From mailnull@www1.ietf.org  Sun Dec  1 11:30:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA09788
	for <ieprep-archive@odin.ietf.org>; Sun, 1 Dec 2002 11:30:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB1GWpE25522
	for ieprep-archive@odin.ietf.org; Sun, 1 Dec 2002 11:32:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1GWpv25519
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 1 Dec 2002 11:32:51 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA09777
	for <ieprep-web-archive@ietf.org>; Sun, 1 Dec 2002 11:29:59 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1GW6v25506;
	Sun, 1 Dec 2002 11:32:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1GVZv25485
	for <ieprep@optimus.ietf.org>; Sun, 1 Dec 2002 11:31:35 -0500
Received: from samsara.kgbinternet.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA09756
	for <ieprep@ietf.org>; Sun, 1 Dec 2002 11:28:39 -0500 (EST)
Received: (from nobody@localhost)
	by samsara.kgbinternet.com (8.11.6/8.11.6) id gB1GVPA19487;
	Sun, 1 Dec 2002 10:31:25 -0600
Date: Sun, 1 Dec 2002 10:31:25 -0600
Message-Id: <200212011631.gB1GVPA19487@samsara.kgbinternet.com>
X-Authentication-Warning: samsara.kgbinternet.com: nobody set sender to carl@ietfwatch.net using -f
From: "Carl Ford" <carl@ietfwatch.net>
To: Bill Woodcock <woody@pch.net>, Carl Ford <carl@ietfwatch.net>,
        Mpierce1@aol.com, <rja@extremenetworks.com>, <ieprep@ietf.org>
X-Mailer: Open WebMail 1.52 20011204
X-OriginatingIP: 67.81.93.146 (ietfwatch.carl)
MIME-Version: 1.0
Content-Type: text/plain; charset=iso-8859-1
Subject: [Ieprep] Re: Looking for governmental feedback as well
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


Bill Woodcock wrote
>  I guess it would
> be more accurate to say that there's a relatively widespread understanding
> that these are desirable purchasing criteria, and that they guide the
> purchasing decisions of many of the larger and better-informed buyers of
> Internet services.  But they're not _majority_ practices, since the
> _majority_ of sites aren't multi-homed in the first place, only larger
> ones are.
> 
So the question that we have to ask for ieprep is do we have a goal of greater 
redundancy for the Internet overall, greater redundancy for the government sites 
specifically, or a specific practise on a government access methods.

Any thoughts appreciated.

Kind Regards,

Carl
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  1 11:58:22 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA10465
	for <ieprep-archive@odin.ietf.org>; Sun, 1 Dec 2002 11:58:22 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB1H0hl31885
	for ieprep-archive@odin.ietf.org; Sun, 1 Dec 2002 12:00:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1H0hv31882
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 1 Dec 2002 12:00:43 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA10450
	for <ieprep-web-archive@ietf.org>; Sun, 1 Dec 2002 11:57:51 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1H02v31859;
	Sun, 1 Dec 2002 12:00:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1Gx9v31690
	for <ieprep@optimus.ietf.org>; Sun, 1 Dec 2002 11:59:09 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA10420
	for <ieprep@ietf.org>; Sun, 1 Dec 2002 11:56:18 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Sun, 1 Dec 2002 11:58:42 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120111585827701
 ; Sun, 01 Dec 2002 11:58:58 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <XNT4AS3J>; Sun, 1 Dec 2002 11:58:46 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E779C6@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Steve Silverman '" <steves@shentel.net>,
        "'RJ Atkinson '" <rja@extremenetworks.com>,
        "'Sean Donelan '" <sean@donelan.com>
Cc: "'ieprep@ietf.org '" <ieprep@ietf.org>
Subject: RE: Try it, you might like it RE: [Ieprep] any commercial usage for "emergency QoS"?
Date: Sun, 1 Dec 2002 11:58:55 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

>I would very much like to hear from anyone with real experience with
>VoIP over a typically loaded WAN.

Maybe you can track down someone at Worldcom for the real experience
http://www.worldcom.com/us/products/connection/enterprise/faq/

Yes, this is designed for shared VoIP&data over an access link to a WAN. 

Congestion?
"What happens to my WorldCom Connection calling when there is a slowdown or
congestion in the network? 

...We have found that congestion is more likely to occur on the access
circuits, again reinforcing the need for a well designed network"

Firewalls?
"How does my firewall impact The WorldCom Connection? 

Today's firewalls are designed to block Session Initiation Protocol (SIP)
signaling. As a result, voice packets must utilize a pinhole in the firewall
to reach their destination. 

However, many firewall vendors have been updating their offerings to support
SIP. WorldCom has certified several SIP enabled firewalls which can be used
alongside your existing firewall and will allow SIP traffic to pass to phone
users."

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  1 15:17:52 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14772
	for <ieprep-archive@odin.ietf.org>; Sun, 1 Dec 2002 15:17:51 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB1KKFo30174
	for ieprep-archive@odin.ietf.org; Sun, 1 Dec 2002 15:20:15 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1KKFv30171
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 1 Dec 2002 15:20:15 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14766
	for <ieprep-web-archive@ietf.org>; Sun, 1 Dec 2002 15:17:20 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1KJHv30116;
	Sun, 1 Dec 2002 15:19:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1KIuv30100
	for <ieprep@optimus.ietf.org>; Sun, 1 Dec 2002 15:18:56 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14734
	for <ieprep@ietf.org>; Sun, 1 Dec 2002 15:16:01 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.140])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 06CD77A0A; Sun,  1 Dec 2002 12:18:46 -0800 (PST)
Date: Sun, 1 Dec 2002 15:18:48 -0500
Subject: Re: Try it, you might like it RE: [Ieprep] any commercial usage for "emergency QoS"?
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: <ieprep@ietf.org>
To: "Steve Silverman" <steves@shentel.net>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <KMEJKAPHBBCMCDDGMBLOIEAHCBAA.steves@shentel.net>
Message-Id: <19749878-056A-11D7-A42D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Nov 29, 2002, at 22:32 America/Montreal, Steve Silverman 
wrote:

> While advertised as VoIP, the Skyline building, like most
> implementations is actually VoLAN.

That's fair, but the bits on the wire are roughly <voice>, UDP, IP,
and Ethernet.  The term "Voice over IP" means *any* deployment where
voice is carried over IP, regardless of whether LAN, MAN, WAN, or
inter-planetary.

> From what I've heard, very few users are using VoIP on a WAN where
> the IP has to contend with moderately loaded (> 30% loaded lines).
> I would very much like to hear from anyone with real experience with
> VoIP over a typically loaded WAN.

That would be Bill Woodcock, who is going blue in the face trying to
explain that all of his experience is that his "real experience with
VoIP over a (wide range of) typically loaded WAN(s)" is that it just
works.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  1 15:30:33 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14960
	for <ieprep-archive@odin.ietf.org>; Sun, 1 Dec 2002 15:30:33 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB1KWv330432
	for ieprep-archive@odin.ietf.org; Sun, 1 Dec 2002 15:32:57 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1KWvv30429
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 1 Dec 2002 15:32:57 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14951
	for <ieprep-web-archive@ietf.org>; Sun, 1 Dec 2002 15:30:02 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1KW1v30408;
	Sun, 1 Dec 2002 15:32:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1KVjv30390
	for <ieprep@optimus.ietf.org>; Sun, 1 Dec 2002 15:31:45 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14942
	for <ieprep@ietf.org>; Sun, 1 Dec 2002 15:28:50 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.140])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 86CF47A0A; Sun,  1 Dec 2002 12:31:35 -0800 (PST)
Date: Sun, 1 Dec 2002 15:31:36 -0500
Subject: Re: [Ieprep] alternate model for edge QoS
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Henning Schulzrinne <hgs@cs.columbia.edu>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <3DE8C92F.3090601@cs.columbia.edu>
Message-Id: <E399CC3B-056B-11D7-A42D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Saturday, Nov 30, 2002, at 09:20 America/Montreal, Henning 
Schulzrinne wrote:
> To be more precise, we have to distinguish three types of states:
>
> (1) scheduling state; this only matters if you have per-flow queueing
> (2) filtering/policing state, which polices flows;
> (3) reservation state, one for each reservation, but only affecting 
> the control processor
>
> From what I can tell, (3) is not a major issue. Given the hundreds of 
> MB on a backbone router and the tens of MB on an edge router and the 
> roughly 500 bytes/reservation, you'd need lots of reservations to even 
> come close to mattering.

	Nope.  Each flow reservation is also likely to add complexity to (1) 
for
packet forwarding of the preferred flows and to (2) because the 
reservation
is probably internally a kind of filter.

	The German academic reports are not by themselves credible or 
sufficient.
More interesting and applicable reports would involve a typical 
mid-range
commercial IP router running a general deployment software load with a
typical packet forwarding load, typical sized routing table, and so 
forth.

> Web servers routinely handle large volumes of TLS connections.

	Web servers these days usually have *expensive* dedicated hardware
to handle TLS and in MANY cases the TLS has been off-loaded to an
*expensive* front-end processor (this box often is also the 
load-balancer
widget in front of the set of servers implementing the web site).

	Commercial routers do not have such dedicated crypto hardware at hand.
It could be added, but would significantly increase complexity of the
router and the user cost of the router -- and what about all the 
deployed
routers that won't be swapped out overnight.

>> Please do explain more.  I don't understand the details of who gives
>> whom a CERT about what, when, and when whom needs to do what with it.
>
> For concreteness, let's assume that FEMA is in charge of this 
> hypothetical program. FEMA sends, in some secure fashion, a cert to 
> each operator it cares to (based on whatever business relationships it 
> might set up). Each operator installs the CA cert in the control 
> processor, just like any other CA root cert.

	Right now, zero CA root certs are in any normally deployed control 
processor,
so you're assuming a bunch of stuff not in any mainstream deployed 
router today.

> Obviously, this happens during non-emergency conditions and can be as 
> simple as a signed S/MIME mail. There's no need for secrecy, just 
> integrity, since this is a public key only. Alternatively, the 
> operator can just download the .p12 file from the FEMA web server. 
> There may even be automated ways of doing this; I haven't followed all 
> the recent PKI-related WGs. This operation is repeated at cert 
> lifetime intervals; these seem to be roughly yearly.

	So the ISP now has to manage certificates in routers, make sure the 
certificate
stays in the router even if the CP gets swapped out due to a fault 
(faults of that
sort happened to me many a time on reasonably modern backbone routers). 
  Increases
the operational complexity and operational costs pretty significantly.

> When a signed resource request comes along that claims to be signed by 
> the FEMA CA, the control processor verifies this assertion. If this 
> checks out, the holder of the cert (presumably, an authorized 
> emergency worker) gets her reservation, if not, the attempt fails. No 
> real-time interaction with outside entities is required.

	It is an expensive computation to verify a signature.  An obvious DDOS 
attack is
to flood the router with forged resource requests and force the router 
to spend
lots of time performing a bunch of expensive computations.  Bad idea 
for one to enable
such router validation if one wants the router to keep BGP in sync, 
forward packets,
and do sundry other stuff needed to keep the Internet up and running.

	Sorry, this just is not deployable stuff right now, even if the fantasy
crypto accelerator hardware were shipping today in most commercial IP 
routers.

Ran


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  1 15:36:42 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15011
	for <ieprep-archive@odin.ietf.org>; Sun, 1 Dec 2002 15:36:42 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB1Kd4w31113
	for ieprep-archive@odin.ietf.org; Sun, 1 Dec 2002 15:39:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1Kd4v31110
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 1 Dec 2002 15:39:04 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15005
	for <ieprep-web-archive@ietf.org>; Sun, 1 Dec 2002 15:36:11 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1Kc2v31083;
	Sun, 1 Dec 2002 15:38:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1Kb7v30687
	for <ieprep@optimus.ietf.org>; Sun, 1 Dec 2002 15:37:07 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA14996
	for <ieprep@ietf.org>; Sun, 1 Dec 2002 15:34:15 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.140])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 921FA7A0A; Sun,  1 Dec 2002 12:36:58 -0800 (PST)
Date: Sun, 1 Dec 2002 15:36:59 -0500
Subject: Re: [Ieprep] Accounting
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <21739.1038667538@cs.ucl.ac.uk>
Message-Id: <A429E53E-056C-11D7-A42D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Saturday, Nov 30, 2002, at 09:45 America/Montreal, Ken Carlberg 
wrote:
> one thing I can probably do in the draft-carlberg-ets-telephony-00.txt
> draft is take a rough inverse of your statement and state that
> the ability to perform accounting must not be precluded.

	That seems quite reasonable to me.

	And I checked with some ISP folks in Japan. They have no requirement
for accounting as part of their *deployed* IEprep service.  In fact,
they claim that their *deployed* IEprep service is identical (*no 
changes*)
to their deployed regular commercial IP service.

	Its fascinating that the several countries with a deployed IEprep 
service
"require" so little, but the country with the big paid staff and zero 
IEprep
deployment "requires" so much.

	Sigh.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  1 15:42:54 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15161
	for <ieprep-archive@odin.ietf.org>; Sun, 1 Dec 2002 15:42:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB1KjG731253
	for ieprep-archive@odin.ietf.org; Sun, 1 Dec 2002 15:45:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1KjGv31250
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 1 Dec 2002 15:45:16 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15158
	for <ieprep-web-archive@ietf.org>; Sun, 1 Dec 2002 15:42:23 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1Ki3v31230;
	Sun, 1 Dec 2002 15:44:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1Kh9v31210
	for <ieprep@optimus.ietf.org>; Sun, 1 Dec 2002 15:43:09 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15143
	for <ieprep@ietf.org>; Sun, 1 Dec 2002 15:40:16 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.140])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id D57767A0A; Sun,  1 Dec 2002 12:42:59 -0800 (PST)
Date: Sun, 1 Dec 2002 15:43:01 -0500
Subject: Re: [Ieprep] a practical VoIP issue
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E779C5@mcl-its-exs01.mail.saic.com>
Message-Id: <7B8A515A-056D-11D7-A42D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Saturday, Nov 30, 2002, at 15:57 America/Montreal, King, Kimberly S. 
wrote:
>> Apparently VoIP systems use a wide range of dynamically
>> assigned UDP ports, which most firewalls block and which
>> most NATs cannot cope with properly.
>>
>> 	If one really wants to be able to use Internet
>> infrastructure during an emergency, designing a VoIP system
>> that can do NAT/Firewall traversal might be a pretty important 
>> component.
>
> There exist SIP aware firewalls that can cope with NAT and dynamically
> assigned UDP ports.  Are you suggesting something different?

	That is interesting and helpful, but maybe not sufficient.

	Not all VoIP deployments use SIP, for openers.

	Most *deployed* firewalls don't open enough of the UDP ports
*because the policy choice of the firewall operator is to block
those ports*.  I don't see how it VoIP service will automatically work
through a majority of deployed firewalls as long as it needs so many 
ports
left wide open for an adversary.

	For pre-planned VoIP deployments, this can be worked through by
getting policies changed, upgrading firewall capabilities, and so forth.
For IEprep use of VoIP, the reality is that most access links have a 
firewall
on the customer end and that firewall is likely to be pre-configured in 
a
manner that blocks IEprep use of VoIP.

	I actually think that this kind of issue (including NAT Traversal,
which is closely related) is a huge architectural headache for The
Internet.  I don't see any sort of quick fix, but it is a real problem.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  1 16:26:24 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA15763
	for <ieprep-archive@odin.ietf.org>; Sun, 1 Dec 2002 16:26:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB1LSk400327
	for ieprep-archive@odin.ietf.org; Sun, 1 Dec 2002 16:28:46 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1LSkv00324
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 1 Dec 2002 16:28:46 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA15747
	for <ieprep-web-archive@ietf.org>; Sun, 1 Dec 2002 16:25:52 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1LRPv32752;
	Sun, 1 Dec 2002 16:27:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1LQgv32716
	for <ieprep@optimus.ietf.org>; Sun, 1 Dec 2002 16:26:42 -0500
Received: from cs.columbia.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA15731
	for <ieprep@ietf.org>; Sun, 1 Dec 2002 16:23:48 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB1LQYl8026963
	(version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NOT);
	Sun, 1 Dec 2002 16:26:34 -0500 (EST)
Received: from cs.columbia.edu (tallgrass.netlab.uky.edu [204.198.76.66])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB1LQWdG004163
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Sun, 1 Dec 2002 16:26:33 -0500 (EST)
Message-ID: <3DEA7EE0.8090501@cs.columbia.edu>
Date: Sun, 01 Dec 2002 16:28:00 -0500
From: Henning Schulzrinne <hgs@cs.columbia.edu>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.2b) Gecko/20021016
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: RJ Atkinson <rja@extremenetworks.com>
CC: ieprep <ieprep@ietf.org>
Subject: Re: [Ieprep] alternate model for edge QoS
References: <E399CC3B-056B-11D7-A42D-00039357A82A@extremenetworks.com>
In-Reply-To: <E399CC3B-056B-11D7-A42D-00039357A82A@extremenetworks.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


>     Nope.  Each flow reservation is also likely to add complexity to (1)
> for
> packet forwarding of the preferred flows

The whole point is that all of them are mapped (after filtering) to one 
class, so there is no additional scheduling state - it's one priority queue.

I just looked, more or less randomly, at the Juniper ERX edge router. It 
claims the ability to support 200,000 filters.


>  and to (2) because the reservation
> is probably internally a kind of filter.
>
>
>
> > Web servers routinely handle large volumes of TLS connections.
>
>
>     Web servers these days usually have *expensive* dedicated hardware
> to handle TLS and in MANY cases the TLS has been off-loaded to an
> *expensive* front-end processor (this box often is also the load-balancer
> widget in front of the set of servers implementing the web site).


The numbers I cited did not use any accelerator (the paper also contains 
numbers for those, which are indeed much higher).

>
>
>     Right now, zero CA root certs are in any normally deployed control
> processor,
> so you're assuming a bunch of stuff not in any mainstream deployed
> router today.

Please recall that the discussion started with my assumption of adding 
NSIS functionality, which is likely to require both IPsec and possibly 
TLS, depending on how the design shapes up. Clearly, this is not 
something that you can plug into today's router. The point of the 
discussion was to see what, if any, requirements might be helpful for 
protocols being developed elsewhere in the IETF.


>
>
> > Obviously, this happens during non-emergency conditions and can be as
> > simple as a signed S/MIME mail. There's no need for secrecy, just
> > integrity, since this is a public key only. Alternatively, the
> > operator can just download the .p12 file from the FEMA web server.
> > There may even be automated ways of doing this; I haven't followed all
> > the recent PKI-related WGs. This operation is repeated at cert
> > lifetime intervals; these seem to be roughly yearly.
>
>
>     So the ISP now has to manage certificates in routers, make sure the
> certificate
> stays in the router even if the CP gets swapped out due to a fault
> (faults of that
> sort happened to me many a time on reasonably modern backbone routers).
>  Increases
> the operational complexity and operational costs pretty significantly.

Again, the worse case scenario is that the cert is gone when the 
earthquake hits. Means you get no-preference service. I suspect many 
routers already support ssh, so they already deal with PK management, so 
this is hardly earth-shattering complexity.


>
>
> > When a signed resource request comes along that claims to be signed by
> > the FEMA CA, the control processor verifies this assertion. If this
> > checks out, the holder of the cert (presumably, an authorized
> > emergency worker) gets her reservation, if not, the attempt fails. No
> > real-time interaction with outside entities is required.
>
>
>     It is an expensive computation to verify a signature.  An obvious
> DDOS attack is
> to flood the router with forged resource requests and force the router
> to spend
> lots of time performing a bunch of expensive computations.  Bad idea for
> one to enable
> such router validation if one wants the router to keep BGP in sync,
> forward packets,
> and do sundry other stuff needed to keep the Internet up and running.

As noted, low-end PCs can do hundreds of such verifications a second. 
This is more than enough for the purposes of what's needed here.


