not use the
>>>>>      zero-checksum for this tunnel.
>>>>>
>>>>>      This must be addressed as I see it. If not the AMT will be so
>>>>>      brittle that it can't be used in a large number of its intended
>>>>>      deployments.
>>>>>
>>>>>
>>>>> You probably have no idea how often this issue was raised and discussed
>>>>> within the group. The consensus was that there was no single solution to
>>>>> the problem. The only workable solution I can suggest is to use an IPv4
>>>>> tunnel to carry the IPv6 packets. The mechanism for deciding whether (or
>>>>> if) to use an IPv4 tunnel is considered out of scope.
>>>>
>>>>
>>>> There is a solution, as mentioned above: feedback loop. If the relay says
>>>> "sending you x packets" and the gateway sees zero, there is an issue. In the
>>>> easiest case, a log file entry is generate saying that the tunneled traffic
>>>> couldn't reach this particular gateway.
>>>>
>>>>
>>>>>
>>>>>
>>>>>      4) MTU issues
>>>>>
>>>>>      This document total fails to discuss the issues of MTU blackholing.
>>>>>      As the IP multicast datagrams as well as the encapsulated IGMP/MLD
>>>>>      messages can with the added tunnel-overhead result in that the sent
>>>>>      packet exceeds the MTU of the path, these packets could be black
>>>>>      holed. This can potentially result in very intermittent transport
>>>>>      behavior for the tunnel. Thus, some discussion of how do handle the
>>>>>      MTU issues in this context should be introduced.
>>>>>
>>>>>
>>>>> Yep. This could be a problem that has no easy solution. Not only might
>>>>> the outer packets be fragmented, but the encapsulated packets might be
>>>>> fragments. Since the protocol obviously has no control over the size of
>>>>> source packets, it cannot really avoid the potential need to fragment
>>>>> the tunnel packets. A gateway can only rely on the underlying UDP/IP
>>>>> stack to provide the necessary reassembly of the tunnel packets. Once
>>>>> the tunnel packet has been reassembled, the gateway, host IP stack or
>>>>> application may still need to reassemble the contents of a sequence of
>>>>> tunnel packets.
>>>>
>>>>
>>>> In the case the WG doesn't have any solution, which I couldn't believe, it
>>>> is worth to document the issues. One operational solution can be to limit
>>>> the MTU in the native multicast network, if this part and the relay are
>>>> under the same administrative control. However, this would require that the
>>>> AMT relay logs any issue with the MTU accross the tunnel.
>>>> Further, Magnus point is about that the tunneled packets can get fragmented,
>>>> less about the inner parts of the tunnel.
>>>>
>>>> Addressing the MTU issue will increase the viability of this a lot.
>>>>
>>>>
>>>>>
>>>>>
>>>>>      I am willing to discuss methods here, but I guess several
>>>>>      alternatives exists and thus which is most appropriate and the level
>>>>>      of AMT support for them varies I would like the protocol designers
>>>>>      to do a first stab at resolving this.
>>>>>
>>>>>      Other Issues
>>>>>      ======================================================================
>>>>>
>>>>>      A) The table of content on page 2 should include more levels of
>>>>>      headings. Most likely down to 4 is needed to make the TOC usable for
>>>>>      finding content in the document.
>>>>>
>>>>>      B) The claimed ASM support
>>>>>      I would like to better understand how one can claim to support ASM
>>>>>      when one has no up stream path to inject the ASM group participants
>>>>>      traffic. When one uses ASM one normally does this for a reason and
>>>>>      needing the possibility to inject packets into the group. These
>>>>>      limitations needs to be clarified.
>>>>>
>>>>>
>>>>> See above. The protocol is intended for multicast reception. It makes no
>>>>> assumptions about application layer use of the multicast traffic. It can
>>>>> only be used to receive ASM IP traffic.
>>>>
>>>>
>>>> Document it, please. Otherwise, the draft is incomplete.