>
>     Sorry, this just is not deployable stuff right now, even if the 
> fantasy
> crypto accelerator hardware were shipping today in most commercial IP
> routers.
>
> Ran
>

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  1 18:22:49 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17878
	for <ieprep-archive@odin.ietf.org>; Sun, 1 Dec 2002 18:22:49 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB1NPDI06072
	for ieprep-archive@odin.ietf.org; Sun, 1 Dec 2002 18:25:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1NPDv06069
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 1 Dec 2002 18:25:13 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17875
	for <ieprep-web-archive@ietf.org>; Sun, 1 Dec 2002 18:22:18 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1NO5v06034;
	Sun, 1 Dec 2002 18:24:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB1NNQv06005
	for <ieprep@optimus.ietf.org>; Sun, 1 Dec 2002 18:23:26 -0500
Received: from sith.maoz.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17860
	for <ieprep@ietf.org>; Sun, 1 Dec 2002 18:20:30 -0500 (EST)
Received: (from dmm@localhost)
	by sith.maoz.com (8.12.6/8.12.6) id gB1NO5sa022309;
	Sun, 1 Dec 2002 15:24:05 -0800
Date: Sun, 1 Dec 2002 15:24:05 -0800
From: David Meyer <dmm@sprint.net>
To: ieprep@ietf.org
Cc: Steve Silverman <steves@shentel.net>, rja@extremenetworks.com
Subject: Re: Try it, you might like it RE: [Ieprep] any commercial usage for "emergency QoS"?
Message-ID: <20021201152405.A22286@sprint.net>
References: <KMEJKAPHBBCMCDDGMBLOIEAHCBAA.steves@shentel.net> <19749878-056A-11D7-A42D-00039357A82A@extremenetworks.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <19749878-056A-11D7-A42D-00039357A82A@extremenetworks.com>; from rja@extremenetworks.com on Sun, Dec 01, 2002 at 03:18:48PM -0500
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

>> > From what I've heard, very few users are using VoIP on a WAN where
>> > the IP has to contend with moderately loaded (> 30% loaded lines).

That is not what I have heard, nor is it the experience that we
have had in the State of Oregon. We trunk many of our Class 5
switches together using VoIP, and that traffic runs over the
state internal network (which aggregates traffic from all of the
universities in the state for delivery either to another
university or to the Internet at large). We haven't seen any real
problems.

>> > I would very much like to hear from anyone with real experience with
>> > VoIP over a typically loaded WAN.
>> 
>> That would be Bill Woodcock, who is going blue in the face trying to
>> explain that all of his experience is that his "real experience with
>> VoIP over a (wide range of) typically loaded WAN(s)" is that it just
>> works.

Add to this that all ISPs have large MNC customers who use their
own VoIP infrastructure over Internet infrastructures (i.e.,
stat-muxed with their Internet traffic that the send to their
service provider) for toll bypass. This is especially prevalent
in the the international cause (although there is quite a bit of
domestic VoIP flying around). 

Dave
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  1 19:51:05 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA19119
	for <ieprep-archive@odin.ietf.org>; Sun, 1 Dec 2002 19:51:05 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB20rUG09938
	for ieprep-archive@odin.ietf.org; Sun, 1 Dec 2002 19:53:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB20rUv09935
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 1 Dec 2002 19:53:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA19113
	for <ieprep-web-archive@ietf.org>; Sun, 1 Dec 2002 19:50:33 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB20n6v09726;
	Sun, 1 Dec 2002 19:49:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB20mqv09706
	for <ieprep@optimus.ietf.org>; Sun, 1 Dec 2002 19:48:52 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA18998
	for <ieprep@ietf.org>; Sun, 1 Dec 2002 19:45:55 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.78])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 4C74A7A0A; Sun,  1 Dec 2002 16:48:38 -0800 (PST)
Date: Sun, 1 Dec 2002 19:47:44 -0500
Subject: Re: [Ieprep] alternate model for edge QoS
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Henning Schulzrinne <hgs@cs.columbia.edu>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <3DEA7EE0.8090501@cs.columbia.edu>
Message-Id: <AB4D4F82-058F-11D7-9B56-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Sunday, Dec 1, 2002, at 16:28 America/Montreal, Henning Schulzrinne 
wrote:
> I just looked, more or less randomly, at the Juniper ERX edge router. 
> It claims the ability to support 200,000 filters.

Probably it does, for the set of filters that are pre-defined in their
ASIC.  That doesn't say anything about the other state that's needed.
And off-hand it doesn't say whether or not any particular vendor's 
existing
ASIC filters can do the QoS filtering that has been discussed here.
No doubt some do and some don't.

Ironically, common layer-3 switches (from multiple vendors) do in fact 
have
an internal flow-cache architecture that lends itself to offering QoS
with less pain from state.  Common backbone routers (from multiple 
vendors)
use a very different internal longest-prefix-match architecture that has
more pain when one has to deal with flow state.  The LPM approach scales
better than a flow-cache approach when inside a backbone router.  In a
typical layer-3 switch application (LAN/MAN, edge, access), that 
dimension
of scalability is not as much of an issue.

> The numbers I cited did not use any accelerator (the paper also 
> contains
> numbers for those, which are indeed much higher).

And most uses of TLS aren't related to what you are proposing
to make routers perform.

>>     Right now, zero CA root certs are in any normally deployed control
>> processor, so you're assuming a bunch of stuff not in any mainstream 
>> deployed
>> router today.
>
> Please recall that the discussion started with my assumption of adding 
> NSIS functionality, which is likely to require both IPsec and possibly 
> TLS, depending on how the design shapes up. Clearly, this is not 
> something that you can plug into today's router. The point of the 
> discussion was to see what, if any, requirements might be helpful for 
> protocols being developed elsewhere in the IETF.

I think a very practical requirement is that the new proposal NOT 
create easily
exploited DOS or DDOS attack vectors.  So far, the proposals being 
discussed
all create such new attack vectors.

>> > Obviously, this happens during non-emergency conditions and can be 
>> as
>> > simple as a signed S/MIME mail. There's no need for secrecy, just
>> > integrity, since this is a public key only. Alternatively, the
>> > operator can just download the .p12 file from the FEMA web server.
>> > There may even be automated ways of doing this; I haven't followed 
>> all
>> > the recent PKI-related WGs. This operation is repeated at cert
>> > lifetime intervals; these seem to be roughly yearly.
>>
>>
>>     So the ISP now has to manage certificates in routers, make sure 
>> the
>> certificate
>> stays in the router even if the CP gets swapped out due to a fault
>> (faults of that
>> sort happened to me many a time on reasonably modern backbone 
>> routers).
>>  Increases
>> the operational complexity and operational costs pretty significantly.
>
> Again, the worse case scenario is that the cert is gone when the 
> earthquake hits. Means you get no-preference service. I suspect many 
> routers already support ssh, so they already deal with PK management, 
> so this is hardly earth-shattering complexity.

No, the worst case is that the router falls over and dies because it 
gets DDOS
attacked -- or the network availability declines visibly due to the 
additional
complexity -- or other consequent operational problems make the network 
brittle
for a myriad of reasons.

Oh, and SSH doesn't really do certificate management today.  Also, the 
operators
apply filters to any deployed SSH (generally these drop all SSH packets 
that don't
claim to originate on the NOC subnet or from a small set of authorised 
hosts;
and filters throughout that domain's network protect against spoofing 
that subnet).
That protection/risk-management approach would not work for any 
customer-initiated
preferred service requests, because the emergency could happen anywhere 
and one
can't know in advance where the IEprep users will happen to jack in.

>>     It is an expensive computation to verify a signature.  An obvious
>> DDOS attack is
>> to flood the router with forged resource requests and force the router
>> to spend
>> lots of time performing a bunch of expensive computations.  Bad idea 
>> for
>> one to enable
>> such router validation if one wants the router to keep BGP in sync,
>> forward packets,
>> and do sundry other stuff needed to keep the Internet up and running.
>
> As noted, low-end PCs can do hundreds of such verifications a second. 
> This is more than enough for the purposes of what's needed here.

An attacker can trivially throw thousands of false packets at a router 
per second,
each of which forces the router to perform the signature verification, 
so your
math does not support the idea that it can trivially be done in a 
mainstream
CPU.  Instead, you are creating a new (and easily exploited) 
Distributed DOS
attack vulnerability for ISPs.  My guess is that they won't decide to 
deploy
that (and roll said as much in Atlanta).

Of course, inside a single private network a bunch of things might be 
useful
that aren't inside a multi-domain network.  My comments are focused on 
the
issues around inter-domain QoS, not on deployment inside a large private
network.  DoD's own IP networks (e.g. NIPRnet) are examples of large 
private
networks, of course.  So this is an area where what DoD IP networks do
and what GETS-like IP networks do will likely vary.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  1 20:52:45 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA19934
	for <ieprep-archive@odin.ietf.org>; Sun, 1 Dec 2002 20:52:45 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB21tBw12893
	for ieprep-archive@odin.ietf.org; Sun, 1 Dec 2002 20:55:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB21tBv12890
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 1 Dec 2002 20:55:11 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA19931
	for <ieprep-web-archive@ietf.org>; Sun, 1 Dec 2002 20:52:14 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB21p5v12783;
	Sun, 1 Dec 2002 20:51:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB21oPv12760
	for <ieprep@optimus.ietf.org>; Sun, 1 Dec 2002 20:50:25 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA19859
	for <ieprep@ietf.org>; Sun, 1 Dec 2002 20:47:28 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB21hZo14894;
	Sun, 1 Dec 2002 17:43:35 -0800 (PST)
Date: Sun, 1 Dec 2002 17:43:35 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: Carl Ford <carl@ietfwatch.net>
cc: Mpierce1@aol.com, <rja@extremenetworks.com>, <ieprep@ietf.org>
Subject: Re: [Ieprep] Re: Looking for governmental feedback as well
In-Reply-To: <200212011631.gB1GVPA19487@samsara.kgbinternet.com>
Message-ID: <Pine.GSO.4.44.0212011432170.13968-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Sun, 1 Dec 2002, Carl Ford wrote:
    > So the question that we have to ask for ieprep is do we have a goal of greater
    > redundancy for the Internet overall, greater redundancy for the government sites
    > specifically, or a specific practise on a government access methods.

Well, this is getting kind of reflexive...  Should the government require
the IETF to require the government to require itself to buy certain
services that it's not currently buying?  :-)  I'd say no, that's no place
for the IETF.

I guess the good news, from the point of view of anyone looking to dirve
increased redundancy at the unreliable edges, is that there's been that
general trend at work in the Internet already for the past two years.
It's been characterized as something of a downward spiral, since what
typically happens is a customer hears that their service might be
interrupted by an outage, and goes and buys some redundant service, but
their total budget for service hasn't really doubled, so they wind up
both buying a new service which is (on average) half as expensive as the
old one, and then replacing the old one with a different one that's half
as expensive.  The problem is, it's nearly twice as expensive for the two
new providers to deliver two bad services as for the old provider to
deliver one good one, so the industry winds up with less money to actually
solve problems and provide decent service.  TANSTAAFL.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  1 22:45:40 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA22326
	for <ieprep-archive@odin.ietf.org>; Sun, 1 Dec 2002 22:45:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB23m5q18823
	for ieprep-archive@odin.ietf.org; Sun, 1 Dec 2002 22:48:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB23m5v18820
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 1 Dec 2002 22:48:05 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA22308
	for <ieprep-web-archive@ietf.org>; Sun, 1 Dec 2002 22:45:08 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB23i3v18646;
	Sun, 1 Dec 2002 22:44:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB23h8v18592
	for <ieprep@optimus.ietf.org>; Sun, 1 Dec 2002 22:43:08 -0500
Received: from mtiwmhc13.worldnet.att.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA22124
	for <ieprep@ietf.org>; Sun, 1 Dec 2002 22:40:11 -0500 (EST)
Received: from cs.columbia.edu ([12.84.248.201])
          by mtiwmhc13.worldnet.att.net
          (InterMail vM.5.01.05.12 201-253-122-126-112-20020820) with ESMTP
          id <20021202034256.SMRR24110.mtiwmhc13.worldnet.att.net@cs.columbia.edu>;
          Mon, 2 Dec 2002 03:42:56 +0000
Message-ID: <3DEAD63E.7050902@cs.columbia.edu>
Date: Sun, 01 Dec 2002 22:40:46 -0500
From: Henning Schulzrinne <hgs@cs.columbia.edu>
Organization: Columbia University
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.1) Gecko/20020826
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: RJ Atkinson <rja@extremenetworks.com>
CC: ieprep <ieprep@ietf.org>
Subject: Re: [Ieprep] alternate model for edge QoS
References: <AB4D4F82-058F-11D7-9B56-00039357A82A@extremenetworks.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

On DDOS, it's a bit more subtle than throwing packets at a router: TLS 
(and others) make sure that only attackers with non-forged IP addresses 
even get checked. That way, you'd have to compromise hundreds of 
distinct hosts to even get one second worth of DOS.

RJ Atkinson wrote:
> 

> 
> And most uses of TLS aren't related to what you are proposing
> to make routers perform.

The crypto operations are similar, so the performance is likely to be in 
the same ballpark.

> 
>>>     Right now, zero CA root certs are in any normally deployed control
>>> processor, so you're assuming a bunch of stuff not in any mainstream 
>>> deployed
>>> router today.
>>
>>
>> Please recall that the discussion started with my assumption of adding 
>> NSIS functionality, which is likely to require both IPsec and possibly 
>> TLS, depending on how the design shapes up. Clearly, this is not 
>> something that you can plug into today's router. The point of the 
>> discussion was to see what, if any, requirements might be helpful for 
>> protocols being developed elsewhere in the IETF.
> 
> 
> I think a very practical requirement is that the new proposal NOT create 
> easily
> exploited DOS or DDOS attack vectors.  So far, the proposals being 
> discussed
> all create such new attack vectors.
> 
>>> > Obviously, this happens during non-emergency conditions and can be as
>>> > simple as a signed S/MIME mail. There's no need for secrecy, just
>>> > integrity, since this is a public key only. Alternatively, the
>>> > operator can just download the .p12 file from the FEMA web server.
>>> > There may even be automated ways of doing this; I haven't followed all
>>> > the recent PKI-related WGs. This operation is repeated at cert
>>> > lifetime intervals; these seem to be roughly yearly.
>>>
>>>
>>>     So the ISP now has to manage certificates in routers, make sure the
>>> certificate
>>> stays in the router even if the CP gets swapped out due to a fault
>>> (faults of that
>>> sort happened to me many a time on reasonably modern backbone routers).
>>>  Increases
>>> the operational complexity and operational costs pretty significantly.
>>
>>
>> Again, the worse case scenario is that the cert is gone when the 
>> earthquake hits. Means you get no-preference service. I suspect many 
>> routers already support ssh, so they already deal with PK management, 
>> so this is hardly earth-shattering complexity.
> 
> 
> No, the worst case is that the router falls over and dies because it 
> gets DDOS
> attacked -- or the network availability declines visibly due to the 
> additional
> complexity -- or other consequent operational problems make the network 
> brittle
> for a myriad of reasons.
> 
> Oh, and SSH doesn't really do certificate management today.  Also, the 
> operators
> apply filters to any deployed SSH (generally these drop all SSH packets 
> that don't
> claim to originate on the NOC subnet or from a small set of authorised 
> hosts;
> and filters throughout that domain's network protect against spoofing 
> that subnet).
> That protection/risk-management approach would not work for any 
> customer-initiated
> preferred service requests, because the emergency could happen anywhere 
> and one
> can't know in advance where the IEprep users will happen to jack in.
> 
>>>     It is an expensive computation to verify a signature.  An obvious
>>> DDOS attack is
>>> to flood the router with forged resource requests and force the router
>>> to spend
>>> lots of time performing a bunch of expensive computations.  Bad idea for
>>> one to enable
>>> such router validation if one wants the router to keep BGP in sync,
>>> forward packets,
>>> and do sundry other stuff needed to keep the Internet up and running.
>>
>>
>> As noted, low-end PCs can do hundreds of such verifications a second. 
>> This is more than enough for the purposes of what's needed here.
> 
> 
> An attacker can trivially throw thousands of false packets at a router 
> per second,
> each of which forces the router to perform the signature verification, 
> so your
> math does not support the idea that it can trivially be done in a 
> mainstream
> CPU.  Instead, you are creating a new (and easily exploited) Distributed 
> DOS
> attack vulnerability for ISPs.  My guess is that they won't decide to 
> deploy
> that (and roll said as much in Atlanta).
> 
> Of course, inside a single private network a bunch of things might be 
> useful
> that aren't inside a multi-domain network.  My comments are focused on the
> issues around inter-domain QoS, not on deployment inside a large private
> network.  DoD's own IP networks (e.g. NIPRnet) are examples of large 
> private
> networks, of course.  So this is an area where what DoD IP networks do
> and what GETS-like IP networks do will likely vary.
> 
> Ran
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 08:06:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA12752
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 08:06:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2D8p528691
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 08:08:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2D8pv28688
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 08:08:51 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA12739
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 08:06:00 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2D89v28651;
	Mon, 2 Dec 2002 08:08:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2D7kv28627
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 08:07:46 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA12698
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 08:04:55 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 2 Dec 2002 08:07:16 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120208073202722
 ; Mon, 02 Dec 2002 08:07:32 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <XNT4B28Z>; Mon, 2 Dec 2002 08:07:19 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E779C9@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson '" <rja@extremenetworks.com>,
        "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>
Cc: "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] a practical VoIP issue & context
Date: Mon, 2 Dec 2002 08:07:32 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Ran said on firewalls:
>For pre-planned VoIP deployments, this can be worked through by
>getting policies changed, upgrading firewall capabilities, 
>and so forth.

Yes, so we have identified that in context such as
"1.	N ETS locations with access links into the same ISP where the ISP
offers CoS. Note that the particular ISP, access link bandwidth, and local
configurations can be determined in advance of the emergency."
that the problem can be solved.

However, 
Ran also said:
>For IEprep use of VoIP, the reality is that most access links have a 
>firewall on the customer end and that firewall is likely to be pre-
>configured in a manner that blocks IEprep use of VoIP.

So, if the problem to be solved is more in a context of 
"2.	Emergency Worker accesses Internet via arbitrary connection (e.g.,
Internet Cafe).  Note that the ISP providing connectivity to the Internet
cannot be determined in advance." then there may be problems with voice
media streams being blocked or VoIP broken due to NAT.

Part of the impediments to ieprep progress (and this not directed at Ran) is
that all problems in all contexts and all requirements are discussed at the
same time.  However, it is more productive to define a particular problem,
the context, and try to find some effective reasonable way to approach it.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 09:44:26 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA17248
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 09:44:26 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2Ekmg03070
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 09:46:48 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Ekmv03067
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 09:46:48 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA17215
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 09:43:55 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Ek3v03040;
	Mon, 2 Dec 2002 09:46:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Ejev03006
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 09:45:40 -0500
Received: from cs.columbia.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA17119
	for <IEPREP@ietf.org>; Mon, 2 Dec 2002 09:42:47 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB2Ejal8009230
	(version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NOT)
	for <IEPREP@ietf.org>; Mon, 2 Dec 2002 09:45:37 -0500 (EST)
Received: from cs.columbia.edu (tallgrass.netlab.uky.edu [204.198.76.66])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB2EjZdG026704
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT)
	for <IEPREP@ietf.org>; Mon, 2 Dec 2002 09:45:36 -0500 (EST)
Message-ID: <3DEB7201.10509@cs.columbia.edu>
Date: Mon, 02 Dec 2002 09:45:21 -0500
From: Henning Schulzrinne <hgs@cs.columbia.edu>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.2b) Gecko/20021016
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: IEPREP <IEPREP@ietf.org>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] SIP requirements draft after WG last call
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Taking into account comments received during WG last call, I will be 
submitting a new draft to the I-D editor today. Until it appears in the 
I-D archives, you can find it at 
http://www.cs.columbia.edu/sip/drafts/draft-ietf-ieprep-sip-reqs-02.txt

There were no major changes. I aligned the topology section with James 
Polk's draft, fixed a few editorial problems and moved sections around, 
so that the requirements now follow the discussion sections. I also 
added a brief Security Considerations section that mostly points to 
Section 10.

If you find any last-minute glitches, please let me know.