>>>>
>>>> Thank you,
>>>>
>>>>    Martin
>>>> --
>>>> IETF Transport Area Director
>>>>
>>>> martin.stiemerling@neclab.eu
>>>>
>>>> NEC Laboratories Europe - Network Research Division NEC Europe Limited
>>>> Registered Office: NEC House, 1 Victoria Road, London W3 6BL
>>>> Registered in England 283
>>>>
>>>> _______________________________________________
>>>> MBONED mailing list
>>>> MBONED@ietf.org
>>>> https://www.ietf.org/mailman/listinfo/mboned
>>>
>>
>>
>> --
>>
>> Magnus Westerlund
>>
>> ----------------------------------------------------------------------
>> Multimedia Technologies, Ericsson Research EAB/TVM
>> ----------------------------------------------------------------------
>> Ericsson AB                | Phone  +46 10 7148287
>> Färögatan 6                | Mobile +46 73 0949079
>> SE-164 80 Stockholm, Sweden| mailto: magnus.westerlund@ericsson.com
>> ----------------------------------------------------------------------
>>

-- 
martin.stiemerling@neclab.eu

NEC Laboratories Europe - Network Research Division NEC Europe Limited
Registered Office: NEC House, 1 Victoria Road, London W3 6BL
Registered in England 283

From gjshep@gmail.com  Thu Jan 17 07:22:14 2013
Return-Path: <gjshep@gmail.com>
X-Original-To: mboned@ietfa.amsl.com
Delivered-To: mboned@ietfa.amsl.com
Received: from localhost (localhost [127.0.0.1]) by ietfa.amsl.com (Postfix) with ESMTP id AFF1421F8596; Thu, 17 Jan 2013 07:22:12 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.599
X-Spam-Level: 
X-Spam-Status: No, score=-103.599 tagged_above=-999 required=5 tests=[BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.30]) by localhost (ietfa.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id wsQh7+5-MvP7; Thu, 17 Jan 2013 07:22:11 -0800 (PST)
Received: from mail-wi0-f179.google.com (mail-wi0-f179.google.com [209.85.212.179]) by ietfa.amsl.com (Postfix) with ESMTP id 1145F21F8581; Thu, 17 Jan 2013 07:22:09 -0800 (PST)
Received: by mail-wi0-f179.google.com with SMTP id o1so2365253wic.12 for <multiple recipients>; Thu, 17 Jan 2013 07:22:09 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=20120113; h=mime-version:x-received:reply-to:in-reply-to:references:date :message-id:subject:from:to:cc:content-type :content-transfer-encoding; bh=mSEEzQYCR49dzWtP5OtttoDTCEegSmif5QOtudDh+tE=; b=DP5eAtqcECUgAUw6ag7u955jsB1eWMvLdTGeQIVZOEczwUOjX6DOHUHeB24GGGXkmt gr3vGdk8TMVxZi0VfPkfKe+pSwAelY4FOfX3JqIiB2tzskw47EAyw+9flDpqJ0ez3W5Y 0j1qjPHg2Qh1gu2zzHWVva0ZF8nsNryCVm1mhh48pOqshbc4dQgQhSLq+0fJWdoRG/zQ ireoITj2jHGYOrqGQuVe6fbbIP5t771lm4tlQCFyYaTpdb9GmHNpzPP4thDRYVNhDDlU yMahPSxNfaAZVr2vvdMPCRobxHi+f25a+dXMEC5XchMBpzp+HMkM0kJFkEhoS04/u16h PyVQ==
MIME-Version: 1.0
X-Received: by 10.180.75.135 with SMTP id c7mr16712743wiw.10.1358436129137; Thu, 17 Jan 2013 07:22:09 -0800 (PST)
Received: by 10.180.97.66 with HTTP; Thu, 17 Jan 2013 07:22:09 -0800 (PST)
In-Reply-To: <50F815C8.102@neclab.eu>
References: <52E4A8FC978E0241AE652516E24CAF000E5FBC@ESESSMB309.ericsson.se> <CAJkfEFqZH3h8JfzxM=5uFpWzYAjTXxrofc9A1JKHiYs8MEwBOQ@mail.gmail.com> <50ED7B8D.3010904@neclab.eu> <CABFReBoCTN-KP=LHrVy+Bg4y6g4aezUH-2rWzvZ0GVmAtejERA@mail.gmail.com> <50F66AA5.6050705@ericsson.com> <CABFReBpiT3=LJbS8_Mh3kigAqsMUAZP9ECB9LaiO99gL9Tj-xQ@mail.gmail.com> <50F815C8.102@neclab.eu>
Date: Thu, 17 Jan 2013 07:22:09 -0800
Message-ID: <CABFReBpcSnnC46rTeWjRgdaHU_w8=a6tjrnW0kEa-cUt-3YwoQ@mail.gmail.com>
From: Greg Shepherd <gjshep@gmail.com>