Henning

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 10:04:38 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA18130
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 10:04:38 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2F6xp03985
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 10:06:59 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2F6xv03982
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 10:06:59 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA18107
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 10:04:07 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2F63v03947;
	Mon, 2 Dec 2002 10:06:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2F5Dv03883
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 10:05:13 -0500
Received: from server07.icaen.uiowa.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA18024
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 10:02:20 -0500 (EST)
Received: from server11.icaen.uiowa.edu (server11.icaen.uiowa.edu [128.255.17.51]) by server07.icaen.uiowa.edu (8.9.3/8.9.3) with ESMTP id JAA04901 for <ieprep@ietf.org> sent by <gveerara@engineering.uiowa.edu>; Mon, 2 Dec 2002 09:05:08 -0600 (CST)
Received: from webmail.engineering.uiowa.edu (server50.engr.uiowa.edu [128.255.18.28])
	by server11.icaen.uiowa.edu (8.9.3/smtp-service-1.6) with ESMTP id JAA26007
	for <ieprep@ietf.org>;  sent by <gveerara@engineering.uiowa.edu>; Mon, 2 Dec 2002 09:05:06 -0600 (CST)
X-WebMail-UserID:  gveerara
Date: Mon, 2 Dec 2002 09:05:05 -0600
From: gveerara <gveerara@engineering.uiowa.edu>
To: ieprep@ietf.org
X-EXP32-SerialNo: 00003240
Message-ID: <3DD4DDB4@webmail.engineering.uiowa.edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="ISO-8859-1"
Content-Transfer-Encoding: 7bit
X-Mailer: WebMail (Hydra) SMTP v3.61.05
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] Ddos
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Hi ,
I am doing a term paper for one of my courses computer networks on  dDos and 
ways to counter it.Could some one please tell me where i could find papers or 
other sources which have info on dDos.I am a little lost  here and I have just
 a week to get it done .I will be really thankful if u could give me some 
suggestions.
Thanks,
Govi.

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 10:17:35 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA18726
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 10:17:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2FJts05252
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 10:19:55 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FJtv05249
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 10:19:55 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA18719
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 10:17:04 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FJ1v05191;
	Mon, 2 Dec 2002 10:19:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FIBv05152
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 10:18:11 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA18615
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 10:15:20 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id E04B17A0A; Mon,  2 Dec 2002 07:18:04 -0800 (PST)
Date: Mon, 2 Dec 2002 10:18:05 -0500
Subject: Re: [Ieprep] alternate model for edge QoS
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Henning Schulzrinne <hgs@cs.columbia.edu>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <3DEAD63E.7050902@cs.columbia.edu>
Message-Id: <41C688FE-0609-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Sunday, Dec 1, 2002, at 22:40 America/Montreal, Henning Schulzrinne 
wrote:
> On DDOS, it's a bit more subtle than throwing packets at a router: TLS 
> (and others) make sure that only attackers with non-forged IP 
> addresses even get checked. That way, you'd have to compromise 
> hundreds of distinct hosts to even get one second worth of DOS.

	Today, it is not possible to put in source address filtering on the 
largest
pipes inter-connecting two (or more) ISPs.  A lot of deployed router 
hardware can't
handle it.  Even if the router hardware can, it is intractably 
complicated to create
and correctly maintain (in real-time) the right source address prefix 
filter set.
Consider the issues faced by someone trying to source address filter an 
OC-192
connecting to one of the top 3 providers, each of which has meaningful 
amounts
of daily change in the set of prefixes that is legitimately connected 
to it.  This
is one of several reasons why the inter-domain QoS situation is 
fundamentally
different from intra-domain QoS.  Intra-domain QoS is not uncommon 
these days
using existing mechanisms (e.g. DiffServ, 802.1p, and so forth).

	Since there is no way to source address filter all inbound links,
there is no way for a backbone router to sort out the non-forged IP 
addresses
from the forgeries.

	And at @Home, our experience was that a single province in Canada (or 
any
large US state where we operated or any Australian state that we 
operated in)
typically had thousands of compromised hosts at any point in time.  
Those compromises
were not something that @Home (or any other ISP) could prevent.  Most 
compromises
were due to the home user not applying OS patches (that were in fact 
already available)
from Microsoft.

	Therefore I have no problem in considering thousands of zombie hosts 
being
available across the country/globe to launch a coordinated DDOS attack 
on an ISP.
It has happened numerous times in the past in the US and overseas.  
There is no reason
to believe that it won't happen in future (or is less likely in 
future).  @Home applied
(and the better of the cable network operators apply) source address 
filtering at the edge
so that IP address forgeries are minimised (not zero, but minimised), 
but lack of IP
address forgeries doesn't help the victim when a large scale DDOS 
attack is in progress.

> RJ Atkinson wrote:
>
>> And most uses of TLS aren't related to what you are proposing
>> to make routers perform.
>
> The crypto operations are similar, so the performance is likely to be 
> in the same ballpark.

	I dispute that the crypto operations are so similar.  Most TLS 
operations
are using symmetric ciphers (e.g. DES, AES), which are computationally
less expensive than for asymmetric ciphers (e.g. RSA).  Only asymmetric 
ciphers
are normally used for signing/verifying.  One could imagine a 
symmetric-only
situation using a Needham-Schroeder KDC (e.g. Kerberos V) to reduce the 
computational
burden, but that would still require a substantial deployment effort of 
its own.

	Adding complexity to any network tends to lead to reduced availability,
increased probability of a fault occurring (because more things can go 
wrong),
and longer mean-time-to-repair (MTTR).  It is important that an 
operator understand
all these tradeoffs before they deploy anything that increases the 
complexity
of their systems and network.  So far, the operators on this list are 
concluding
that the proposed inter-domain QoS would add so much complexity that it 
would
cause much more harm than good.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 10:26:35 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19263
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 10:26:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2FStq05879
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 10:28:55 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FStv05876
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 10:28:55 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19202
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 10:26:04 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FS5v05782;
	Mon, 2 Dec 2002 10:28:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FRIv05736
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 10:27:18 -0500
Received: from imo-d04.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19091
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 10:24:27 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d04.mx.aol.com (mail_out_v34.13.) id j.b8.329cb1e2 (657);
	Mon, 2 Dec 2002 10:26:50 -0500 (EST)
Message-ID: <b8.329cb1e2.2b1cd5b9@aol.com>
Date: Mon, 2 Dec 2002 10:26:49 EST
Subject: Re: [Ieprep] Accounting
To: rja@extremenetworks.com, K.Carlberg@cs.ucl.ac.uk
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_b8.329cb1e2.2b1cd5b9_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_b8.329cb1e2.2b1cd5b9_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/1/2002 3:38:27 PM Eastern Standard Time, 
rja@extremenetworks.com writes:


>    And I checked with some ISP folks in Japan. They have no requirement
> for accounting as part of their *deployed* IEprep service.  In fact,
> they claim that their *deployed* IEprep service is identical (*no changes*)
> to their deployed regular commercial IP service.
> 
>     Its fascinating that the several countries with a deployed IEprep 
> service
> "require" so little, but the country with the big paid staff and zero IEprep
> deployment "requires" so much.
> 



[MAP] Perhaps you could elaborate. We've had a lack of info about "deployed 
IEPREP services" in other countries (besides the US), either in the 
circuit-switched world or IP. As far as I know, GETS in the US and MLPP are 
the only ones today in the circuit switched world and there are none in IP. 
The IAA in Japan is something completely different, as is the sevice in UK 
that was described as a way to get priority dial tone.

What specific service in Japan are you referring to? Your statement that 
their "deployed IEPREP service is identical to their deployed regular 
commercial IP service" sounds like it means that they have no actual IEPREP 
service.

Mike Pierce


--part1_b8.329cb1e2.2b1cd5b9_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/1/2002 3:38:27 PM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px"> &nbsp;&nbsp;And I checked with some ISP folks in Japan. They have no requirement
<BR>for accounting as part of their *deployed* IEprep service. &nbsp;In fact,
<BR>they claim that their *deployed* IEprep service is identical (*no changes*)
<BR>to their deployed regular commercial IP service.
<BR>
<BR> &nbsp;&nbsp;&nbsp;Its fascinating that the several countries with a deployed IEprep service
<BR>"require" so little, but the country with the big paid staff and zero IEprep
<BR>deployment "requires" so much.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>
<BR>[MAP] Perhaps you could elaborate. We've had a lack of info about "deployed IEPREP services" in other countries (besides the US), either in the circuit-switched world or IP. As far as I know, GETS in the US and MLPP are the only ones today in the circuit switched world and there are none in IP. The IAA in Japan is something completely different, as is the sevice in UK that was described as a way to get priority dial tone.
<BR>
<BR>What specific service in Japan are you referring to? Your statement that their "deployed IEPREP service is identical to their deployed regular commercial IP service" sounds like it means that they have no actual IEPREP service.
<BR>
<BR>Mike Pierce
<BR></FONT></HTML>

--part1_b8.329cb1e2.2b1cd5b9_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 10:26:38 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19279
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 10:26:37 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2FSwS05895
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 10:28:58 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FSvv05892
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 10:28:57 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19206
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 10:26:06 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FS9v05807;
	Mon, 2 Dec 2002 10:28:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FRQv05748
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 10:27:26 -0500
Received: from imo-r10.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19109
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 10:24:35 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r10.mx.aol.com (mail_out_v34.13.) id h.1c6.26b575b (657);
	Mon, 2 Dec 2002 10:27:08 -0500 (EST)
Message-ID: <1c6.26b575b.2b1cd5cb@aol.com>
Date: Mon, 2 Dec 2002 10:27:07 EST
Subject: Re: [Ieprep] Re: Looking for governmental feedback as well
To: carl@ietfwatch.net, woody@pch.net, rja@extremenetworks.com,
        ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_1c6.26b575b.2b1cd5cb_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_1c6.26b575b.2b1cd5cb_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/1/2002 11:30:20 AM Eastern Standard Time, 
carl@ietfwatch.net writes:


> So the question that we have to ask for ieprep is do we have a goal of 
> greater 
> redundancy for the Internet overall, greater redundancy for the government 
> sites 
> specifically, or a specific practise on a government access methods.
> 


[MAP] It's an interesting question or idea that IEPREP could write a series 
of BCPs advising people how to engineer their networks for maximum 
survivablity (i.e., through redundancy, route diversion, hardened sites, 
etc), but I don't believe that is what this WG was intended for.

This WG needs to define how to deal with the situation which might occur in a 
disaster when, in spite of all the above things being done, there is still 
congestion.

Mike Pierce
Artel


--part1_1c6.26b575b.2b1cd5cb_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/1/2002 11:30:20 AM Eastern Standard Time, carl@ietfwatch.net writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">So the question that we have to ask for ieprep is do we have a goal of greater 
<BR>redundancy for the Internet overall, greater redundancy for the government sites 
<BR>specifically, or a specific practise on a government access methods.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>[MAP] It's an interesting question or idea that IEPREP could write a series of BCPs advising people how to engineer their networks for maximum survivablity (i.e., through redundancy, route diversion, hardened sites, etc), but I don't believe that is what this WG was intended for.
<BR>
<BR>This WG needs to define how to deal with the situation which might occur in a disaster when, in spite of all the above things being done, there is still congestion.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_1c6.26b575b.2b1cd5cb_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 10:27:34 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19356
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 10:27:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2FTst05986
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 10:29:54 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FTsv05983
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 10:29:54 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19313
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 10:27:02 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FT2v05941;
	Mon, 2 Dec 2002 10:29:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FSXv05859
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 10:28:33 -0500
Received: from imo-d05.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19175
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 10:25:41 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d05.mx.aol.com (mail_out_v34.13.) id j.146.3f435a1 (657);
	Mon, 2 Dec 2002 10:28:20 -0500 (EST)
Message-ID: <146.3f435a1.2b1cd614@aol.com>
Date: Mon, 2 Dec 2002 10:28:20 EST
Subject: Re: [Ieprep] problem context
To: rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_146.3f435a1.2b1cd614_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_146.3f435a1.2b1cd614_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 11/29/2002 6:27:53 PM Eastern Standard Time, 
rja@extremenetworks.com writes:


> On Friday, Nov 29, 2002, at 13:09 America/Montreal, Mpierce1@aol.com 
> wrote:
> > [MAP] Yes, we're aware that the Internet is unaware of voice calls. 
> > That is why we have proposed that some form of "packet marking" will 
> > be needed to achieve the preferential treament goals. Isn't that 
> > exactly what AF and EF did?
> 
>     [RJA] Neither AF nor EF is widely deployed.  


[MAP] Whether or not something is currently widely deployed is not the issue. 
We're talking about whether or not the necessary mechanisms are 
"standardized". Our goal, as Don explained, is to get commercial standards 
(IETF or others) which DoD (and others) can then reference to ask for (pay 
for) things to be deployed.

------------------------------------------------------------------------------

----------
> 
>   [RJA]  So if DoD wants to use DiffServ *inside* DoD, I think that is 
> practical (and
> think I've said so in the past).  It is the deployment of inter-domain QoS
> (of any sort, including DiffServ) that seems intractible right now.  One of
> 


[MAP] Then please let on get on with the "standardization" work so that what 
we need can be defined.

------------------------------------------------------------------------------

--------
> >
> > [MAP] As I believe we have made clear, overprovisioning is not an 
> > acceptable solution for the emergency/disaster situations that we are 
> > addressing. Could we eliminate that from any further discussion?
> 
>   [RJA]  We can't eliminate that until the folks who want to eliminate that 
> can
> deliver a persuasive argument that it can't work (and explaining why
> the past scenarios where it DID work aren't applicable is probably
> a part of that).
> 
>     So far, I keep hearing NCS/DISA-funded folks say that "overprovisioning 
> is not
> acceptable" but I haven't heard any technical reasons why that is the case.
> 


[MAP] You misinterpreted what I (and probably others) have said. I did not 
say that "overprovisioning is not acceptable" (your quote) or that we need to 
eliminate overprovisioning. Any carrier or network operator can overprovision 
if they want. I said that "overprovisioning is not an acceptable solution" to 
the problem that the IEPREP WG was formed to solve.

------------------------------------------------------------------------------

------------
> >
> > [MAP] But the bottom line, after all the existing solutions are laid on 
> the 
> > table, is that none have been able to explain how priority calls get 
> > the required treatment over regular calls on IP-based networks in a 
> > situation of congestion using existing standards.
> 
> [RJA]    And I thought that was NOT Don's stated objective.  His objective 
> was that the
> preferred calls were able to get through, which is one that the operators 
> here seem
> to think can be achieved (through some combination of over-provisioning and 
> these other
> mechanisms I discussed earlier).  Maybe I mis-read his objective ?
> 

[MAP] I believe it is exactly the objective, stated in slightly different 
words.

------------------------------------------------------------------------------

------------

[RJA]
>     The DoD tactical nets are a huge pain, due to lack of bandwidth/capacity,
> but DoD operates all the equipment on those nets and so can deploy 
> link-specific
> solutions (e.g. router-queuing, link-layer QoS of whatever is available, 
> others)
> 


[MAP] So back to the need for "commercial" standards to solve that "pain".

Mike Pierce
Artel



--part1_146.3f435a1.2b1cd614_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 11/29/2002 6:27:53 PM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">On Friday, Nov 29, 2002, at 13:09 America/Montreal, Mpierce1@aol.com 
<BR>wrote:
<BR>&gt; [MAP] Yes, we're aware that the Internet is unaware of voice calls. 
<BR>&gt; That is why we have proposed that some form of "packet marking" will 
<BR>&gt; be needed to achieve the preferential treament goals. Isn't that 
<BR>&gt; exactly what AF and EF did?
<BR>
<BR> &nbsp;&nbsp;&nbsp;[RJA] Neither AF nor EF is widely deployed. &nbsp;Each has some deployment. </BLOCKQUOTE>
<BR>
<BR>
<BR>[MAP] Whether or not something is currently widely deployed is not the issue. We're talking about whether or not the necessary mechanisms are "standardized". Our goal, as Don explained, is to get commercial standards (IETF or others) which DoD (and others) can then reference to ask for (pay for) things to be deployed.
<BR>
<BR>----------------------------------------------------------------------------------------
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR> &nbsp;[RJA] &nbsp;So if DoD wants to use DiffServ *inside* DoD, I think that is practical (and
<BR>think I've said so in the past). &nbsp;It is the deployment of inter-domain QoS
<BR>(of any sort, including DiffServ) that seems intractible right now. &nbsp;One of
<BR>the IAB documents (RFC-2990 ??) talks about this as well.</BLOCKQUOTE>
<BR>
<BR>
<BR>[MAP] Then please let on get on with the "standardization" work so that what we need can be defined.
<BR>
<BR>--------------------------------------------------------------------------------------
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">&gt;
<BR>&gt; [MAP] As I believe we have made clear, overprovisioning is not an 
<BR>&gt; acceptable solution for the emergency/disaster situations that we are 
<BR>&gt; addressing. Could we eliminate that from any further discussion?
<BR>
<BR> &nbsp;[RJA] &nbsp;We can't eliminate that until the folks who want to eliminate that can
<BR>deliver a persuasive argument that it can't work (and explaining why
<BR>the past scenarios where it DID work aren't applicable is probably
<BR>a part of that).
<BR>
<BR> &nbsp;&nbsp;&nbsp;So far, I keep hearing NCS/DISA-funded folks say that "overprovisioning is not
<BR>acceptable" but I haven't heard any technical reasons why that is the case.
<BR>To my ears, it just sounds like policy rather than engineering.</BLOCKQUOTE>
<BR>
<BR>
<BR>[MAP] You misinterpreted what I (and probably others) have said. I did not say that "overprovisioning is not acceptable" (your quote) or that we need to eliminate overprovisioning. Any carrier or network operator can overprovision if they want. I said that "overprovisioning is not an acceptable solution" to the problem that the IEPREP WG was formed to solve.
<BR>
<BR>------------------------------------------------------------------------------------------
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">&gt;
<BR>&gt; [MAP] But the bottom line, after all the existing solutions are laid on the 
<BR>&gt; table, is that none have been able to explain how priority calls get 
<BR>&gt; the required treatment over regular calls on IP-based networks in a 
<BR>&gt; situation of congestion using existing standards.
<BR>
<BR>[RJA] &nbsp;&nbsp;&nbsp;And I thought that was NOT Don's stated objective. &nbsp;His objective was that the
<BR>preferred calls were able to get through, which is one that the operators here seem
<BR>to think can be achieved (through some combination of over-provisioning and these other
<BR>mechanisms I discussed earlier). &nbsp;Maybe I mis-read his objective ?
<BR></BLOCKQUOTE>
<BR>
<BR>[MAP] I believe it is exactly the objective, stated in slightly different words.
<BR>
<BR>------------------------------------------------------------------------------------------
<BR>
<BR>[RJA]
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px"> &nbsp;&nbsp;&nbsp;The DoD tactical nets are a huge pain, due to lack of bandwidth/capacity,
<BR>but DoD operates all the equipment on those nets and so can deploy link-specific
<BR>solutions (e.g. router-queuing, link-layer QoS of whatever is available, others)
<BR>on those networks.</BLOCKQUOTE>
<BR>
<BR>
<BR>[MAP] So back to the need for "commercial" standards to solve that "pain".
<BR>
<BR>Mike Pierce
<BR>Artel
<BR>
<BR></FONT></HTML>

--part1_146.3f435a1.2b1cd614_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 10:29:40 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19501
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 10:29:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2FW0V06229
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 10:32:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FW0v06226
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 10:32:00 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19483
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 10:29:08 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FV8v06168;
	Mon, 2 Dec 2002 10:31:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FUvv06067
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 10:30:57 -0500
Received: from imo-r04.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19411
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 10:28:06 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r04.mx.aol.com (mail_out_v34.13.) id 7.19e.cd816ec (657);
	Mon, 2 Dec 2002 10:30:43 -0500 (EST)
Message-ID: <19e.cd816ec.2b1cd6a3@aol.com>
Date: Mon, 2 Dec 2002 10:30:43 EST
Subject: Re: [Ieprep] alternate model for edge QoS
To: hgs@cs.columbia.edu, rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_19e.cd816ec.2b1cd6a3_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_19e.cd816ec.2b1cd6a3_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 11/30/2002 9:20:47 AM Eastern Standard Time, 
hgs@cs.columbia.edu writes:


> > That was not the problem with RSVP deployments and is not the problem with 
> RSVP MD5.
> > The problem with RSVP deployments included at least (1) too much state in 
> routers,
> 
> This might have been a problem a few years ago, but I'm having a hard 
> time believing that a router that can handle on the order of 100,000 
> route entries has trouble with, say, a few hundred reservation entries. 
> As was pointed out by Fred, there are large MPLS path counts in 
> operation today, with no noticeable state problems. (Not counting the 
> pain of configuration.)
> 
> To be more precise, we have to distinguish three types of states:
> 
> (1) scheduling state; this only matters if you have per-flow queueing
> (2) filtering/policing state, which polices flows;
> (3) reservation state, one for each reservation, but only affecting the 
> control processor
> 



[MAP] Thanks Henning.