To: Martin Stiemerling <martin.stiemerling@neclab.eu>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Cc: "mboned@ietf.org" <mboned@ietf.org>, Greg Bumgardner <gbumgard@gmail.com>, "iesg@ietf.org" <iesg@ietf.org>, "tsv-dir@ietf.org" <tsv-dir@ietf.org>
Subject: Re: [MBONED] Last Call: <draft-ietf-mboned-auto-multicast-14.txt> (Automatic Multicast Tunneling) to Proposed Standard
X-BeenThere: mboned@ietf.org
X-Mailman-Version: 2.1.12
Precedence: list
Reply-To: gjshep@gmail.com
List-Id: Mail List for the Mboned Working Group <mboned.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/options/mboned>, <mailto:mboned-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/mboned>
List-Post: <mailto:mboned@ietf.org>
List-Help: <mailto:mboned-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/mboned>, <mailto:mboned-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 17 Jan 2013 15:22:15 -0000

On Thu, Jan 17, 2013 at 7:16 AM, Martin Stiemerling
<martin.stiemerling@neclab.eu> wrote:
>
>
> On 01/16/2013 03:04 PM, Greg Shepherd wrote:
>>
>> Why would you require such restriction? If your concern is about
>> support for RTCP we already have RFC 5760 to address this. It's an
>> application layer problem.
>
>
> RFC 5760 has some nice security considerations, e.g., possible attacks on
> the Receiver-to-Distribution-Source Communication. It might be also good =
to
> document how AMT in conjunction with RFC 5760 is handled in actual
> deployments. E.g., AMT can introduce new issues, as the multicast source =
is
> located in a firewall protected network that blocks RTCP traffic coming f=
rom
> receivers somewhere else in the Internet.

Sourcing SSM content services behind a RTCP-blocking firewall while
expecting RTCP feedback from receivers is a massive operator error.

> AMT should discuss and document the changes in the deployment model as
> described in RFC 5760.

Wouldn't this be more suited for an AMT BCP draft?

Greg

>   Martin
>
>
>>
>> Greg
>>
>> On Wed, Jan 16, 2013 at 12:53 AM, Magnus Westerlund
>> <magnus.westerlund@ericsson.com> wrote:
>>>
>>> Hi,
>>>
>>> I think the major question here, are anyone really objecting to
>>> restricting the applicability of AMT to managed environments, i.e. wher=
e
>>> the full path between relay and client are in domains that are
>>> explicitly aware that they will have AMT traffic and approve of it. The
>>> comments appears to indicate that this is where it is deployed. If one
>>> makes such an applicability statement most of the significant comments
>>> can be more easily resolved.
>>>
>>> Cheers
>>>
>>> Magnus
>>>
>>> On 2013-01-10 17:35, Greg Shepherd wrote:
>>>>
>>>> On Wed, Jan 9, 2013 at 6:15 AM, Martin Stiemerling
>>>> <martin.stiemerling@neclab.eu> wrote:
>>>>>
>>>>> Hi Greg,
>>>>>
>>>>> I'm joining the discussion as the DISCUSS holder for most of the issu=
es
>>>>> below.
>>>>>
>>>>>
>>>>> On 01/09/2013 03:26 AM, Greg Bumgardner wrote:
>>>>>>
>>>>>>
>>>>>> Magnus,
>>>>>>
>>>>>> I would like to thank you for providing valuable feedback regarding
>>>>>> the
>>>>>> AMT spec.
>>>>>>
>>>>>> I would first like to state that many of your concerns and criticism=
s
>>>>>> regarding the protocol were raised during MBONED group discussions.
>>>>>> Many
>>>>>> alternative solutions were identified but ultimately rejected by the
>>>>>> group, primarily due to the reluctance of group members to make
>>>>>> significant changes to the protocol after several years of slow
>>>>>> development. My comments are inlined below...