I think it's time to end this circular argument that:

1. States are bad.
2. Any notion of "connections" requires "states".
3. Any notion of "connections" is bad (I regard RSVP as a way to make IP look 
"connection oriented")
4. IP is connectionless, therefore any notion of "connections" destroys its 
beauty.
5. Any notion of "state" destroys its beauty.

As you've pointed out, there are different kinds of state. Telephone switches 
have been designed which maintain "state" for 10's of thousands of calls, and 
that was before memory cost only hundreds of dollars per Gigabyte.

Mike Pierce
Artel


--part1_19e.cd816ec.2b1cd6a3_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 11/30/2002 9:20:47 AM Eastern Standard Time, hgs@cs.columbia.edu writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">&gt; That was not the problem with RSVP deployments and is not the problem with RSVP MD5.
<BR>&gt; The problem with RSVP deployments included at least (1) too much state in routers,
<BR>
<BR>This might have been a problem a few years ago, but I'm having a hard 
<BR>time believing that a router that can handle on the order of 100,000 
<BR>route entries has trouble with, say, a few hundred reservation entries. 
<BR>As was pointed out by Fred, there are large MPLS path counts in 
<BR>operation today, with no noticeable state problems. (Not counting the 
<BR>pain of configuration.)
<BR>
<BR>To be more precise, we have to distinguish three types of states:
<BR>
<BR>(1) scheduling state; this only matters if you have per-flow queueing
<BR>(2) filtering/policing state, which polices flows;
<BR>(3) reservation state, one for each reservation, but only affecting the 
<BR>control processor
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>
<BR>[MAP] Thanks Henning.
<BR>
<BR>I think it's time to end this circular argument that:
<BR>
<BR>1. States are bad.
<BR>2. Any notion of "connections" requires "states".
<BR>3. Any notion of "connections" is bad (I regard RSVP as a way to make IP look "connection oriented")
<BR>4. IP is connectionless, therefore any notion of "connections" destroys its beauty.
<BR>5. Any notion of "state" destroys its beauty.
<BR>
<BR>As you've pointed out, there are different kinds of state. Telephone switches have been designed which maintain "state" for 10's of thousands of calls, and that was before memory cost only hundreds of dollars per Gigabyte.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_19e.cd816ec.2b1cd6a3_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 10:29:41 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19514
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 10:29:40 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2FW1g06246
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 10:32:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FW1v06241
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 10:32:01 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19487
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 10:29:09 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FV7v06149;
	Mon, 2 Dec 2002 10:31:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FUYv06047
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 10:30:34 -0500
Received: from imo-d03.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19379
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 10:27:43 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d03.mx.aol.com (mail_out_v34.13.) id b.12a.1c3acee1 (657);
	Mon, 2 Dec 2002 10:30:11 -0500 (EST)
Message-ID: <12a.1c3acee1.2b1cd683@aol.com>
Date: Mon, 2 Dec 2002 10:30:11 EST
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
To: woody@pch.net, rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_12a.1c3acee1.2b1cd683_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_12a.1c3acee1.2b1cd683_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 11/30/2002 9:24:30 AM Eastern Standard Time, woody@pch.net 
writes:


>    > Another question: Was it stated during the presentation that the
>     > packet loss ratio was 40%?
> 
> Four people measured loss between 27% and 30% during that call.  Henning
> measured 0%, and then subsequently amended that to, I believe, 7%.  30%
> loss is typical on the connection to the phone in Nairobi, sometimes
> lower, sometimes higher.
> 
>     > Whatever the loss ratio, why are packets lost? Is it because of
>     > congestion at some point (and discard because the buffer is full)?
> 
> The simple answer is that they're discarded because they weren't paid for.
> There are several thousand 128K DSL circuits terminating on a 384K
> satellite link there, which means, in ISP-speak, about a thousand-to-one
> subscriber ratio, which is one order of magnitude higher than you'd
> typically find here in the States, for instance.  If they could afford to
> pay ten times more for their service, which they can't, because it's not a
> wealthy country, they'd receive service comparable to that which we
> recieve here.  To put things in perspective, if you were in Korea, your
> service would be as much better than US service as US service is than
> Kenyan service.  So there's certainly a lot that can be done if the
> government wanted to improve the quality of service, rather than mandate
> QoS.
> 


[MAP] So take this simple case and apply an emergency. (I'd like to see the 
performance last week right after the bombing in Kenya!)

If 10 or 20 or whatever number of those "several thousands" of end users 
tried to make a VoIP call via that single 384K link, what would have 
happened? It is my understanding that, at some point, the overload would have 
caused either:

1. all calls to be of unacceptable quality
2. new call attempts would have been rejected

In either case, how would one individual who is supposed to have authorized 
emergency access, get their call through? I and others have asked this 
question numerous times, but have not seen any hint of an answer, just claims 
that one should "overprovision" to prevent this from happening. Clearly, 
overprovisioning or adding capacity in real time can not solve the problem in 
this situation. I'm looking for a concrete solution. I have been lead to 
believe that there is none at present, thus the existance of IEPREP.

Mike Pierce
Artel



--part1_12a.1c3acee1.2b1cd683_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 11/30/2002 9:24:30 AM Eastern Standard Time, woody@pch.net writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px"> &nbsp;&nbsp;&gt; Another question: Was it stated during the presentation that the
<BR> &nbsp;&nbsp;&nbsp;&gt; packet loss ratio was 40%?
<BR>
<BR>Four people measured loss between 27% and 30% during that call. &nbsp;Henning
<BR>measured 0%, and then subsequently amended that to, I believe, 7%. &nbsp;30%
<BR>loss is typical on the connection to the phone in Nairobi, sometimes
<BR>lower, sometimes higher.
<BR>
<BR> &nbsp;&nbsp;&nbsp;&gt; Whatever the loss ratio, why are packets lost? Is it because of
<BR> &nbsp;&nbsp;&nbsp;&gt; congestion at some point (and discard because the buffer is full)?
<BR>
<BR>The simple answer is that they're discarded because they weren't paid for.
<BR>There are several thousand 128K DSL circuits terminating on a 384K
<BR>satellite link there, which means, in ISP-speak, about a thousand-to-one
<BR>subscriber ratio, which is one order of magnitude higher than you'd
<BR>typically find here in the States, for instance. &nbsp;If they could afford to
<BR>pay ten times more for their service, which they can't, because it's not a
<BR>wealthy country, they'd receive service comparable to that which we
<BR>recieve here. &nbsp;To put things in perspective, if you were in Korea, your
<BR>service would be as much better than US service as US service is than
<BR>Kenyan service. &nbsp;So there's certainly a lot that can be done if the
<BR>government wanted to improve the quality of service, rather than mandate
<BR>QoS.
<BR></BLOCKQUOTE>
<BR>
<BR>
<BR>[MAP] So take this simple case and apply an emergency. (I'd like to see the performance last week right after the bombing in Kenya!)
<BR>
<BR>If 10 or 20 or whatever number of those "several thousands" of end users tried to make a VoIP call via that single 384K link, what would have happened? It is my understanding that, at some point, the overload would have caused either:
<BR>
<BR>1. all calls to be of unacceptable quality
<BR>2. new call attempts would have been rejected
<BR>
<BR>In either case, how would one individual who is supposed to have authorized emergency access, get their call through? I and others have asked this question numerous times, but have not seen any hint of an answer, just claims that one should "overprovision" to prevent this from happening. Clearly, overprovisioning or adding capacity in real time can not solve the problem in this situation. I'm looking for a concrete solution. I have been lead to believe that there is none at present, thus the existance of IEPREP.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR>
<BR></FONT></HTML>

--part1_12a.1c3acee1.2b1cd683_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 10:36:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19946
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 10:36:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2FcqO07336
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 10:38:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Fcpv07333
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 10:38:51 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19919
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 10:36:00 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Fc1v07260;
	Mon, 2 Dec 2002 10:38:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FbFv06958
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 10:37:15 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA19754
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 10:34:24 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 940797A0A; Mon,  2 Dec 2002 07:37:09 -0800 (PST)
Date: Mon, 2 Dec 2002 10:37:10 -0500
Subject: Re: [Ieprep] IEprep in JP
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <b8.329cb1e2.2b1cd5b9@aol.com>
Message-Id: <EC1E0B20-060B-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 10:26 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] Perhaps you could elaborate. We've had a lack of info about 
> "deployed IEPREP services" in other countries (besides the US), either 
> in the circuit-switched world or IP. As far as I know, GETS in the US 
> and MLPP are the only ones today in the circuit switched world and 
> there are none in IP. The IAA in Japan is something completely 
> different, as is the sevice in UK that was described as a way to get 
> priority dial tone.

IAA is an IEprep service.  IEprep is not scoped to only be voice 
services
such as GETS.  And IEprep is not scoped to inherently require MLPP.

Kindly recall that IETF operates on a global basis.  Just because the
USG thinks it wants MLPP does not mean that it is a global all-countries
requirement from an IETF perspective.  Our standards have to be equally
applicable to all countries (including less connected countries and 
non-western
countries) [1] and must not have nationalised biases.  RFC-1984 is one 
example
of this global perspective on IETF work.

> What specific service in Japan are you referring to? Your statement 
> that their "deployed IEPREP service is identical to their deployed 
> regular commercial IP service" sounds like it means that they have no 
> actual IEPREP service.

No, they reportedly do have an IEprep service to support IAA, but the 
Japanese
reportedly do not believe that MLPP or other QoS mechanisms is a 
requirement for
IEprep.   Instead, they reportedly believe that they need to use 
applications
that are well adapted for the Internet -- rather than trying to make 
the Internet
into a circuit-switched network (as some here seem to be trying to do).

(I'll see if I can't get one of the Japanese folks I talked with to 
describe their
service in their own words on this list, either directly or by 
forwarding a note
from them (with their prior permission before forwarding of course)).

Ran

[1] One of the reasons that IETF still does its main work via email and 
that email list
consensus is the official metric that WG Chairs are supposed to use 
(rather
than using WG meeting in-person consenus) is that we want to permit 
folks in remote
countries without travel budget to be able to participate in IETF 
Working Groups.
This property has been very important in a number of cases (e.g. MIME).

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 10:50:36 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA20900
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 10:50:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2FqvJ08161
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 10:52:57 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Fqvv08158
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 10:52:57 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA20847
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 10:50:05 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Fq1v08122;
	Mon, 2 Dec 2002 10:52:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Fp4v08051
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 10:51:04 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA20707
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 10:48:13 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 7622D7A0A; Mon,  2 Dec 2002 07:50:58 -0800 (PST)
Date: Mon, 2 Dec 2002 10:50:59 -0500
Subject: Re: [Ieprep] Re: Looking for governmental feedback as well
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <1c6.26b575b.2b1cd5cb@aol.com>
Message-Id: <DA58C4B5-060D-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 10:27 America/Montreal, Mpierce1@aol.com 
wrote:
> This WG needs to define how to deal with the situation which might 
> occur in a disaster when, in spite of all the above things being done, 
> there is still congestion.

FIRST, you need to produce a credible rationale for believing that the 
core
network will have congestion.  So far, all the disasters that have 
happened
(including some with multiple large fibre cuts) have NOT experienced 
congestion
in the core.  You and a few others keep asserting that we need to worry
about congestion in the core, despite all evidence (including the 
massive
fibre cuts and power outages in lower Manhatten on 9/11/01) saying that 
such
is not a credible problem.

Congestion in access links is of course routine, but is not something 
that
can be solved by an ISP since the access customer controls that link.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 10:51:03 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA20927
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 10:51:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2FrOF08195
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 10:53:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FrOv08192
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 10:53:24 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA20891
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 10:50:32 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Fp4v08049;
	Mon, 2 Dec 2002 10:51:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2FoLv08006
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 10:50:21 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA20665
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 10:47:29 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAGMDV>; Mon, 2 Dec 2002 10:47:52 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A41499D@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'Carl Ford'" <carl@ietfwatch.net>, ieprep@ietf.org
Subject: redundancy RE: [Ieprep] any commercial usage for "emergency QoS"?
Date: Mon, 2 Dec 2002 10:47:43 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

At least one perspective on IEPREP is "opportunistic" use of Internet
resources.  Many of the organizations involved in emergency preparedness and
reaction DO have "private" communications facilities of some sort.  However,
these facilities are no where near as ubiquitous as the public
communications facilities (PSTN and Internet).  The idea is to be able to
take advantage of "whatever IS there" when the emergency takes place, until
the dedicated facilities can be installed (e.g., a FEMA portable "site").

I think all your suggestions on redundancy are very valid (for both
emergency and on-emergency uses), but I don't know whether they fall within
the scope of IEPREP, or even IETF.

-----Original Message-----
From: Carl Ford [mailto:carl@ietfwatch.net]
Sent: Thursday, November 28, 2002 9:11 AM
To: Bill Woodcock; Carl Ford; Mpierce1@aol.com; rja@extremenetworks.com;
ieprep@ietf.org
Subject: Re: [Ieprep] any commercial usage for "emergency QoS"?


I was hoping for a discussion on redundancy.  

As a "bell-head" I have been involved in providing customers with alternate
serving 
wire centers and SONET.  Both were aimed at making the communication path
more 
redundant, but both were from the same provider.  With the Internet my sense
is that 
we can provide better redundancy by saying the following.

1 ISPs should be chosen that are not utilizing the same methodology of
transport.  
Wireless, dial-up, dedicated (particularly if selected from an alternative
carrier) all add 
to redundancy.

2) Internal Intranets that support multiple connection points to ISPs
provide even more 
redundancy.  I am not sure here if it should be stated that some reserve to
rebuild a 
backbone may be a good idea.

3) Security and QoS methodologies should be thought of as when available,
and 
ingress into them should be planned for an any to any methodology.  Meaning
it is 
great if we give the emergency services there own edge devices with priority
services, 
but it probably more likely that they are going to have to connect up at a
library / 
Internet Cafe using some low quality edge point to reach the Internet.

Am I making sense?

Kind Regards,

Carl Ford 




> On Thu, 28 Nov 2002, Carl Ford wrote:
>     > My first question would be what if access were to be unattainable
>     > to the Internet.
> 
> That certainly happens, as we've all noticed from time to time.  And when
> it does, no amount of QoS or additional rules or more fragile routers or
> guys with guns are going to move your packets for you.  The only thing
> that's going to help is having your connectivity restored, and the only
> way that'll happen quickly is if you share fate with the largest number of
> other users.  That is, when there's a problem, you _really_ want to be
> _just like everybody else_, since the aggregated bulk of the users are the
> ones who'll be restored first.
> 
>     > But fundamentally the question that should be asked is what
>     > scenarios are there that would make the Internet inaccessible and
>     > what should be done to prepare for that.
> 
> Keep a candle, some matches, and a good book handy?  :-)
> 
>                                 -Bill



--
Sent from KGB Internet Solutions OpenWebMail
http://www.kgbinternet.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 10:52:26 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA21004
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 10:52:26 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2FskU08291
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 10:54:46 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Fskv08288
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 10:54:46 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA20985
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 10:51:54 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Fs1v08246;
	Mon, 2 Dec 2002 10:54:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Frov08220
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 10:53:50 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA20923
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 10:50:59 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 8A3D27A0A; Mon,  2 Dec 2002 07:53:44 -0800 (PST)
Date: Mon, 2 Dec 2002 10:53:45 -0500
Subject: Re: [Ieprep] defining a particular problem
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep '" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E779C9@mcl-its-exs01.mail.saic.com>
Message-Id: <3D417A69-060E-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 08:07 America/Montreal, King, Kimberly S. 
wrote:
> Part of the impediments to ieprep progress (and this not directed at 
> Ran) is
> that all problems in all contexts and all requirements are discussed 
> at the
> same time.  However, it is more productive to define a particular 
> problem,
> the context, and try to find some effective reasonable way to approach 
> it.

In this respect, James Polk's draft enumerating a wide (but incomplete) 
set
of scenarios is part of the reason that we keep talking about all the 
contexts
simultaneously.

If someone wants to narrow the scope of the WG activity, please someone
make a specific proposal.  I'd be happy to *reduce* the scope of this 
WG in
nearly any way at all (specifically including eliminating the WG out 
right).

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 11:04:40 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA22008
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 11:04:40 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2G71209180
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 11:07:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2G71v09171
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 11:07:01 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA21980
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 11:04:09 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2G64v09108;
	Mon, 2 Dec 2002 11:06:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2G58v09036
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 11:05:08 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA21855
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 11:02:15 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 87D117A0A; Mon,  2 Dec 2002 08:04:56 -0800 (PST)
Date: Mon, 2 Dec 2002 11:04:57 -0500
Subject: Re: [Ieprep] problem context
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep@ietf.org
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <146.3f435a1.2b1cd614@aol.com>
Message-Id: <CDDC5E86-060F-11D7-960D-00039357A82A@extremenetworks.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gB2G58v09037
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


On Monday, Dec 2, 2002, at 10:28 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] Whether or not something is currently widely deployed is not the 
> issue. We're talking about whether or not the necessary mechanisms are 
> "standardized". Our goal, as Don explained, is to get commercial 
> standards (IETF or others) which DoD (and others) can then reference 
> to ask for (pay for) things to be deployed.

DiffServ using EF and AF are already on the IETF standards-track.
If you just want that, can we please shutdown this WG now ?

> [MAP] Then please let on get on with the "standardization" work so 
> that what we need can be defined.

DoD deployment of VoIP on DoD's internal networks is NOT an IEprep 
scenario.
It is out of scope for this WG.  And as noted above, DiffServ is 
already on
the IETF standards-track, so there is nothing left to be done if EF or 
AF
approaches work (and AF absolutely can provide MLPP, including optional
pre-emption, at the IP-layer on packets).  Both AF and EF are widely
available in shipping products (though implementation quality varies).

So what exactly do you want from this WG given that we have a mechanism
on the standards-track that can perform MLPP if the operator chooses to
use it ?

Requiring operators to deploy any mechanism is outside IETF scope, so
trying to require that operators deploy inter-domain MLPP is very very
far outside IETF scope.

> [MAP] You misinterpreted what I (and probably others) have said. I did 
> not say that "overprovisioning is not acceptable" (your quote) or that 
> we need to eliminate overprovisioning. Any carrier or network operator 
> can overprovision if they want. I said that "overprovisioning is not 
> an acceptable solution" to the problem that the IEPREP WG was formed 
> to solve.

You all are obliged first to explain WHY it is not acceptable to you 
all,
particularly since it clearly has worked in every natural (e.g. Kobe, 
Northridge,
others) and terrorist (on 9/11) disaster so far.

> > [MAP] But the bottom line, after all the existing solutions are laid 
> on the
> > table, is that none have been able to explain how priority calls get
> > the required treatment over regular calls on IP-based networks in a
> > situation of congestion using existing standards.
>
> [RJA]    And I thought that was NOT Don's stated objective.  His 
> objective was that the
> preferred calls were able to get through, which is one that the 
> operators here seem
> to think can be achieved (through some combination of 
> over-provisioning and these other
> mechanisms I discussed earlier).  Maybe I mis-read his objective ?
>
>
>
> [MAP] I believe it is exactly the objective, stated in slightly 
> different words.

Don,
	Could you please clarify whether "getting the calls through" is the 
objective
or forcing non-DoD operators to deploy a particular mechanism (e.g. 
MLPP) of your
liking is the objective ?  To several of us, Mike's words are 
materially different
than your earlier words, not merely "stated in slightly different 
words".

> [MAP] So back to the need for "commercial" standards to solve that 
> "pain".

The commercial standards needed for DoD's own networks ALREADY exist.
The set include DiffServ with EF and/or AF, IEEE 802.1p, and others.
That standards work is already done and DoD can buy COTS product with
those capabilities today (e.g. ours, other vendors).  And stuff for use
solely inside DoD (or any other large private) networks is outside the
scope of IEprep.

Can we PLEASE shutdown this WG since we already have standard mechanisms
for MLPP defined by IETF and that is what Mike *now* says he wants from
this WG ?

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 11:15:40 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA23105
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 11:15:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2GI0h10391
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 11:18:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2GI0v10388
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 11:18:00 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA23040
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 11:15:08 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2GH2v10340;
	Mon, 2 Dec 2002 11:17:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2GGqv10309
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 11:16:52 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA22991
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 11:14:00 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id E03FA7A0A; Mon,  2 Dec 2002 08:16:41 -0800 (PST)
Date: Mon, 2 Dec 2002 11:16:43 -0500
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <12a.1c3acee1.2b1cd683@aol.com>
Message-Id: <7248D878-0611-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 10:30 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] So take this simple case and apply an emergency. (I'd like to 
> see the performance last week right after the bombing in Kenya!)
>
> If 10 or 20 or whatever number of those "several thousands" of end 
> users tried to make a VoIP call via that single 384K link, what would 
> have happened? It is my understanding that, at some point, the 
> overload would have caused either:
>
> 1. all calls to be of unacceptable quality
> 2. new call attempts would have been rejected

	The IP-layer doesn't have the concept of a "call", so (B) can't happen 
at
the IP-layer.  Bill's SIP server is not on the access link (as I 
understand it),
so would not be subject to the emergency -- it would be able to process 
calls
normally without any MLPP.  (Bill please clarify these points.)

	If congestion on the last mile link occurred (which we don't know 
right now),
then it could ONLY be solved using link-specific deployed mechanisms 
(e.g. router
CoS like Sprint/Genuity described would be sufficient).  If congestion 
occurred,
then what would have happened is probably packet loss (after the packet 
buffers
on the access routers filled up).  How much packet loss occurred 
depends on the
depth of those buffers, the offered load, the link capacity, etc.  So 
we don't
know whether congestion occurred (yet, Bill should ask) or exactly what 
happened
if it occurred.

	It is entirely possible that the Nairobi VoIP user was in fact having 
no
problems using his link during that situation.  As of this second, we 
don't
know (and Bill should ask).

> I and others have asked this question numerous times, but have not 
> seen any hint of an answer, just claims that one should 
> "overprovision" to prevent this from happening. Clearly, 
> overprovisioning or adding capacity in real time can not solve the 
> problem in this situation. I'm looking for a concrete solution. I have 
> been lead to believe that there is none at present, thus the existance 
> of IEPREP.

Concrete existing solutions:
	1)	Deploy link-specific QoS.  For DSL see T1/ANSI/ITU-T standards,
		for Ethernet see 802.1p, for cable modems see DOCSIS standards.
	2) 	Deploy access router CoS (a la Sprint/Genuity product offerings).
	     This is already widely available and is outside of IETF's scope 
since
	     it is an implementation detail inside of a router.

Please STOP claiming that concrete solutions to the access link 
congestion
issue have not been offered.  It is a bold-faced lie and it is quite 
tiresome
at this point.

And please STOP claiming that IETF lacks mechanisms that DoD could use 
to
deploy MLPP internally.  DiffServ is on the standards track already with
AF and EF.  So that is another bold-faced lie.

It is just not productive when you keep repeating claims that are 
obviously untrue
to anyone reading the list's archives.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 11:22:39 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA23626
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 11:22:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2GP0w10850
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 11:25:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2GP0v10847
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 11:25:00 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA23578
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 11:22:08 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2GO1v10753;
	Mon, 2 Dec 2002 11:24:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2GL1v10556
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 11:21:01 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA23277
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 11:18:09 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 025917A0A; Mon,  2 Dec 2002 08:20:54 -0800 (PST)
Date: Mon, 2 Dec 2002 11:20:56 -0500
Subject: Re: [Ieprep] alternate model for edge QoS
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <19e.cd816ec.2b1cd6a3@aol.com>
Message-Id: <092B2C48-0612-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 10:30 America/Montreal, Mpierce1@aol.com 
wrote:
> I think it's time to end this circular argument that:
>
> 1. States are bad.
> 2. Any notion of "connections" requires "states".
> 3. Any notion of "connections" is bad (I regard RSVP as a way to make 
> IP look "connection oriented")
> 4. IP is connectionless, therefore any notion of "connections" 
> destroys its beauty.
> 5. Any notion of "state" destroys its beauty.

1 is true if edited to be "excessive state is bad".
2 and 3 are unequivocally true for an IP router.

If you ever were to build a router, you'd learn this.
Folks who build routers and folks who operate lots of
routers generally have already learned this.

4 and 5 above aren't technical comments.

And this WG will NOT be making radical changes to the Internet
Architecture as that is NOT within scope here.

MPLS exists and can be used to build a circuit-switched overlay
on top of an IP network.  If folks want a circuit-switched network
but want to call it "IP", then that would be what one could do.
Those sorts of networks are MORE prone to failure in a crisis
of course.  Our existing IP network does really well.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 11:41:46 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25052
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 11:41:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2Gi7d12580
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 11:44:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Gi7v12577
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 11:44:07 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA25027
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 11:41:15 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Gh6v12478;
	Mon, 2 Dec 2002 11:43:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2GgOv12427
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 11:42:24 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA24914
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 11:39:32 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAGMQJ>; Mon, 2 Dec 2002 11:39:56 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A41499F@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        "Choi, Don"
	 <choid@ncr.disa.mil>
Cc: ieprep <ieprep@ietf.org>
Subject: NCS RE: [Ieprep] problem context
Date: Mon, 2 Dec 2002 11:39:47 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

NO.
I believe that the NCS objective is NOT the same as the DoD objective.

I am NOT speaking for the NCS, but saying what I, personally, THINK the NCS
objective is.  And I would be glad for correction/confirmation.

First, that the current NCS CSN programs (e.g., GETS, WPS, and a couple of
others) continue to provide "higher probability of completion" WITHIN the
CSN when a) islands of IP are deployed WITHIN the CSN, and b) calls
(specifically voice calls)  originate in, terminate in, or cross an IP
network (whether an ISP, a corporate network, a VPN, or a private network).

Second, that as "convergence" leads to a blurring of the distinction between
voice and data networks, and as non voice application (both inelastic and
elastic) become a critical part of managing emergency situations, there be
relevant labels and/or other "hooks" within the protocol suite so that 
IF these critical non-voice applications DO suffer under congestion (i.e.,
"there is a problem")
AND  a program is launched to alleviate the problem
THEN deployment of the program need not be delayed because of the lack of
relevant labels or other "hooks" within the protocol suite. 

-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Friday, November 29, 2002 9:43 AM
To: Choi, Don
Cc: ieprep
Subject: Re: [Ieprep] problem context



On Wednesday, Nov 27, 2002, at 11:37 America/Montreal, Choi, Don wrote:
> What DoD wants is, simply put, when a mission critical user makes a 
> call,
> the call goes through and stays up. It does not say calls have to be
> pre-empted nor packets should be dropped.

	  Is the NCS objective for the 
commercial
world identical to the DISA objective for DoD IP networks ?

	
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 12:22:07 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA27127
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 12:22:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2HOTJ15588
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 12:24:29 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2HOTv15585
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 12:24:29 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA27088
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 12:21:36 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2HNBv15543;
	Mon, 2 Dec 2002 12:23:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2HKav15413
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 12:20:36 -0500
Received: from [140.32.132.66] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA26884
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 12:17:44 -0500 (EST)
Received: from mail.nps.navy.mil by [140.32.132.66]
          via smtpd (for [132.151.6.1]) with ESMTP; Mon, 2 Dec 2002 09:16:21 -0800
Received: from ziggy ([131.120.179.220])
	by capella.nps.navy.mil (8.12.2/8.12.2) with ESMTP id gB2HH1bR026211;
	Mon, 2 Dec 2002 09:17:02 -0800 (PST)
Subject: Re: [Ieprep] Accounting
From: Rex Buddenberg <budden@nps.navy.mil>
To: RJ Atkinson <rja@extremenetworks.com>
Cc: Mpierce1@aol.com, ieprep@ietf.org
In-Reply-To: <E706EF3C-03EF-11D7-9A09-00039357A82A@extremenetworks.com>
References: <E706EF3C-03EF-11D7-9A09-00039357A82A@extremenetworks.com>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) 
Date: 02 Dec 2002 09:12:57 -0800
Message-Id: <1038849178.1219.4.camel@ziggy>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

On Fri, 2002-11-29 at 15:11, RJ Atkinson wrote:
> 
> On Friday, Nov 29, 2002, at 13:07 America/Montreal, Mpierce1@aol.com 
> wrote:
> > [RJA] So I remain very confused about what the accounting advocates
> > are asking for.  I might be slower than most to puzzle some
> > stuff out, but what I got out of the Accounting presentation
> > was roughly this:
> >
> >     - Accounting is a requirement.
> >     - We don't mean that packet accounting is a requirement.
> >

> Mike's reply has been very helpful.  There are a couple of things that 
> I'd
> like folks to read/correct still.  And if anyone has different answers 
> than
> Mike had, please speak up and help me understand the differences.


Ran,  are you binning key management under accounting?

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 12:23:32 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA27227
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 12:23:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2HPsh15679
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 12:25:54 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2HPsv15676
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 12:25:54 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA27206
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 12:23:01 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2HPCv15657;
	Mon, 2 Dec 2002 12:25:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2HOmv15607
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 12:24:48 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA27097
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 12:21:56 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 959E67A0A; Mon,  2 Dec 2002 09:24:41 -0800 (PST)
Date: Mon, 2 Dec 2002 12:24:43 -0500
Subject: Re: [Ieprep] Accounting
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Rex Buddenberg <budden@nps.navy.mil>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <1038849178.1219.4.camel@ziggy>
Message-Id: <F222D553-061A-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 12:12 America/Montreal, Rex Buddenberg wrote:
> Ran,  are you binning key management under accounting?

Hmm.  No.  Had not occurred to me that they were related.
Are they related ?

Ran

PS:	And I do think key management is its usual nightmare here, just as
it is in other places where one wants to do authentication of stuff.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 12:24:37 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA27287
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 12:24:37 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2HQw015730
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 12:26:58 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2HQwv15727
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 12:26:58 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA27262
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 12:24:05 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2HQ1v15705;
	Mon, 2 Dec 2002 12:26:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2HP5v15644
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 12:25:05 -0500
Received: from c015.snv.cp.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id MAA27141
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 12:22:12 -0500 (EST)
Received: (cpmta 5650 invoked from network); 2 Dec 2002 09:24:59 -0800
Received: from 195.242.54.251 (HELO pavarotti)
  by smtp.globalipsound.com (209.228.35.122) with SMTP; 2 Dec 2002 09:24:59 -0800
X-Sent: 2 Dec 2002 17:24:59 GMT
Reply-To: <alan.duric@globalipsound.com>
From: "Alan Duric" <alan.duric@globalipsound.com>
To: "'ieprep'" <ieprep@ietf.org>
Cc: "'RJ Atkinson'" <rja@extremenetworks.com>, <Mpierce1@aol.com>
Subject: RE: [Ieprep] Internet faults, VoIP demo, etc and others
Date: Mon, 2 Dec 2002 18:21:24 +0100
Message-ID: <001f01c29a27$418e44c0$fb36f2c3@pavarotti>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook, Build 10.0.4024
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106
Importance: Normal
In-Reply-To: <7248D878-0611-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

> Concrete existing solutions:
> 	1)	Deploy link-specific QoS.  For DSL see 
> T1/ANSI/ITU-T standards,
> 		for Ethernet see 802.1p, for cable modems see 
> DOCSIS standards.
> 	2) 	Deploy access router CoS (a la Sprint/Genuity 
> product offerings).
> 	     This is already widely available and is outside of 
> IETF's scope 
> since
> 	     it is an implementation detail inside of a router.
 
This is very good start, but not utterly sufficient. Since, mentioned
mechanisms would still not guarantee that packet loss will not happen,
and there's more that can be done in order to have IEPREP call at
certain speech quality level where the purpose of the call can be
conveyed between involved parties, unless the purpose of IEPREP work
related to IEPREP call is to "count" the delivered and lost packets.
There's number of things related to choice of codecs, transcoding,
packet loss concealment mechanisms, jitter buffers (all of the packets
delivered are not necessarily usable if not on time, .) that will effect
conversation in a great manner for the same values of packet loss, so as
said in Yokohama during IEPREP session from packet counting and network
perspective "launch" and "lunch" might sound pretty much the same, but I
guess we all agree in reality of IEPREP initiated call, (figuratively
speaking) those words should not be confused.
So, my question is - does this make sense for IEPREP related work?
			if does how can we contribute to this part?

Regards,
Alan Duric


________________________________________________________________________
____________
Cuius rei demonstrationem mirabilem sane detexi hanc marginis exiguitas
non caperet. 





_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 12:51:41 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28843
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 12:51:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2Hs3p17876
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 12:54:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Hs3v17872
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 12:54:03 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28773
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 12:51:10 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Hr4v17745;
	Mon, 2 Dec 2002 12:53:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Hq2v17676
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 12:52:02 -0500
Received: from imo-m05.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28629
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 12:49:09 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m05.mx.aol.com (mail_out_v34.13.) id j.6d.36a09dd (25508);
	Mon, 2 Dec 2002 12:51:55 -0500 (EST)
Message-ID: <6d.36a09dd.2b1cf7bb@aol.com>
Date: Mon, 2 Dec 2002 12:51:55 EST
Subject: Re: [Ieprep] IEprep in JP
To: rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_6d.36a09dd.2b1cf7bb_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_6d.36a09dd.2b1cf7bb_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/2/2002 10:37:39 AM Eastern Standard Time, 
rja@extremenetworks.com writes:


> On Monday, Dec 2, 2002, at 10:26 America/Montreal, Mpierce1@aol.com 
> wrote:
> > [MAP] Perhaps you could elaborate. We've had a lack of info about 
> > "deployed IEPREP services" in other countries (besides the US), either 
> > in the circuit-switched world or IP. As far as I know, GETS in the US 
> > and MLPP are the only ones today in the circuit switched world and 
> > there are none in IP. The IAA in Japan is something completely 
> > different, as is the sevice in UK that was described as a way to get 
> > priority dial tone.
> 
> IAA is an IEprep service.  IEprep is not scoped to only be voice services
> such as GETS.  And IEprep is not scoped to inherently require MLPP.
> 

[MAP] Yes, I understand that IAA is an "IEPREP service" in the general sense. 
What I meant is that it is not a service which can or does operate on the 
basis that certain users are given authority to get a better "priority" than 
others. My points are all related to a service with special authorized users.


> Kindly recall that IETF operates on a global basis.  Just because the
> USG thinks it wants MLPP does not mean that it is a global all-countries
> requirement from an IETF perspective.  Our standards have to be equally
> applicable to all countries (including less connected countries and 
> non-western
> countries) [1] and must not have nationalised biases.  RFC-1984 is one 
> example
> of this global perspective on IETF work.
> 
[MAP] Please note that the "MLPP" that we're using as a model for an IP 
equivalent is also a "global standard" (in ITU-T). It can be applied in any 
country. I would agree that it is mostly the US that implements it in private 
neworks (as far as I am aware).

Mike Pierce
Artel


--part1_6d.36a09dd.2b1cf7bb_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/2/2002 10:37:39 AM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">On Monday, Dec 2, 2002, at 10:26 America/Montreal, Mpierce1@aol.com 
<BR>wrote:
<BR>&gt; [MAP] Perhaps you could elaborate. We've had a lack of info about 
<BR>&gt; "deployed IEPREP services" in other countries (besides the US), either 
<BR>&gt; in the circuit-switched world or IP. As far as I know, GETS in the US 
<BR>&gt; and MLPP are the only ones today in the circuit switched world and 
<BR>&gt; there are none in IP. The IAA in Japan is something completely 
<BR>&gt; different, as is the sevice in UK that was described as a way to get 
<BR>&gt; priority dial tone.
<BR>
<BR>IAA is an IEprep service. &nbsp;IEprep is not scoped to only be voice services
<BR>such as GETS. &nbsp;And IEprep is not scoped to inherently require MLPP.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR>
<BR>[MAP] Yes, I understand that IAA is an "IEPREP service" in the general sense. What I meant is that it is not a service which can or does operate on the basis that certain users are given authority to get a better "priority" than others. My points are all related to a service with special authorized users.
<BR>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">Kindly recall that IETF operates on a global basis. &nbsp;Just because the
<BR>USG thinks it wants MLPP does not mean that it is a global all-countries
<BR>requirement from an IETF perspective. &nbsp;Our standards have to be equally
<BR>applicable to all countries (including less connected countries and non-western
<BR>countries) [1] and must not have nationalised biases. &nbsp;RFC-1984 is one example
<BR>of this global perspective on IETF work.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">[MAP] Please note that the "MLPP" that we're using as a model for an IP equivalent is also a "global standard" (in ITU-T). It can be applied in any country. I would agree that it is mostly the US that implements it in private neworks (as far as I am aware).
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_6d.36a09dd.2b1cf7bb_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 12:51:45 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28870
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 12:51:45 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2Hs7Z17905
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 12:54:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Hs7v17902
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 12:54:07 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28802
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 12:51:14 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Hr6v17761;
	Mon, 2 Dec 2002 12:53:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2HqUv17698
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 12:52:30 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28648
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 12:49:37 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAGM5J>; Mon, 2 Dec 2002 12:50:01 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149A0@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>, Mpierce1@aol.com
Cc: ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: Looking for governmental feedback as well
Date: Mon, 2 Dec 2002 12:49:52 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

I don't think anyone is asking the ISPs to solve the problem of congestion
on access links (though it seems that some ISPs (e.g., Sprint) are
addressing it).

What is being asked is for IETF to help identify emergency traffic WHEN SOME
ENTITY is addressing congestion on the access links.

-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Monday, December 02, 2002 10:51 AM
To: Mpierce1@aol.com
Cc: ieprep
Subject: Re: [Ieprep] Re: Looking for governmental feedback as well


Congestion in access links is of course routine, but is not something 
that
can be solved by an ISP since the access customer controls that link.

Ran

__
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 12:56:19 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29192
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 12:56:19 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2Hwfk18226
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 12:58:41 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Hwfv18223
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 12:58:41 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29136
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 12:55:46 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Hu3v18045;
	Mon, 2 Dec 2002 12:56:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Ht6v17980
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 12:55:06 -0500
Received: from imo-m03.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28907
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 12:52:12 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m03.mx.aol.com (mail_out_v34.13.) id j.29.326c7f94 (25508);
	Mon, 2 Dec 2002 12:54:39 -0500 (EST)
Message-ID: <29.326c7f94.2b1cf85e@aol.com>
Date: Mon, 2 Dec 2002 12:54:38 EST
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
To: rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_29.326c7f94.2b1cf85e_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_29.326c7f94.2b1cf85e_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/2/2002 11:18:10 AM Eastern Standard Time, 
rja@extremenetworks.com writes:


> On Monday, Dec 2, 2002, at 10:30 America/Montreal, Mpierce1@aol.com 
> wrote:
> > [MAP] So take this simple case and apply an emergency. (I'd like to 
> > see the performance last week right after the bombing in Kenya!)
> >
> > If 10 or 20 or whatever number of those "several thousands" of end 
> > users tried to make a VoIP call via that single 384K link, what would 
> > have happened? It is my understanding that, at some point, the 
> > overload would have caused either:
> >
> > 1. all calls to be of unacceptable quality
> > 2. new call attempts would have been rejected
> 
>   [RJA]  The IP-layer doesn't have the concept of a "call", so (B) can't 
> happen at
> the IP-layer.  Bill's SIP server is not on the access link (as I understand 
> it),
> so would not be subject to the emergency -- it would be able to process 
> calls
> normally without any MLPP.  (Bill please clarify these points.)
> 


[MAP] I understand fully that the IP-layer is unaware of "calls" or "voice 
packets". That wasn't the point. If some higher layer function doesn't limit 
the setup of new calls (2. above) when it knows that there are already too 
many being stuffed through a finite size link, then the IP-layer would be 
overwhelmed with packets which it can't handle (1. above).