>>>>>
>>>>>
>>>>>
>>>>> Can you give some insights why the protocol development took so long?
>>>>
>>>>
>>>> Yes, I can, though I'm not sure how relevant the incubation period is
>>>> to its adoption. And since I've breaking multicast networks for 15+
>>>> years now, asking a question like this may get you more than you were
>>>> expecting :)
>>>>
>>>> The need for multicast transition services is as old as multicast
>>>> itself. Deering's first work - DVMRP - had to assume there was no
>>>> 'native' path and so incorporated encapsulation as an integral part of
>>>> the protocol. I believe DVMRP may be the first use of overlay
>>>> encapsulation in the IETF, but I welcome you to dig for your self and
>>>> back with me. It's the earliest I've been able to find.
>>>>
>>>> Once we moved to PIM, encapsulation for receiver delivery was thrown
>>>> out and we all pushed to "go native". The reality was - and still is -
>>>> that those most wanting end-to-end native multicast (content owners)
>>>> had no way to motivate network operators to enable multicast services.
>>>> Deployment experience with PIM revealed to many the need for
>>>> encapsulation overlay mechanism, and from this several drafts were
>>>> introduced.
>>>>
>>>> In 2003 we held a BOF at IETF57 with the intention of agreeing on a
>>>> list of requirements and consolidating the efforts into a single
>>>> draft. We selected the AMT draft as being the closest fit, though
>>>> significant changes still need to be addressed. You see the list of
>>>> drafts and read the BOF minutes here:
>>>>
>>>> http://www.ietf.org/meeting/57/session/mlm.txt
>>>>
>>>> Authors came and went.. Progress was slow, until ATT picked up the
>>>> torch, demonstrated running code, and outlined their deployment plans.
>>>> Greg Bumgardner also wrote a java Gateway implementation and took over
>>>> as primary author. Finally we had solid industry support and a
>>>> qualified and motivated set of contributors.
>>>>
>>>> Last year we finally reached WG consensus, along with at least three
>>>> shipping implementations of both the Relay and the Gateway , and at
>>>> least one active and growing commercial deployment of a
>>>> IPMulticast/AMT service. I doubt many drafts in the editors' queue can
>>>> claim that much support. :)
>>>>
>>>>> Anyhow, I can understand the reluctance of changing parts of a protoc=
ol
>>>>> after such a long time. But there are a lot of rough edges in the
>>>>> protocol
>>>>> that need attention before moving it forward on the Standards Track.
>>>>
>>>>
>>>> Time is irrelevant to any reluctance to change - commercial
>>>> implementations and deployments are, however quite relevant.
>>>>
>>>> Cheers,
>>>> Greg
>>>>
>>>>>>
>>>>>> Thanks,
>>>>>>
>>>>>> -g.b.
>>>>>>
>>>>>>
>>>>>> On Wed, Dec 26, 2012 at 5:01 AM, Magnus Westerlund
>>>>>> <magnus.westerlund@ericsson.com
>>>>>> <mailto:magnus.westerlund@ericsson.com>>
>>>>>>
>>>>>> wrote:
>>>>>>
>>>>>>      Hi,
>>>>>>
>>>>>>      Sorry for being late with this IETF last call comments. I will
>>>>>>      partly blame the ADs requesting this Transport Directorate revi=
ew
>>>>>> a
>>>>>>      bit late, the other part is all mine and the holidays. Anyway, =
I
>>>>>> do
>>>>>>      hope you will consider these issues and comments as I believe I
>>>>>>      found some serious ones in addition to a number of clarificatio=
ns
>>>>>>      that should be made.
>>>>>>
>>>>>>
>>>>>>      Significant Issues:
>>>>>>
>>>>>>      1. Congestion Control
>>>>>>      This is clearly a tunnel establishment protocol of something th=
at
>>>>>> is
>>>>>>      IP traffic. Thus normally the responsibility for congestion
>>>>>> control
>>>>>>      is with the tunnelled traffic. However, I would like argue that
>>>>>> this
>>>>>>      does not apply in this case due to the nature of the tunnelled
>>>>>>      traffic, i.e. multicast traffic and secondary due to limitation=
s
>>>>>> in
>>>>>>      the tunnel protocol.
>>>>>>
>>>>>>      Lets start with the second part. This protocol claims to suppor=
t
>>>>>> ASM
>>>>>>      still don't provide a upstream delivery mechanism, i.e. an ASM
>>>>>>      receiver is not capable of sending as it should. This prevents
>>>>>>      several existing mechanism for congestion control that exist in
>>>>>>      protocols supporting multicast. The first is using RTCP for
>>>>>>      congestion control in ASM [RFC3550], the second isTCP-Friendly
>>>>>>      Multicast Congestion Control [RFC4654] that can be used in the
>>>>>> RMT
>>>>>>      suite of protocols, and I know has been implemented in some NOR=
M
>>>>>>      implementation. Thus onlystrictly receiver based mechanisms, su=
ch
>>>>>> as
>>>>>>
>>>>>>      Wave and Equation Based Rate Control [RFC3738] are available in
>>>>>> this
>>>>>>      context.
>>>>>>
>>>>>>
>>>>>> You are correct, the AMT protocol provides no method for injecting
>>>>>> multicast traffic into a multicast enabled network, and thus does no=
t
>>>>>> provide the functionality required for something like ASM RTP/RTCP
>>>>>> teleconferencing. The protocol is primarily intended for use in
>>>>>> receiving SSM traffic. Because the AMT protocol uses IGMP/MLD messag=
es
>>>>>> to establish reception/subscription state, the protocol MAY be used =
to
>>>>>> subscribe to ASM broadcasts, but cannot generate them.
>>>>>>
>>>>>> I should note that earlier versions of the AMT spec described a meth=
od
>>>>>> for broadcasting SSM via a gateway/relay tunnel, but this capability
>>>>>> was
>>>>>> removed due to feasibility considerations.
>>>>>
>>>>>
>>>>>
>>>>> This means that the draft should state that it does not support ASM,
>>>>> but
>>>>> only a subset of it. Further, it cuts away the feedback over RTCP and
>>>>> RFC
>>>>> 4654.
>>>>>
>>>>>>
>>>>>>
>>>>>>      Secondly, many multicast usages are in fact deployed without an=
y
>>>>>>      congestion control. This based on that the deploying entity
>>>>>> controls
>>>>>>      the scope and authorization for requesting multicast delivery.
>>>>>>      However, does restrictions does not apply to AMT delivery of
>>>>>>      multicast. If the gateway can reach using unicast the relay it
>>>>>> can
>>>>>>      be delivered the multicast group from the domain the relay is
>>>>>>      attached to. Thus, this protocol changes the deployment
>>>>>> restrictions
>>>>>>      of multicast which many non-congestion controlled delivery is
>>>>>> based
>>>>>>      on. Instead the non-congestion controlled trafficcan now sent
>>>>>> over
>>>>>>
>>>>>>      an IP/UDP tunnel over Internet where neither relay nor gateway
>>>>>> may
>>>>>>      have any knowledge about the path the traffic may take.
>>>>>>
>>>>>>
>>>>>> The AMT protocol does not attempt to address the issue of congestion
>>>>>> control. The group decided to treat the detection and remediation of
>>>>>> congestion (of encapsulated multicast packets) as an application-lay=
er
>>>>>> responsibility. While we might have attempted to provide some level =
of
>>>>>> QoS control on tunnel packets by copying ToS/DS bits from inner to
>>>>>> outer
>>>>>> headers, this would only occur on a packet-by-packet basis on not on
>>>>>> the
>>>>>> tunnel as a whole.
>>>>>
>>>>>
>>>>>
>>>>> What's the group's view on how the application layer is handling
>>>>> congestion
>>>>> control in this case when using AMT?
>>>>>
>>>>>
>>>>>>
>>>>>>
>>>>>>      Based on this I would like to see two changes to this protocol
>>>>>>      specification. First a section discussing the issue of congesti=
on
>>>>>>      control.
>>>>>>
>>>>>>
>>>>>> I'm not sure what, if anything, I might add to address this. I think
>>>>>> the
>>>>>> spec makes it pretty clear that the tunneling is provided by
>>>>>> best-effort
>>>>>> delivery of UDP packets (perhaps restricted by relay/router
>>>>>> configuration). Perhaps you could suggest some text.