> [RJA]    If congestion on the last mile link occurred (which we don't know 
> right now),
> then it could ONLY be solved using link-specific deployed mechanisms (e.g. 
> router
> CoS like Sprint/Genuity described would be sufficient).  If congestion 
> occurred,
> then what would have happened is probably packet loss (after the packet 
> buffers
> on the access routers filled up).  How much packet loss occurred depends on 
> the
> depth of those buffers, the offered load, the link capacity, etc.  So we 
> don't
> know whether congestion occurred (yet, Bill should ask) or exactly what 
> happened if it occurred.
> 



[MAP] So I take that as an agreement with what I said, that in the case of 
congestion, a lot of packets would be lost. However, you still seem to 
question whether or not congestion can occur. Please explain: If 10 users 
each try to make a phone call (each of which I understand needs 80+ kbit/s) 
and they all have to go across the same 384 kbit/s link through a satallite, 
how can you argue that congestion does not occur? (Change the numbers if you 
want, but a case such as this can always exist.)
> 
> [RJA] Concrete existing solutions:
>     1)    Deploy link-specific QoS.  For DSL see T1/ANSI/ITU-T standards,
>         for Ethernet see 802.1p, for cable modems see DOCSIS standards.
>     2)     Deploy access router CoS (a la Sprint/Genuity product offerings).
>          This is already widely available and is outside of IETF's scope 
> since
>          it is an implementation detail inside of a router.
> 
> Please STOP claiming that concrete solutions to the access link congestion
> issue have not been offered.  It is a bold-faced lie and it is quite 
> tiresome
> at this point.
> 
> And please STOP claiming that IETF lacks mechanisms that DoD could use to
> deploy MLPP internally.  DiffServ is on the standards track already with
> AF and EF.  So that is another bold-faced lie.
> 
> It is just not productive when you keep repeating claims that are obviously 
> untrue
> to anyone reading the list's archives.
> 

[MAP] I'll repeat the question which I don't believe I've gotten any concrete 
answer to: In the case above of the 10 callers from Nairobi, if one of them 
was authorized to make emergency calls and the other 9 weren't, what specific 
standardized mechanism (not products) could be used to provide the required 
service across the congested 384 kbit/s satellite link? (The required service 
is that their call gets set up and they can talk.) If you don't believe that 
10 users would cause congestion, then make it 20 or however many of the 
several thousand it would take.

I would find it hard to believe that there are no standardized solutions to 
this. You're right, maybe there is no need for this WG, but I haven't heard 
the solution yet.

Mike Pierce
Artel


--part1_29.326c7f94.2b1cf85e_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/2/2002 11:18:10 AM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">On Monday, Dec 2, 2002, at 10:30 America/Montreal, Mpierce1@aol.com 
<BR>wrote:
<BR>&gt; [MAP] So take this simple case and apply an emergency. (I'd like to 
<BR>&gt; see the performance last week right after the bombing in Kenya!)
<BR>&gt;
<BR>&gt; If 10 or 20 or whatever number of those "several thousands" of end 
<BR>&gt; users tried to make a VoIP call via that single 384K link, what would 
<BR>&gt; have happened? It is my understanding that, at some point, the 
<BR>&gt; overload would have caused either:
<BR>&gt;
<BR>&gt; 1. all calls to be of unacceptable quality
<BR>&gt; 2. new call attempts would have been rejected
<BR>
<BR> &nbsp;[RJA] &nbsp;The IP-layer doesn't have the concept of a "call", so (B) can't happen at
<BR>the IP-layer. &nbsp;Bill's SIP server is not on the access link (as I understand it),
<BR>so would not be subject to the emergency -- it would be able to process calls
<BR>normally without any MLPP. &nbsp;(Bill please clarify these points.)
<BR></BLOCKQUOTE>
<BR>
<BR>
<BR>[MAP] I understand fully that the IP-layer is unaware of "calls" or "voice packets". That wasn't the point. If some higher layer function doesn't limit the setup of new calls (2. above) when it knows that there are already too many being stuffed through a finite size link, then the IP-layer would be overwhelmed with packets which it can't handle (1. above).
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">[RJA] &nbsp;&nbsp;&nbsp;If congestion on the last mile link occurred (which we don't know right now),
<BR>then it could ONLY be solved using link-specific deployed mechanisms (e.g. router
<BR>CoS like Sprint/Genuity described would be sufficient). &nbsp;If congestion occurred,
<BR>then what would have happened is probably packet loss (after the packet buffers
<BR>on the access routers filled up). &nbsp;How much packet loss occurred depends on the
<BR>depth of those buffers, the offered load, the link capacity, etc. &nbsp;So we don't
<BR>know whether congestion occurred (yet, Bill should ask) or exactly what 
<BR>happened if it occurred.
<BR></BLOCKQUOTE>
<BR>
<BR>
<BR>
<BR>[MAP] So I take that as an agreement with what I said, that in the case of congestion, a lot of packets would be lost. However, you still seem to question whether or not congestion can occur. Please explain: If 10 users each try to make a phone call (each of which I understand needs 80+ kbit/s) and they all have to go across the same 384 kbit/s link through a satallite, how can you argue that congestion does not occur? (Change the numbers if you want, but a case such as this can always exist.)
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>[RJA] Concrete existing solutions:
<BR> &nbsp;&nbsp;&nbsp;1) &nbsp;&nbsp;&nbsp;Deploy link-specific QoS. &nbsp;For DSL see T1/ANSI/ITU-T standards,
<BR> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;for Ethernet see 802.1p, for cable modems see DOCSIS standards.
<BR> &nbsp;&nbsp;&nbsp;2) &nbsp;&nbsp;&nbsp;&nbsp;Deploy access router CoS (a la Sprint/Genuity product offerings).
<BR> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;This is already widely available and is outside of IETF's scope since
<BR> &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;it is an implementation detail inside of a router.
<BR>
<BR>Please STOP claiming that concrete solutions to the access link congestion
<BR>issue have not been offered. &nbsp;It is a bold-faced lie and it is quite tiresome
<BR>at this point.
<BR>
<BR>And please STOP claiming that IETF lacks mechanisms that DoD could use to
<BR>deploy MLPP internally. &nbsp;DiffServ is on the standards track already with
<BR>AF and EF. &nbsp;So that is another bold-faced lie.
<BR>
<BR>It is just not productive when you keep repeating claims that are obviously untrue
<BR>to anyone reading the list's archives.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>[MAP] I'll repeat the question which I don't believe I've gotten any concrete answer to: In the case above of the 10 callers from Nairobi, if one of them was authorized to make emergency calls and the other 9 weren't, what specific standardized mechanism (not products) could be used to provide the required service across the congested 384 kbit/s satellite link? (The required service is that their call gets set up and they can talk.) If you don't believe that 10 users would cause congestion, then make it 20 or however many of the several thousand it would take.
<BR>
<BR>I would find it hard to believe that there are no standardized solutions to this. You're right, maybe there is no need for this WG, but I haven't heard the solution yet.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_29.326c7f94.2b1cf85e_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 12:58:13 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29335
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 12:58:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2I0ZD18448
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 13:00:35 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2I0Zv18445
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 13:00:35 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29322
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 12:57:41 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Hx0v18300;
	Mon, 2 Dec 2002 12:59:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Hwov18253
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 12:58:50 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29150
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 12:55:57 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAGM59>; Mon, 2 Dec 2002 12:56:21 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149A1@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>, Mpierce1@aol.com
Cc: ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] Internet faults, VoIP demo, etc and others
Date: Mon, 2 Dec 2002 12:56:10 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

OK.
For these "concrete solutions to access link congestion issues", how do you
suggest we identify "emergency" traffic?

-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Monday, December 02, 2002 11:17 AM
To: Mpierce1@aol.com
Cc: ieprep
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others


Please STOP claiming that concrete solutions to the access link 
congestion
issue have not been offered.  It is a bold-faced lie and it is quite 
tiresome
at this point.
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 13:07:50 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29875
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 13:07:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2IACF19601
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 13:10:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2IACv19586
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 13:10:12 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29863
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 13:07:18 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2I8Hv19388;
	Mon, 2 Dec 2002 13:08:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2I0mv18468
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 13:00:48 -0500
Received: from [140.32.132.66] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29331
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 12:57:54 -0500 (EST)
Received: from mail.nps.navy.mil by [140.32.132.66]
          via smtpd (for [132.151.6.1]) with ESMTP; Mon, 2 Dec 2002 09:56:32 -0800
Received: from ziggy ([131.120.179.220])
	by capella.nps.navy.mil (8.12.2/8.12.2) with ESMTP id gB2Hv8bR028769;
	Mon, 2 Dec 2002 09:57:09 -0800 (PST)
Subject: Re: [Ieprep] Re: Looking for governmental feedback as well
From: Rex Buddenberg <budden@nps.navy.mil>
To: Carl Ford <carl@ietfwatch.net>
Cc: Bill Woodcock <woody@pch.net>, Mpierce1@aol.com, rja@extremenetworks.com,
        ieprep@ietf.org
In-Reply-To: <200212011631.gB1GVPA19487@samsara.kgbinternet.com>
References: <200212011631.gB1GVPA19487@samsara.kgbinternet.com>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) 
Date: 02 Dec 2002 09:53:02 -0800
Message-Id: <1038851586.1224.21.camel@ziggy>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

On Sun, 2002-12-01 at 08:31, Carl Ford wrote:
> 
> Bill Woodcock wrote
> >  I guess it would
> > be more accurate to say that there's a relatively widespread understanding
> > that these are desirable purchasing criteria, and that they guide the
> > purchasing decisions of many of the larger and better-informed buyers of
> > Internet services.  But they're not _majority_ practices, since the
> > _majority_ of sites aren't multi-homed in the first place, only larger
> > ones are.
> > 
> So the question that we have to ask for ieprep is do we have a goal of greater 
> redundancy for the Internet overall, greater redundancy for the government sites 
> specifically, or a specific practise on a government access methods.
> 
> Any thoughts appreciated.
>

Carl,

The notion of 'government owned networks and sites' is of little and
decreasing value:
	- the most important reason is that in disaster relief type of
operations the real workers are at the county and local level and none
of those folks are their own ISPs.  
	- at the federal level, the trend is increasingly away from owning your
own plumbing.  Navy's NMCI contract outsourced both a bunch of
intra-Navy WAN plumbing and a fair amount of what had previously been
obtained from DISA.  
	- even in satellite comms networks, a great deal of intra-DOD traffic
is carried over commercial satellites.  For instance, in the Navy's ADNS
program, they simply wrote a Channel Access Protocol around INMARSAT B
plumbing so it accepts and spews IP datagrams at both ends.  


Sea story.  On 7 Dec 1941, when the last installment of the Purple
decrypt breaking off relations between Japan and US arrived, it was
promptly decrypted and walked up the chain of command in Washington
(before Ambassador Nomura arrived).  Stimson, Marshall, etc reacted by
understanding that this was indeed a war warning and dispatched alerts. 
The Army and Navy-owned circuits were congested so the warnings were
duplicated through commercial channels (e.g. Western Union).  
	In Hawaii, the alerts via the military channels seem to be unaccounted
for in the histories, but the commercial telegram spent the morning in
the messenger bag of the local (Nisei:) messenger while he was in a
ditch taking cover from the battle.  


The proper goal should be greater availability and survivability.  

The provisioning means is usable diverse alternate routes, and when
appropriate, duplicated end systems.  

The business means can be a variety:
	- some 'rules' imposed on the ISPs.  (This is much harder to do in the
unregulated ISP business than it is in the regulated PSTN one).
	- some carrots in the form of operating subsidies like we've done for
commercial shipbuilders and operators for years.  These carrots can be 
		o embedded in contracts for government-procured network services or 
		o they can be overt subsidies as offsets for the tragedy of the
commons.  Public good.




_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 13:09:41 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA00040
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 13:09:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2IC4P19724
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 13:12:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2IC3v19721
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 13:12:03 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29990
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 13:09:10 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2IB3v19661;
	Mon, 2 Dec 2002 13:11:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2I90v19419
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 13:09:00 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29821
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 13:06:06 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAGM7P>; Mon, 2 Dec 2002 13:06:31 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149A2@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        Henning Schulzrinne
	 <hgs@cs.columbia.edu>
Cc: ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] alternate model for edge QoS
Date: Mon, 2 Dec 2002 13:06:22 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

There is no such thing as a "GETS-like" network, IP or otherwise. GETS (and
all the related programs I am aware of) are designed to "exploit" existing
(typically, but not necessarily exclusively, public) networks.

-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Sunday, December 01, 2002 7:48 PM
To: Henning Schulzrinne
Cc: ieprep
Subject: Re: [Ieprep] alternate model for edge QoS


  So this is an area where what DoD IP networks do
and what GETS-like IP networks do will likely vary.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 13:16:53 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA00465
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 13:16:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2IJGt20183
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 13:19:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2IJCv20176
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 13:19:12 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA00426
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 13:16:17 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2II1v20088;
	Mon, 2 Dec 2002 13:18:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2IHRv20048
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 13:17:27 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA00282
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 13:14:33 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 34A6F7A0A; Mon,  2 Dec 2002 10:17:19 -0800 (PST)
Date: Mon, 2 Dec 2002 13:17:20 -0500
Subject: Re: [Ieprep] packet loss guarantees
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep'" <ieprep@ietf.org>
To: <alan.duric@globalipsound.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <001f01c29a27$418e44c0$fb36f2c3@pavarotti>
Message-Id: <4C2A7B0C-0622-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 12:21 America/Montreal, Alan Duric wrote:
> This is very good start, but not utterly sufficient. Since, mentioned
> mechanisms would still not guarantee that packet loss will not happen,

*Nothing* can guarantee that packet loss will not happen.  A trivial 
example
is that if the offered load of legitimate "preferred" traffic exceeds 
the link
capacity, then some packets will be dropped.  Another trivial example 
is that
if there is no path from A to B, then packets from A to B will be lost.
Less trivial examples consider that an adversary might be flooding the 
link
with forged traffic claiming to be "preferred" or other kinds of attacks
or that equipment might fail (e.g. get hit by lightning or a falling 
tree).

Hence, the IEprep question cannot be about "guarantee of zero packet 
loss",
but instead must be (by limtations from physics) of "probability of 
traffic
getting through".

Some of us think that good system engineering, for example including
access link QoS, path-diverse capacity, and over-provisioned backbones
(and for the VoIP case, thoughtful selection of robust codecs), has the
best probability of getting sufficient traffic through.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 13:39:36 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28844
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 12:51:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2Hs4G17885
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 12:54:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Hs3v17873
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 12:54:03 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28769
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 12:51:10 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Hr6v17781;
	Mon, 2 Dec 2002 12:53:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2HqXv17706
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 12:52:33 -0500
Received: from imo-m05.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA28651
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 12:49:40 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m05.mx.aol.com (mail_out_v34.13.) id i.7f.3012729a (25508);
	Mon, 2 Dec 2002 12:52:23 -0500 (EST)
Message-ID: <7f.3012729a.2b1cf7d6@aol.com>
Date: Mon, 2 Dec 2002 12:52:22 EST
Subject: Re: [Ieprep] Accounting
To: K.Carlberg@cs.ucl.ac.uk
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_7f.3012729a.2b1cf7d6_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_7f.3012729a.2b1cf7d6_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 11/30/2002 9:46:06 AM Eastern Standard Time, 
K.Carlberg@cs.ucl.ac.uk writes:


> > [MAP] We start with the requirement that any system that handles IEPREP
> > or any kind of priority telephone calls must be capable of performing
> > accounting, i.e., maintaining Call Detail Records. 
> 
> why?  I don't mean to be dense, but I'm curious if your response is
> rooted in reasons that relate to supporting a business model you have
> in mind.
> 
> part of the reason I have not been convinced about accounting as a
> requirement is that my exposure to it has never been one in which it
> operates in real-time as a means of affecting the flow that transits a
> network.  labeling and subsequent authentication/policing of flows
> represent mechanisms that operates at real time.  accounting is an
> 


[MAP] I never implied that it should operate in real-time to affect the flow. 
My presentation talked about the "auditing" requirement, or "after-the-fact" 
as you said above.

> 
> customers that would pay for a service like GETS, or one based on MLPP, 
> would certainly want accounting and so I'm not saying there isn't a need
> for it.  but since IEPREP is broader in scope than these two systems,
> we can't tie in a third mechanism that in certain cases plays no role.
> for example, the land-line based system of GTPS (in the UK) has no
> need for accounting. 
> 
> 
[MAP] I'm at a loss to explain further to resolve your lack of being 
"convinced", since you say in the final paragraph above that you agree that 
accounting is needed for GETS and MLPP, which is the only point I've made. 
What is it you don't agree with that I said?

Mike Pierce
Artel



--part1_7f.3012729a.2b1cf7d6_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 11/30/2002 9:46:06 AM Eastern Standard Time, K.Carlberg@cs.ucl.ac.uk writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">&gt; [MAP] We start with the requirement that any system that handles IEPREP
<BR>&gt; or any kind of priority telephone calls must be capable of performing
<BR>&gt; accounting, i.e., maintaining Call Detail Records. 
<BR>
<BR>why? &nbsp;I don't mean to be dense, but I'm curious if your response is
<BR>rooted in reasons that relate to supporting a business model you have
<BR>in mind.
<BR>
<BR>part of the reason I have not been convinced about accounting as a
<BR>requirement is that my exposure to it has never been one in which it
<BR>operates in real-time as a means of affecting the flow that transits a
<BR>network. &nbsp;labeling and subsequent authentication/policing of flows
<BR>represent mechanisms that operates at real time. &nbsp;accounting is an
<BR>after-the-fact third mechanism that may or may not be used.</BLOCKQUOTE>
<BR>
<BR>
<BR>[MAP] I never implied that it should operate in real-time to affect the flow. My presentation talked about the "auditing" requirement, or "after-the-fact" as you said above.
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>customers that would pay for a service like GETS, or one based on MLPP, 
<BR>would certainly want accounting and so I'm not saying there isn't a need
<BR>for it. &nbsp;but since IEPREP is broader in scope than these two systems,
<BR>we can't tie in a third mechanism that in certain cases plays no role.
<BR>for example, the land-line based system of GTPS (in the UK) has no
<BR>need for accounting. 
<BR>
<BR></BLOCKQUOTE>
<BR>[MAP] I'm at a loss to explain further to resolve your lack of being "convinced", since you say in the final paragraph above that you agree that accounting is needed for GETS and MLPP, which is the only point I've made. What is it you don't agree with that I said?
<BR>
<BR>Mike Pierce
<BR>Artel
<BR>
<BR></FONT></HTML>

--part1_7f.3012729a.2b1cf7d6_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 13:42:34 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA02075
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 13:42:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2IiuH22337
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 13:44:56 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Iiuv22334
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 13:44:56 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA02046
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 13:42:02 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2If5v22091;
	Mon, 2 Dec 2002 13:41:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Iemv22038
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 13:40:48 -0500
Received: from newdev.harvard.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA01673
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 13:37:54 -0500 (EST)
Received: (from sob@localhost)
	by newdev.harvard.edu (8.12.2/8.12.2) id gB2Ie5iF028770;
	Mon, 2 Dec 2002 13:40:05 -0500 (EST)
Date: Mon, 2 Dec 2002 13:40:05 -0500 (EST)
From: Scott  Bradner <sob@harvard.edu>
Message-Id: <200212021840.gB2Ie5iF028770@newdev.harvard.edu>
To: Mpierce1@aol.com, rja@extremenetworks.com, woody@pch.net
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Cc: ieprep@ietf.org
In-Reply-To: <12a.1c3acee1.2b1cd683@aol.com>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

> 2. new call attempts would have been rejected

no- the congested link would not cause the SIP proxies to reject calls
since they have no knowledge of the congestion

Scott
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 13:43:29 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA02314
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 13:43:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2IjqN22399
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 13:45:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Ijqv22396
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 13:45:52 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA02111
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 13:42:58 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Ig1v22148;
	Mon, 2 Dec 2002 13:42:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Ifiv22117
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 13:41:44 -0500
Received: from c015.snv.cp.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id NAA01697
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 13:38:50 -0500 (EST)
Received: (cpmta 4456 invoked from network); 2 Dec 2002 10:41:37 -0800
Received: from 195.242.54.251 (HELO pavarotti)
  by smtp.globalipsound.com (209.228.35.119) with SMTP; 2 Dec 2002 10:41:37 -0800
X-Sent: 2 Dec 2002 18:41:37 GMT
Reply-To: <alan.duric@globalipsound.com>
From: "Alan Duric" <alan.duric@globalipsound.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>, "'ieprep'" <ieprep@ietf.org>
Subject: RE: [Ieprep] packet loss guarantees
Date: Mon, 2 Dec 2002 19:38:02 +0100
Message-ID: <002601c29a31$f622dc70$fb36f2c3@pavarotti>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook, Build 10.0.4024
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106
Importance: Normal
In-Reply-To: <4C2A7B0C-0622-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Ran,

> *Nothing* can guarantee that packet loss will not happen.  A trivial 
> example ...

Great! Then, I think we are fully in agreement in respect to this point.

> Some of us think that good system engineering, for example 
> including access link QoS, path-diverse capacity, and 
> over-provisioned backbones (and for the VoIP case, thoughtful 
> selection of robust codecs), has the best probability of 
> getting sufficient traffic through.