>>>>>
>>>>>
>>>>>
>>>>> There are multiple issues here:
>>>>> - there is no feedback loop between relay and gateway so that any
>>>>> information about the tunneled multicast traffic can be sent back fro=
m
>>>>> the
>>>>> gateway to the relay. This feedback loop can be used for congestion
>>>>> control,
>>>>> but also to figure out if the non-zero IPv6/UDP traffic gets black
>>>>> holed
>>>>> somewhere.
>>>>> - the tunneled traffic can exceed the capacity of the network path
>>>>> between
>>>>> relay and gateway. This needs to be monitored and acted upon any
>>>>> congestion.
>>>>> - since a single relay can feed many gateways, even tunneled multicas=
t
>>>>> flows
>>>>> from the same relay can compete for network path capacity. I.e., if t=
he
>>>>> tunneled traffic shares a network path. In the worst case, an added
>>>>> multicast flow can lead to congestion for all other tunneled multicas=
t
>>>>> flows.
>>>>> - any tunneled flow can impact other traffic from other hosts in the
>>>>> Internet.
>>>>>
>>>>>>
>>>>>>      Secondly, I think this protocol should have an applicability
>>>>>>      statement limiting its deployment to restricted environments
>>>>>> where
>>>>>>      the relay and gateway deployers can provide certain resource
>>>>>>      provisions between the entities to avoid the multicast traffic
>>>>>>      affecting other traffic sharing the same bottlenecksin ways not
>>>>>>
>>>>>>      allowed by the network provider.
>>>>>>
>>>>>>
>>>>>> While I understand your concern, this would essentially negate the
>>>>>> entire reason for the protocol, which is to provide multicast transi=
t
>>>>>> to
>>>>>> non-multicast connected receivers. However, the AMT spec does not
>>>>>> preclude the implementation/deployment of AMT-enabled gateway device=
s
>>>>>> for which AMT support may be enabled or disabled (or blocked via a
>>>>>> firewall).
>>>>>
>>>>>
>>>>>
>>>>> The point is that it is not safe to operate in the public Internet wi=
th
>>>>> the
>>>>> current protocol. Either with fix the issues in the protocol or we
>>>>> limit the
>>>>> applicability.
>>>>>
>>>>>
>>>>>>
>>>>>>
>>>>>>
>>>>>>      2. Security
>>>>>>
>>>>>>      This protocols is frank with it having limited security feature=
s
>>>>>> and
>>>>>>      says this is similar to the IGMP and MLD protocols being used.
>>>>>>      However, I think this is a failure to propoerly consider the
>>>>>> threat
>>>>>>      model. If one uses AMT over general Internet it will run in a
>>>>>>      network where the one deploying the multicast and the relays no
>>>>>>      longer control requirements on source address verification or
>>>>>>      possibilities for traffic separation as they can do within the
>>>>>>      domains where multicast currently are deployed. The security
>>>>>>      vulnerabilities in IGMP and MLD are much more contained and
>>>>>>      controllable in a LAN environment where one has chosen to deplo=
y
>>>>>>      multicast compared to an Relay exposing this to the whole
>>>>>> Internet.
>>>>>>      Once more I think there is only two choices here.
>>>>>>
>>>>>>      A) Beef up the security to general Internet threat model, i.e. =
at
>>>>>> a
>>>>>>      minimal provide a real model for gateway authentication using
>>>>>>      identities, not only return routability based verifications.
>>>>>>
>>>>>>      B) Limit the applicability of AMT to managed environments and
>>>>>> make
>>>>>>      it clear that the relay will need to limit which gateways are
>>>>>>      allowed to access the relay based on addressing.
>>>>>>
>>>>>>      Based on the first significant issue with congestion control I
>>>>>>      expect that there is little meaning to do A) unless also one is
>>>>>>      willing to beef up AMT to provide congestion control. Which I
>>>>>> think
>>>>>>      is not according to the design wishes for the protocol designer=
s.
>>>>>>
>>>>>>
>>>>>> The security topic was visited and revisited numerous times over the
>>>>>> last couple of years. The group made the decision to limit the
>>>>>> protocol
>>>>>> to simple source address verification and leave it at that. An
>>>>>> emphasis
>>>>>> was placed on protecting relays against simple DoS attacks but not
>>>>>> gateways.
>>>>>
>>>>>
>>>>>
>>>>> I can see the point, but I assume that most people are just looking t=
o
>>>>> closed environments, e.g., multicast in access networks, such as xDSL
>>>>> or
>>>>> 3GPP. However, AMT assumes to run across the public Internet and
>>>>> threats are
>>>>> extremely different there. Magnus made a good proposal that is
>>>>> relatively
>>>>> easy to implement.
>>>>>
>>>>>>
>>>>>>
>>>>>>      3. Use of Zero Checksum
>>>>>>
>>>>>>      The AMT specification enables the use of Zero UDP checksum with
>>>>>>      IPv6, i.e.
>>>>>>      draft-ietf-6man-udpchecksums-06
>>>>>>      <http://datatracker.ietf.org/doc/draft-ietf-6man-udpchecksums/>
>>>>>> and
>>>>>>      draft-ietf-6man-udpzero-08
>>>>>>      <http://datatracker.ietf.org/doc/draft-ietf-6man-udpzero/>.
>>>>>> Nothing
>>>>>>
>>>>>>      against this in principal. However, I have noticed that AMT fai=
ls
>>>>>> to
>>>>>>      properly address the failure modes of using a zero-checksum. AM=
T
>>>>>> is
>>>>>>      a typical example of a protocol that actually need active
>>>>>>      verification of each tunnel that zero-checksum functions. This =
as
>>>>>>      AMT is clearly intended to be deployed with its Gateway part in
>>>>>>      end-hosts and residential network devices or routers. This mean=
s
>>>>>> the
>>>>>>      tunnel will pass through both firewalls and NATs on its path
>>>>>> between
>>>>>>      the relay and the gateway. Unless these devices are not upgrade=
d
>>>>>> to
>>>>>>      support zero-checksum in UDP for IPv6 the traffic may actually
>>>>>>      become black holed. The most likely is a simple firewall that h=
as
>>>>>> a
>>>>>>      rule for IPv6/UDP which doesn't allow zero checksum as it is
>>>>>> against
>>>>>>      RFC 2460. Thus all the Multicast data packets will disappear on
>>>>>>      route in the tunnel. There is no mechanism in the AMT protocol =
to
>>>>>>      detect this and negotiate with the relay so that it will not us=
e
>>>>>> the
>>>>>>      zero-checksum for this tunnel.
>>>>>>
>>>>>>      This must be addressed as I see it. If not the AMT will be so
>>>>>>      brittle that it can't be used in a large number of its intended
>>>>>>      deployments.
>>>>>>
>>>>>>
>>>>>> You probably have no idea how often this issue was raised and
>>>>>> discussed
>>>>>> within the group. The consensus was that there was no single solutio=
n
>>>>>> to
>>>>>> the problem. The only workable solution I can suggest is to use an
>>>>>> IPv4
>>>>>> tunnel to carry the IPv6 packets. The mechanism for deciding whether
>>>>>> (or
>>>>>> if) to use an IPv4 tunnel is considered out of scope.
>>>>>
>>>>>
>>>>>
>>>>> There is a solution, as mentioned above: feedback loop. If the relay
>>>>> says
>>>>> "sending you x packets" and the gateway sees zero, there is an issue.
>>>>> In the
>>>>> easiest case, a log file entry is generate saying that the tunneled
>>>>> traffic
>>>>> couldn't reach this particular gateway.
>>>>>
>>>>>
>>>>>>
>>>>>>
>>>>>>      4) MTU issues
>>>>>>
>>>>>>      This document total fails to discuss the issues of MTU
>>>>>> blackholing.
>>>>>>      As the IP multicast datagrams as well as the encapsulated
>>>>>> IGMP/MLD
>>>>>>      messages can with the added tunnel-overhead result in that the
>>>>>> sent
>>>>>>      packet exceeds the MTU of the path, these packets could be blac=
k
>>>>>>      holed. This can potentially result in very intermittent transpo=
rt
>>>>>>      behavior for the tunnel. Thus, some discussion of how do handle
>>>>>> the
>>>>>>      MTU issues in this context should be introduced.