Glad to see that You recognize among other mechanisms the value of
robust codecs (so that's one more point where we are in agreement) for
IEPREP. Nevertheless, my question to the group is (cc-ed from my
previous posting, with certain overhead info, to stress IMHO a few
additional significant factors):

"There's number of things related to choice of codecs, transcoding,
packet loss concealment mechanisms, jitter buffers (all of the packets
delivered are not necessarily usable if not on time, .) that will effect
conversation in a great manner for the same values of packet loss, so as
said in Yokohama during IEPREP session from packet counting and network
perspective "launch" and "lunch" might sound pretty much the same, but I
guess we all agree in reality of IEPREP initiated call, (figuratively
speaking) those words should not be confused. So, my question is - does
this make sense for IEPREP related work?" 

Best regards,
Alan Duric



________________________________________________________________________
____________
Cuius rei demonstrationem mirabilem sane detexi hanc marginis exiguitas
non caperet.  


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 14:41:58 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA06215
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 14:41:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2JiMu26719
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 14:44:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2JiMv26716
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 14:44:22 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA06182
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 14:41:27 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2JfDv26599;
	Mon, 2 Dec 2002 14:41:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Je1v26541
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 14:40:01 -0500
Received: from bells.cs.ucl.ac.uk (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id OAA05929
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 14:37:06 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.11511-0@bells.cs.ucl.ac.uk>; Mon, 2 Dec 2002 19:39:50 +0000
To: Mpierce1@aol.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Accounting
In-reply-to: Your message of "Mon, 02 Dec 2002 12:52:22 EST." <7f.3012729a.2b1cf7d6@aol.com>
Date: Mon, 02 Dec 2002 19:39:50 +0000
Message-ID: <28198.1038857990@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


>> customers that would pay for a service like GETS, or one based on MLPP, 
>> would certainly want accounting and so I'm not saying there isn't a need
>> for it.  but since IEPREP is broader in scope than these two systems,
>> we can't tie in a third mechanism that in certain cases plays no role.
>> for example, the land-line based system of GTPS (in the UK) has no
>> need for accounting. 
>> 
>> 
> [MAP] I'm at a loss to explain further to resolve your lack of being
> "convinced", since you say in the final paragraph above that you agree
> that accounting is needed for GETS and MLPP, which is the only point
> I've made.  What is it you don't agree with that I said?

well, to be precise, I stated ones "based on MLPP", and not MLPP itself.

ok, I think the mixup comes from me.  my previous posting had in mind
the requirements drafts of carlberg-ets-* that aim at general statements
that can be made.  this is why I stated (in the excerpt you omitted)
that I can add something to the carlberg-ets-telephony-* draft about
accounting.

on the other hand, I suspect that you have in mind the pierce-ieprep-
assured-service-req-* draft.  so we probably had a ships-in-the-night
email exchange.

-ken

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 15:19:21 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA07939
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 15:19:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2KLj829396
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 15:21:45 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2KLjv29393
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 15:21:45 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA07905
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 15:18:50 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2KI1v29149;
	Mon, 2 Dec 2002 15:18:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2KHav29126
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 15:17:36 -0500
Received: from [140.32.132.66] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA07782
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 15:14:40 -0500 (EST)
Received: from mail.nps.navy.mil by [140.32.132.66]
          via smtpd (for [132.151.6.1]) with ESMTP; Mon, 2 Dec 2002 12:13:18 -0800
Received: from ziggy ([131.120.179.220])
	by capella.nps.navy.mil (8.12.2/8.12.2) with ESMTP id gB2KE0bR006522;
	Mon, 2 Dec 2002 12:14:01 -0800 (PST)
Subject: Re: [Ieprep] Accounting
From: Rex Buddenberg <budden@nps.navy.mil>
To: RJ Atkinson <rja@extremenetworks.com>
Cc: ieprep <ieprep@ietf.org>
In-Reply-To: <F222D553-061A-11D7-960D-00039357A82A@extremenetworks.com>
References: <F222D553-061A-11D7-960D-00039357A82A@extremenetworks.com>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) 
Date: 02 Dec 2002 12:09:54 -0800
Message-Id: <1038859795.1219.27.camel@ziggy>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Authentication of data is a critical item for emergency services
networks.  If this isn't in lights in a requirements statement, it
should be.

Therefore key management has to go somewhere.  Agreed that it may be a
local item (and it's an end-end issue, not one for routers to worry
about) but from the customer's point of view, I'd think some coaching on
how to proceed would be a Good Thing to have ... to offset some of the
coaching I'd be getting from n vendors who are all in the mode of 'never
mind what your requirement is, we've got the solution'.

Given that key management is a necessary part of a Compleat Emergency
Services structure, where does it go on the tree?  Accounting management
sounds like as good an answer as any...


On Mon, 2002-12-02 at 09:24, RJ Atkinson wrote:
> 
> On Monday, Dec 2, 2002, at 12:12 America/Montreal, Rex Buddenberg wrote:
> > Ran,  are you binning key management under accounting?
> 
> Hmm.  No.  Had not occurred to me that they were related.
> Are they related ?
> 
> Ran
> 
> PS:	And I do think key management is its usual nightmare here, just as
> it is in other places where one wants to do authentication of stuff.
> 
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 15:48:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA09418
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 15:48:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2KoZn31359
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 15:50:35 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2KoZv31356
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 15:50:35 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA09384
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 15:47:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2KkRv31191;
	Mon, 2 Dec 2002 15:46:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2KjWv31142
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 15:45:32 -0500
Received: from dcament482.dc.deloitte.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA09034
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 15:42:35 -0500 (EST)
Received: by dcament482.dc.deloitte.com with Internet Mail Service (5.5.2653.19)
	id <XAXVYCPR>; Mon, 2 Dec 2002 15:45:23 -0500
Message-ID: <9259B650F8A90A489A6FA0630BA93FF4C21652@DCAMENT412.dc.deloitte.com>
From: "McCaffrey, Bryan (CA - Toronto)" <bmccaffrey@dc.com>
To: RJ Atkinson <rja@extremenetworks.com>, Mpierce1@aol.com
Cc: ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] Internet faults, VoIP demo, etc and others
Date: Mon, 2 Dec 2002 15:45:10 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Could end-to-end call admission control also be used (vs. just link-specific
controls) for "b" below? As far as I know, Cisco's implementation of CAC
uses the concept of a "call" described as a unit of bandwidth, 64Kbps at
present. 

-Bryan

-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com] 
Sent: Monday, December 02, 2002 11:17 AM
To: Mpierce1@aol.com
Cc: ieprep
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others



On Monday, Dec 2, 2002, at 10:30 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] So take this simple case and apply an emergency. (I'd like to
> see the performance last week right after the bombing in Kenya!)
>
> If 10 or 20 or whatever number of those "several thousands" of end
> users tried to make a VoIP call via that single 384K link, what would 
> have happened? It is my understanding that, at some point, the 
> overload would have caused either:
>
> 1. all calls to be of unacceptable quality
> 2. new call attempts would have been rejected

	The IP-layer doesn't have the concept of a "call", so (B) can't
happen 
at
the IP-layer.  Bill's SIP server is not on the access link (as I 
understand it),
so would not be subject to the emergency -- it would be able to process 
calls
normally without any MLPP.  (Bill please clarify these points.)

	If congestion on the last mile link occurred (which we don't know 
right now),
then it could ONLY be solved using link-specific deployed mechanisms 
(e.g. router
CoS like Sprint/Genuity described would be sufficient).  If congestion 
occurred,
then what would have happened is probably packet loss (after the packet 
buffers
on the access routers filled up).  How much packet loss occurred 
depends on the
depth of those buffers, the offered load, the link capacity, etc.  So 
we don't
know whether congestion occurred (yet, Bill should ask) or exactly what 
happened
if it occurred.

	It is entirely possible that the Nairobi VoIP user was in fact
having 
no
problems using his link during that situation.  As of this second, we 
don't
know (and Bill should ask).

> I and others have asked this question numerous times, but have not
> seen any hint of an answer, just claims that one should 
> "overprovision" to prevent this from happening. Clearly, 
> overprovisioning or adding capacity in real time can not solve the 
> problem in this situation. I'm looking for a concrete solution. I have 
> been lead to believe that there is none at present, thus the existance 
> of IEPREP.

Concrete existing solutions:
	1)	Deploy link-specific QoS.  For DSL see T1/ANSI/ITU-T
standards,
		for Ethernet see 802.1p, for cable modems see DOCSIS
standards.
	2) 	Deploy access router CoS (a la Sprint/Genuity product
offerings).
	     This is already widely available and is outside of IETF's scope

since
	     it is an implementation detail inside of a router.

Please STOP claiming that concrete solutions to the access link 
congestion
issue have not been offered.  It is a bold-faced lie and it is quite 
tiresome
at this point.

And please STOP claiming that IETF lacks mechanisms that DoD could use 
to
deploy MLPP internally.  DiffServ is on the standards track already with AF
and EF.  So that is another bold-faced lie.

It is just not productive when you keep repeating claims that are 
obviously untrue
to anyone reading the list's archives.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 15:56:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA09824
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 15:56:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2KwtC31735
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 15:58:55 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Kwtv31732
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 15:58:55 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA09804
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 15:55:59 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Kt2v31604;
	Mon, 2 Dec 2002 15:55:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Ks7v31545
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 15:54:07 -0500
Received: from seahorse.shentel.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA09596
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 15:51:11 -0500 (EST)
Received: from steves (ha30s390.d.shentel.net [204.111.30.134])
	by seahorse.shentel.net (8.11.6/8.11.6) with SMTP id gB2Kr3M27723;
	Mon, 2 Dec 2002 15:53:03 -0500
From: "Steve Silverman" <steves@shentel.net>
To: "Rex Buddenberg" <budden@nps.navy.mil>,
        "RJ Atkinson" <rja@extremenetworks.com>
Cc: "ieprep" <ieprep@ietf.org>
Subject: RE: [Ieprep] Accounting
Date: Mon, 2 Dec 2002 15:53:00 -0500
Message-ID: <KMEJKAPHBBCMCDDGMBLOKECDCBAA.steves@shentel.net>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook IMO, Build 9.0.2416 (9.0.2911.0)
In-reply-to: <1038859795.1219.27.camel@ziggy>
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106
Importance: Normal
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

I have to point out that if a call purports to be an emergency call,
it should be transmitted with or without
authentication.  I have pointed out the  need for enough accounting
info to detect misuse & DDOS attacks. But if the building is
burning, we do want to allow a kid to summon the appropriate
response even if he can't remember his password amidst the smoke or
other distractions.  Perhaps a limit on the number of high priority
calls from a location per minute?  And of course the call info is
saved for analysis.

Steve Silverman

> -----Original Message-----
> From: ieprep-admin@ietf.org
> [mailto:ieprep-admin@ietf.org]On Behalf Of
> Rex Buddenberg
> Sent: Monday, December 02, 2002 3:10 PM
> To: RJ Atkinson
> Cc: ieprep
> Subject: Re: [Ieprep] Accounting
>
>
> Authentication of data is a critical item for emergency services
> networks.  If this isn't in lights in a requirements statement, it
> should be.
>
> Therefore key management has to go somewhere.  Agreed
> that it may be a
> local item (and it's an end-end issue, not one for
> routers to worry
> about) but from the customer's point of view, I'd think
> some coaching on
> how to proceed would be a Good Thing to have ... to
> offset some of the
> coaching I'd be getting from n vendors who are all in the
> mode of 'never
> mind what your requirement is, we've got the solution'.
>
> Given that key management is a necessary part of a
> Compleat Emergency
> Services structure, where does it go on the tree?
> Accounting management
> sounds like as good an answer as any...
>
>
> On Mon, 2002-12-02 at 09:24, RJ Atkinson wrote:
> >
> > On Monday, Dec 2, 2002, at 12:12 America/Montreal, Rex
> Buddenberg wrote:
> > > Ran,  are you binning key management under accounting?
> >
> > Hmm.  No.  Had not occurred to me that they were related.
> > Are they related ?
> >
> > Ran
> >
> > PS:	And I do think key management is its usual
> nightmare here, just as
> > it is in other places where one wants to do
> authentication of stuff.
> >
> >
> > _______________________________________________
> > Ieprep mailing list
> > Ieprep@ietf.org
> > https://www1.ietf.org/mailman/listinfo/ieprep
>
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep
>
>


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 16:21:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11188
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 16:21:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2LNuk01233
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 16:23:56 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LNuv01230
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 16:23:56 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11085
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 16:21:00 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LMIv01027;
	Mon, 2 Dec 2002 16:22:18 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LJuv00766
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 16:19:56 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA10745
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 16:16:58 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id B2B7C67123; Mon,  2 Dec 2002 12:45:08 -0500 (EST)
Date: Mon, 2 Dec 2002 16:18:50 -0500
Subject: Re: [Ieprep] Re: Looking for governmental feedback as well
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Rex Buddenberg <budden@nps.navy.mil>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <1038851586.1224.21.camel@ziggy>
Message-Id: <A708E52E-063B-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 12:53 America/Montreal, Rex Buddenberg wrote:
> The notion of 'government owned networks and sites' is of little and
> decreasing value:
> 	- the most important reason is that in disaster relief type of
> operations the real workers are at the county and local level and none
> of those folks are their own ISPs.

s/none/few/

A small number of US localities have their own fibre-based MANs.
A rather larger number of Swedish localities have their own MANs.
Situations vary from country to country, region to region, etc.

IETF needs to maintain a global perspective and avoid US-centrism.

> The proper goal should be greater availability and survivability.
>
> The provisioning means is usable diverse alternate routes, and when
> appropriate, duplicated end systems.

All those seem reasonable to me.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 16:21:32 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11201
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 16:21:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2LNvL01248
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 16:23:57 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LNvv01245
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 16:23:57 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11091
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 16:21:01 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LMHv01008;
	Mon, 2 Dec 2002 16:22:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LJrv00762
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 16:19:53 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA10741
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 16:16:56 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id E9A0267115; Mon,  2 Dec 2002 12:45:05 -0500 (EST)
Date: Mon, 2 Dec 2002 15:05:55 -0500
Subject: Re: [Ieprep] IEprep in JP
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep@ietf.org
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <6d.36a09dd.2b1cf7bb@aol.com>
Message-Id: <77836EF1-0631-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 12:51 America/Montreal, Mpierce1@aol.com 
wrote:

> [MAP] Yes, I understand that IAA is an "IEPREP service" in the general 
> sense.

The general definition is the one IETF uses.

> What I meant is that it is not a service which can or does operate on 
> the basis that certain users are given authority to get a better 
> "priority" than others.

That is consistent with what I hear:
	IAA in Japan does not require that any users get preferred treatment.
	IAA does want a very high probability that the IAA traffic get 
delivered.

> My points are all related to a service with special authorized users.

Interesting, but here in IETF, IEprep does NOT necessarily mean such a 
service.

> [MAP] Please note that the "MLPP" that we're using as a model for an 
> IP equivalent is also a "global standard" (in ITU-T). It can be 
> applied in any country.

That ITU spec is not any sort of Internet Standard and this is not the 
PSTN,
so while amusing that spec is not applicable to actual INTERNET 
Emergency
Preparedness.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 16:21:34 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11215
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 16:21:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2LNxM01264
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 16:23:59 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LNxv01261
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 16:23:59 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11104
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 16:21:03 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LM9v00946;
	Mon, 2 Dec 2002 16:22:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LJov00742
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 16:19:50 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA10728
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 16:16:52 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 592D767114; Mon,  2 Dec 2002 12:45:05 -0500 (EST)
Date: Mon, 2 Dec 2002 15:00:20 -0500
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149A1@chntex02.is.dyncorp.com>
Message-Id: <AFBFBF0D-0630-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 12:56 America/Montreal, Gunn, Janet wrote:
> For these "concrete solutions to access link congestion issues", how 
> do you
> suggest we identify "emergency" traffic?

	For traffic originating on a site behind that congested access link, 
one
could configure such a local policy into the access routers (a la 
Sprint's
example web page).  There are a very wide range of policies that could 
be
configured into the routers.  For voice traffic, one could prioritise 
voice
based on combinations of (protocol/port number/IP address) if desired.  
For
traffic originating inside a given domain, access lists and other COTS 
methods
can be used to prevent unauthorised access from within that domain.  
ToS bits
can be used within a domain and on the access link between that domain 
and its
uplink ISP.  Using ToS bits in the IP header or Ethernet header is 
pretty common
for intra-domain QoS deployments existing today.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 16:21:51 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11243
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 16:21:51 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2LOG001311
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 16:24:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LOGv01308
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 16:24:16 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11170
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 16:21:19 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LMAv00964;
	Mon, 2 Dec 2002 16:22:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LJpv00751
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 16:19:51 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA10730
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 16:16:53 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 307FF67117; Mon,  2 Dec 2002 12:45:07 -0500 (EST)
Date: Mon, 2 Dec 2002 15:23:50 -0500
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep@ietf.org
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <29.326c7f94.2b1cf85e@aol.com>
Message-Id: <F8485C5C-0633-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 12:54 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] I understand fully that the IP-layer is unaware of "calls" or 
> "voice packets". That wasn't the point. If some higher layer function 
> doesn't limit the setup of new calls (2. above) when it knows that 
> there are already too many being stuffed through a finite size link, 
> then the IP-layer would be overwhelmed with packets which it can't 
> handle (1. above).

The IP-layer can be overwhelmed anyway since IP does not have any 
equivalent
to "call setup".  Even if SIP were informed (which it isn't) and had 
some way
to control the voice calls (which it doesn't since anyone on the 
Internet can
use any cooperating SIP server, not just the local one), other 
unrelated applications
like mail or ftp or http don't use SIP (and won't be using it in 
future).

Now one could use IP ToS bits to mark the packets from inside a domain 
for
the uplink router to prioritise, if the uplink router happened to have 
that
policy configured in it by the router's operator.  One could also put 
various
other policies into the router based on some combinations of (address, 
protocol, port),
if one did not want to use the IP ToS bits.  If the link is Ethernet, 
one could
use 802.1p.  If the link is something else, then one could use other 
link-specific
mechanisms (e.g. ATM has PSTN-like calls).

> [MAP] So I take that as an agreement with what I said, that in the 
> case of congestion, a lot of packets would be lost.

No.  The number of packets lost would vary with (offered load, link 
capacity,
traffic pattern, packet memory depth on each side of the link).  In 
many cases
there is minimal or no congestion, because the router buffer is deep 
enough
to handle a trans-oceanic RTT, which in turn gives TCP or SCTP enough 
time
to reduce the transmission rate on existing sessions, which then leaves
enough room for the new traffic to be carried on the link.

> [MAP] I'll repeat the question which I don't believe I've gotten any 
> concrete answer to: In the case above of the 10 callers from Nairobi, 
> if one of them was authorized to make emergency calls and the other 9 
> weren't, what specific standardized mechanism (not products) could be 
> used to provide the required service across the congested 384 kbit/s 
> satellite link? (The required service is that their call gets set up 
> and they can talk.) If you don't believe that 10 users would cause 
> congestion, then make it 20 or however many of the several thousand it 
> would take.
>
> I would find it hard to believe that there are no standardized 
> solutions to this. You're right, maybe there is no need for this WG, 
> but I haven't heard the solution yet.

One could use IP ToS bits, link-specific ToS bits (802.1p in the case 
of Ethernet),
or some combination of (address/protocol/port) to tell the access 
router which
traffic to prefer from the site connected to that access router.  
Preference can be
implemented using the CoS mechanism that Sprint is using, various 
flavours of which
are available from many vendors.  Whether that policy (either to obey 
ToS bits or to sort
based on the address/protocol/port) is configured into that router is a 
question
for whoever  owns/operates that access router.

If the offered load of preferred traffic exceeds link capacity, then 
packets
still will be dropped.

Just don't be expecting ISPs to deploy any of this stuff inside their 
backbones,
because that would create a huge DDOS vulnerability for them (thus 
reducing the
probability of "call" completion well below what is the case today).  
It is both
cheaper and more effective to overprovision backbones.

Ran



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 16:24:32 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11311
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 16:24:30 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2LQtd01428
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 16:26:55 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LQtv01425
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 16:26:55 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11306
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 16:23:59 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LM6v00930;
	Mon, 2 Dec 2002 16:22:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LJov00747
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 16:19:50 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA10729
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 16:16:52 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 81DF467116; Mon,  2 Dec 2002 12:45:06 -0500 (EST)
Date: Mon, 2 Dec 2002 15:08:58 -0500
Subject: Re: [Ieprep] identification of packets
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149A0@chntex02.is.dyncorp.com>
Message-Id: <E48DC8FC-0631-11D7-960D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 12:49 America/Montreal, Gunn, Janet wrote:
> What is being asked is for IETF to help identify emergency traffic 
> WHEN SOME
> ENTITY is addressing congestion on the access links.