>>>>>>
>>>>>>
>>>>>> Yep. This could be a problem that has no easy solution. Not only mig=
ht
>>>>>> the outer packets be fragmented, but the encapsulated packets might =
be
>>>>>> fragments. Since the protocol obviously has no control over the size
>>>>>> of
>>>>>> source packets, it cannot really avoid the potential need to fragmen=
t
>>>>>> the tunnel packets. A gateway can only rely on the underlying UDP/IP
>>>>>> stack to provide the necessary reassembly of the tunnel packets. Onc=
e
>>>>>> the tunnel packet has been reassembled, the gateway, host IP stack o=
r
>>>>>> application may still need to reassemble the contents of a sequence =
of
>>>>>> tunnel packets.
>>>>>
>>>>>
>>>>>
>>>>> In the case the WG doesn't have any solution, which I couldn't believ=
e,
>>>>> it
>>>>> is worth to document the issues. One operational solution can be to
>>>>> limit
>>>>> the MTU in the native multicast network, if this part and the relay a=
re
>>>>> under the same administrative control. However, this would require th=
at
>>>>> the
>>>>> AMT relay logs any issue with the MTU accross the tunnel.
>>>>> Further, Magnus point is about that the tunneled packets can get
>>>>> fragmented,
>>>>> less about the inner parts of the tunnel.
>>>>>
>>>>> Addressing the MTU issue will increase the viability of this a lot.
>>>>>
>>>>>
>>>>>>
>>>>>>
>>>>>>      I am willing to discuss methods here, but I guess several
>>>>>>      alternatives exists and thus which is most appropriate and the
>>>>>> level
>>>>>>      of AMT support for them varies I would like the protocol
>>>>>> designers
>>>>>>      to do a first stab at resolving this.
>>>>>>
>>>>>>      Other Issues
>>>>>>
>>>>>> =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>>>>>>
>>>>>>      A) The table of content on page 2 should include more levels of
>>>>>>      headings. Most likely down to 4 is needed to make the TOC usabl=
e
>>>>>> for
>>>>>>      finding content in the document.
>>>>>>
>>>>>>      B) The claimed ASM support
>>>>>>      I would like to better understand how one can claim to support
>>>>>> ASM
>>>>>>      when one has no up stream path to inject the ASM group
>>>>>> participants
>>>>>>      traffic. When one uses ASM one normally does this for a reason
>>>>>> and
>>>>>>      needing the possibility to inject packets into the group. These
>>>>>>      limitations needs to be clarified.
>>>>>>
>>>>>>
>>>>>> See above. The protocol is intended for multicast reception. It make=
s
>>>>>> no
>>>>>> assumptions about application layer use of the multicast traffic. It
>>>>>> can
>>>>>> only be used to receive ASM IP traffic.
>>>>>
>>>>>
>>>>>
>>>>> Document it, please. Otherwise, the draft is incomplete.
>>>>>
>>>>> Thank you,
>>>>>
>>>>>    Martin
>>>>> --
>>>>> IETF Transport Area Director
>>>>>
>>>>> martin.stiemerling@neclab.eu
>>>>>
>>>>> NEC Laboratories Europe - Network Research Division NEC Europe Limite=
d
>>>>> Registered Office: NEC House, 1 Victoria Road, London W3 6BL
>>>>> Registered in England 283
>>>>>
>>>>> _______________________________________________
>>>>> MBONED mailing list
>>>>> MBONED@ietf.org
>>>>> https://www.ietf.org/mailman/listinfo/mboned
>>>>
>>>>
>>>
>>>
>>> --
>>>
>>> Magnus Westerlund
>>>
>>> ----------------------------------------------------------------------
>>> Multimedia Technologies, Ericsson Research EAB/TVM
>>> ----------------------------------------------------------------------
>>> Ericsson AB                | Phone  +46 10 7148287
>>> F=E4r=F6gatan 6                | Mobile +46 73 0949079
>>> SE-164 80 Stockholm, Sweden| mailto: magnus.westerlund@ericsson.com
>>> ----------------------------------------------------------------------
>>>
>
> --
> martin.stiemerling@neclab.eu
>
> NEC Laboratories Europe - Network Research Division NEC Europe Limited
> Registered Office: NEC House, 1 Victoria Road, London W3 6BL
> Registered in England 283