	Interesting.

	IP ToS bits can be used to identify preferred traffic inside a domain
(and on that domain's access links to other sites or its ISPs).  
Enterprise users
often do that today (or use the 802.1p Ethernet precedence bits or 
other link-specific
methods for the same end).

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 16:36:21 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12260
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 16:36:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2LciU02689
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 16:38:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Lciv02686
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 16:38:44 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12188
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 16:35:50 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LaVv02022;
	Mon, 2 Dec 2002 16:36:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2La1v01982
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 16:36:01 -0500
Received: from imo-m10.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA11998
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 16:33:06 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m10.mx.aol.com (mail_out_v34.13.) id c.129.1c5fbf0f (4394);
	Mon, 2 Dec 2002 16:35:45 -0500 (EST)
Message-ID: <129.1c5fbf0f.2b1d2c31@aol.com>
Date: Mon, 2 Dec 2002 16:35:45 EST
To: rohan@cisco.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_129.1c5fbf0f.2b1d2c31_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Subject: [Ieprep] IEPREP Charter
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_129.1c5fbf0f.2b1d2c31_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 11/26/2002 4:03:42 PM Eastern Standard Time, 
rohan@cisco.com writes:


> I don't feel there is anything especially sinister about dealing with 
> military networking requirements provided the participants understand 
> that the military will have to pay for their services just like 
> everyone else. Certainly many of the requirements are similar, but I 
> really don't care as long as we get a decision from the ADs or a 
> consensus from the WG.  Whatever the outcome, I think the charter needs 
> to get clarified.  My proposed text (for both possible outcomes) is 
> below.
> 


<snip>

[MAP] I would like to see changes to the IEPREP charter to make the following 
points very clear:

1. Scope of work includes both a.) civilian emergency access to 
telecommunications facilities by individuals as authorized by a government 
(e.g., GETS in the US) and b.) military priority systems within essentially 
closed networks (e.g. MLPP). The WG explicitly accepts governments and 
military organizations as legitimate users of IP and of IETF standards and as 
legitimate participants in the standards process (through the individual 
participants they support).

2. IEPREP is to examine ways to provide preferential treatment to calls 
(within the IP environment) only in cases in which congestion occurs. (It 
will not engage in discussions of whether or not congestion can occur. It is 
presumed that it can.)

3. The situation of most concern is a congestion case that would be caused by 
massive failures in the infrastructure, caused by a natural or man-made 
disaster of a magnitude which has not ever occurred (and we hope never does). 
(That means that the WG may consider evensts like 9/11, the Baltimore Train 
tunnel, calls from Nairobi, etc. as interesting case studies, but not as 
proof that there is no problem.)

4. Since over-provisioning is an operational choice made by a carrier and can 
not be used in all cases, it is not to be discussed in IEPREP.

5. The primary results of IEPREP are "requirements" to give to other WGs to 
identify where changes are needed in protocols, etc. Nothing out of IEPREP is 
a mandate on any vendor or carrier to provide anything. (WIth this 
understanding in the Charter, there should be no need for further concerns to 
be raised that the IETF should not require carriers to provide this or that 
functionality.)

By clearly stating these points and thereby eliminating the ongoing 
discussions related to several non-issues, the WG should be able to focus on 
the real issues that it was created to address (or clearly identify that 
there are already solutions which can be applied so the WG can be disbanded).

Mike Pierce
Artel



--part1_129.1c5fbf0f.2b1d2c31_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 11/26/2002 4:03:42 PM Eastern Standard Time, rohan@cisco.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">I don't feel there is anything especially sinister about dealing with 
<BR>military networking requirements provided the participants understand 
<BR>that the military will have to pay for their services just like 
<BR>everyone else. Certainly many of the requirements are similar, but I 
<BR>really don't care as long as we get a decision from the ADs or a 
<BR>consensus from the WG. &nbsp;Whatever the outcome, I think the charter needs 
<BR>to get clarified. &nbsp;My proposed text (for both possible outcomes) is 
<BR>below.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>&lt;snip&gt;
<BR>
<BR>[MAP] I would like to see changes to the IEPREP charter to make the following points very clear:
<BR>
<BR>1. Scope of work includes both a.) civilian emergency access to telecommunications facilities by individuals as authorized by a government (e.g., GETS in the US) and b.) military priority systems within essentially closed networks (e.g. MLPP). The WG explicitly accepts governments and military organizations as legitimate users of IP and of IETF standards and as legitimate participants in the standards process (through the individual participants they support).
<BR>
<BR>2. IEPREP is to examine ways to provide preferential treatment to calls (within the IP environment) only in cases in which congestion occurs. (It will not engage in discussions of whether or not congestion can occur. It is presumed that it can.)
<BR>
<BR>3. The situation of most concern is a congestion case that would be caused by massive failures in the infrastructure, caused by a natural or man-made disaster of a magnitude which has not ever occurred (and we hope never does). (That means that the WG may consider evensts like 9/11, the Baltimore Train tunnel, calls from Nairobi, etc. as interesting case studies, but not as proof that there is no problem.)
<BR>
<BR>4. Since over-provisioning is an operational choice made by a carrier and can not be used in all cases, it is not to be discussed in IEPREP.
<BR>
<BR>5. The primary results of IEPREP are "requirements" to give to other WGs to identify where changes are needed in protocols, etc. Nothing out of IEPREP is a mandate on any vendor or carrier to provide anything. (WIth this understanding in the Charter, there should be no need for further concerns to be raised that the IETF should not require carriers to provide this or that functionality.)
<BR>
<BR>By clearly stating these points and thereby eliminating the ongoing discussions related to several non-issues, the WG should be able to focus on the real issues that it was created to address (or clearly identify that there are already solutions which can be applied so the WG can be disbanded).
<BR>
<BR>Mike Pierce
<BR>Artel
<BR>
<BR></FONT></HTML>

--part1_129.1c5fbf0f.2b1d2c31_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 16:41:06 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12461
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 16:41:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2LhTx02963
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 16:43:29 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LhSv02960
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 16:43:28 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12438
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 16:40:34 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Lf2v02843;
	Mon, 2 Dec 2002 16:41:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Le3v02776
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 16:40:03 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12321
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 16:37:07 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAGN53>; Mon, 2 Dec 2002 16:37:29 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149A7@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>
Cc: ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
Date: Mon, 2 Dec 2002 16:37:20 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

OK,
Then take it a step further

Is there a consistent way to identify emergency traffic across several
different "entities that are addressing congestion on the access links"?
Does this have to be worked out on an entity-by-entity basis, or is there
the possibility of a "standard" (de facto or officially sanctioned) way of
marking such traffic?  If so that would take us a long way in the right
direction.  The appropriate treatment of the marked traffic would still have
to be negotiated on an entity-by-entity basis. 

And, is there a way to pass that "identification as emergency traffic" from
the access links at the originating node to the access links at the
terminating node (if the terminating node is in another domain)? (From what
I have read here, and ToS bits based marking would be lost in the
intervening ISP.)

-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Monday, December 02, 2002 3:09 PM
To: Gunn, Janet
Cc: ieprep
Subject: Re: [Ieprep] identification of packets



On Monday, Dec 2, 2002, at 12:49 America/Montreal, Gunn, Janet wrote:
> What is being asked is for IETF to help identify emergency traffic 
> WHEN SOME
> ENTITY is addressing congestion on the access links.

	Interesting.

	IP ToS bits can be used to identify preferred traffic inside a
domain
(and on that domain's access links to other sites or its ISPs).  
Enterprise users
often do that today (or use the 802.1p Ethernet precedence bits or 
other link-specific
methods for the same end).

Ran
rja@extremenetworks.com
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 16:41:15 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12476
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 16:41:15 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2Lhdq02986
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 16:43:39 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Lhcv02983
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 16:43:38 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12443
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 16:40:44 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Ld3v02716;
	Mon, 2 Dec 2002 16:39:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Lc0v02666
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 16:38:00 -0500
Received: from imo-m03.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12147
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 16:35:06 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m03.mx.aol.com (mail_out_v34.13.) id 5.1ac.ce9e43a (4394);
	Mon, 2 Dec 2002 16:37:24 -0500 (EST)
Message-ID: <1ac.ce9e43a.2b1d2c94@aol.com>
Date: Mon, 2 Dec 2002 16:37:24 EST
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
To: sob@harvard.edu, rja@extremenetworks.com, woody@pch.net
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_1ac.ce9e43a.2b1d2c94_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_1ac.ce9e43a.2b1d2c94_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/2/2002 1:42:11 PM Eastern Standard Time, 
sob@harvard.edu writes:


> > 2. new call attempts would have been rejected
> 
> no- the congested link would not cause the SIP proxies to reject calls
> since they have no knowledge of the congestion
> 
> 


[MAP] Although it must be technically possible if the mechanisms were 
defined, that is not what I was suggesting. I meant that whatever entity was 
setting up calls (SIP proxy) would have some knowledge of how many it could 
attempt to setup simultaneously, at least based on the provisioned capacity. 
(The circuit switched world had it easy in this regard.) When the SIP proxy 
reached that limit, it would stop trying to set up more (since they probably 
wouldn't work anyway).

I had approached this work with the (naive?) idea that such CAC functions 
already existed. That is why my draft-pierce-ieprep-pref-treat-examples 
(previously in SIPPING) suggested "higher call acceptance limits" for 
priority calls as one possible method of providing "preferential treatment".

If you're telling me that no such CAC is possible in a SIP proxy, then I 
believe it is impossible to do what we are attempting (VoIP). I don't believe 
that is the case, though.

I have learned, from some other comments, that this lack of being able to 
respond to congestion (by routing changes or blocking calls) is a significant 
problem not only for priority traffic.

Mike Pierce
Artel

--part1_1ac.ce9e43a.2b1d2c94_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/2/2002 1:42:11 PM Eastern Standard Time, sob@harvard.edu writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">&gt; 2. new call attempts would have been rejected
<BR>
<BR>no- the congested link would not cause the SIP proxies to reject calls
<BR>since they have no knowledge of the congestion
<BR>
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>[MAP] Although it must be technically possible if the mechanisms were defined, that is not what I was suggesting. I meant that whatever entity was setting up calls (SIP proxy) would have some knowledge of how many it could attempt to setup simultaneously, at least based on the provisioned capacity. (The circuit switched world had it easy in this regard.) When the SIP proxy reached that limit, it would stop trying to set up more (since they probably wouldn't work anyway).
<BR>
<BR>I had approached this work with the (naive?) idea that such CAC functions already existed. That is why my draft-pierce-ieprep-pref-treat-examples (previously in SIPPING) suggested "higher call acceptance limits" for priority calls as one possible method of providing "preferential treatment".
<BR>
<BR>If you're telling me that no such CAC is possible in a SIP proxy, then I believe it is impossible to do what we are attempting (VoIP). I don't believe that is the case, though.
<BR>
<BR>I have learned, from some other comments, that this lack of being able to respond to congestion (by routing changes or blocking calls) is a significant problem not only for priority traffic.
<BR>
<BR>Mike Pierce
<BR>Artel</FONT></HTML>

--part1_1ac.ce9e43a.2b1d2c94_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 16:41:16 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12478
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 16:41:16 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2LhdM03001
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 16:43:39 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Lhdv02998
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 16:43:39 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12444
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 16:40:44 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Ld6v02732;
	Mon, 2 Dec 2002 16:39:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Lcfv02681
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 16:38:41 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12183
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 16:35:45 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 2 Dec 2002 16:38:00 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120216381601783
 ; Mon, 02 Dec 2002 16:38:16 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YDVM5FGX>; Mon, 2 Dec 2002 16:38:03 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E779D8@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Mpierce1@aol.com '" <Mpierce1@aol.com>,
        "'rja@extremenetworks.com '" <rja@extremenetworks.com>
Cc: "'ieprep@ietf.org '" <ieprep@ietf.org>
Subject: RE: [Ieprep] Internet faults, VoIP demo, etc and others
Date: Mon, 2 Dec 2002 16:38:14 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


On Monday, Dec 2, 2002, at 10:30 America/Montreal, Mpierce1@aol.com 
wrote: 
> [MAP] So take this simple case and apply an emergency. (I'd like to 
> see the performance last week right after the bombing in Kenya!) 
> 
> If 10 or 20 or whatever number of those "several thousands" of end 
> users tried to make a VoIP call via that single 384K link, what would 
> have happened? It is my understanding that, at some point, the 
> overload would have caused either: 
> 
> 1. all calls to be of unacceptable quality 
> 2. new call attempts would have been rejected 

[MAP] I'll repeat the question which I don't believe I've gotten any
concrete answer to: In the case above of the 10 callers from Nairobi, if
one of them was authorized to make emergency calls and the other 9
weren't, what specific standardized mechanism (not products) could be
used to provide the required service across the congested 384 kbit/s
satellite link? (The required service is that their call gets set up and
they can talk.) If you don't believe that 10 users would cause
congestion, then make it 20 or however many of the several thousand it
would take. 

Mike,

The answers to your question depends upon the assumptions you make (i.e.,
problem context).  Do you assume your 10 or however many users are just
random folks who happen to be using the same access link owned by an
arbitrary ISP?  Then your answer is very likely to be "the call quality for
all users is going to be poor".  

Let's us examine why...well, we could try to identify emergency users with
SIP signaling and try to use "call admission control" in the telco sense.
Fine except your 10 users might use 10 different SIP service providers
completely unconnected to the ISP who happens to be providing the access
link.  Those SIP providers certainly have no way of knowing what other SIP
service providers are doing nor do they track bandwidth utilization across
some random link (they have no way to do this) hence they can't provide a
"call admission control".  So, you can then assume all 10 users are allowed
to setup calls.  Your access link provider is quite likely to be just moving
packets.

Now call quality has to do with lots of things including jitter and loss.
Trying to stuff more voice packets (UDP) into a slow link is going to result
in drops on both ends of the tiny pipe.  There is no way to identify
emergency traffic packets.  This is because there are 10,000 or so ISPs and
it is almost always true that all packets are created equal at the transport
layer.  Trying to change this for an arbitrary ISP over an arbitrary link is
just about impossible.  Remember identification mechanisms like CoS are the
exception, not the norm and the assumption that this is just some arbitrary
connection means no CoS is available.

So, with these set of assumptions, the answer is the call quality will be
poor for all until some people hang up.

But, under a different set of assumptions, the answer is different.  Assume
all 10 users have a local soft PBX that provides a "call admission control"
for all off net calls.  It also tracks who's got priority and can even
preempt lower priority calls.  Then 10 users try to setup but only the
emergency worker and the max number the access link can support are allowed
to complete.  This could actually support Assured Service.  

See how important is it to define the context?

Kimberly

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 16:44:01 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12617
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 16:44:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2LkOQ03090
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 16:46:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LkOv03087
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 16:46:24 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12592
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 16:43:30 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Lg2v02914;
	Mon, 2 Dec 2002 16:42:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Lfev02883
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 16:41:40 -0500
Received: from [140.32.132.66] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA12374
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 16:38:44 -0500 (EST)
Received: from mail.nps.navy.mil by [140.32.132.66]
          via smtpd (for [132.151.6.1]) with ESMTP; Mon, 2 Dec 2002 13:37:20 -0800
Received: from ziggy ([131.120.179.220])
	by capella.nps.navy.mil (8.12.2/8.12.2) with ESMTP id gB2Lc0bR011207;
	Mon, 2 Dec 2002 13:38:00 -0800 (PST)
Subject: RE: [Ieprep] Accounting
From: Rex Buddenberg <budden@nps.navy.mil>
To: Steve Silverman <steves@shentel.net>
Cc: RJ Atkinson <rja@extremenetworks.com>, ieprep <ieprep@ietf.org>
In-Reply-To: <KMEJKAPHBBCMCDDGMBLOKECDCBAA.steves@shentel.net>
References: <KMEJKAPHBBCMCDDGMBLOKECDCBAA.steves@shentel.net>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) 
Date: 02 Dec 2002 13:33:54 -0800
Message-Id: <1038864835.1224.70.camel@ziggy>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

On Mon, 2002-12-02 at 12:53, Steve Silverman wrote:
> I have to point out that if a call purports to be an emergency call,
> it should be transmitted with or without
> authentication. 

Agreed; failure to authenticate is not a reason for denial of service;
it's a means for the receiver to judge the authenticity of the data.  

But if I'm going to build a neighborhood internet that my neighborhood
police (among many others) uses as it's comms system, I want those cops
to be able to know that they are talking with each other, not some
imposter.  I've bored far too many holes in the ocean on false alarms...

Quibling aside, an authentication mechanism needs to be part of the
infrastructure ... where do we say that?



As a data point, the Common Alerting Protocol working group is working
out this problem.  There is a plethora of warning systems out there --
lots of stovepipes.  One for emergency weather, another for attack
warning, ... and many are heavily media dependent -- feeds to the TV
system, etc.  CAP is working on the media-independent (and comms
independent, for that matter) data element standardization problem so
that warnings (and their modifications and cancellations) can be
distributed.  
	- authentication is a core requirement and recognized as such.  CAP is
XML-based so adopting a digital signature approach is fairly
straightforward as far as the protocols are concerned (Ran's point about
messy PKI administration acknowledged).  
	- anyone on the internet can originate a CAP message ... just pirate
the format and fill in the blanks (indeed the test version up now
encourages you to do just that).  No different than calling the fire
department on PSTN with a false alarm.  

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 16:55:20 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA13426
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 16:55:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2LvhD03597
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 16:57:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Lvhv03594
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 16:57:43 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA13411
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 16:54:49 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LtIv03513;
	Mon, 2 Dec 2002 16:55:18 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2LsCv03466
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 16:54:12 -0500
Received: from [209.22.88.17] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id QAA13201
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 16:51:17 -0500 (EST)
Received: from mtassp3.ncr.disa.mil by [209.22.88.17]
          via smtpd (for ietf-mx.ietf.org [132.151.6.1]) with SMTP; 2 Dec 2002 21:47:29 UT
Received: by mtassp3.ncr.disa.mil with Internet Mail Service (5.5.2653.19)
	id <XKBZC2Y4>; Mon, 2 Dec 2002 16:54:05 -0500
Message-ID: <7F18415E4D63CB45BB9B3A591F68D12D5B3DA4@emshqs1.ncr.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'RJ Atkinson'" <rja@extremenetworks.com>, Mpierce1@aol.com
Cc: ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: Looking for governmental feedback as well
Date: Mon, 2 Dec 2002 16:54:01 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>



-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Monday, December 02, 2002 10:51 AM
To: Mpierce1@aol.com
Cc: ieprep
Subject: Re: [Ieprep] Re: Looking for governmental feedback as well
Ran,

>Congestion in access links is of course routine, but is not something 
>that can be solved by an ISP since the access customer controls that link.

I still think congestion in access links needs to be resolved. Below is an
analogy I always think of when it comes to congestion.

Let's say an ambulance or a fire truck that responds to a 911 call during
rush hours. It needs to get on the Beltway to get to the emergency location.
And the Beltway is like a parking lot at that time. Everyone on the road
(including you and I if we happen to be on it) has to make room for the
ambulance or the fire truck to get through. I don't mind a little
inconvenience at all because it could save someone's life. Can you imagine
what happens if there is no established mechanism in place for those
emergency vehicles?

When those 911 calls are originated from our home or our relatives', we
would definitely have different perspectives.

Regards,

An
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 17:16:40 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA14750
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 17:16:40 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2MJ4W05682
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 17:19:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2MJ4v05679
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 17:19:04 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA14728
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 17:16:09 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2MF3v05466;
	Mon, 2 Dec 2002 17:15:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2ME3v05428
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 17:14:03 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA14470
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 17:11:08 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id BFEBD67106; Mon,  2 Dec 2002 13:39:23 -0500 (EST)
Date: Mon, 2 Dec 2002 17:13:55 -0500
Subject: Re: [Ieprep] identification of packets
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149A7@chntex02.is.dyncorp.com>
Message-Id: <5924E36C-0643-11D7-9562-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 16:37 America/Montreal, Gunn, Janet wrote:
> Is there a consistent way to identify emergency traffic across several
> different "entities that are addressing congestion on the access 
> links"?
> Does this have to be worked out on an entity-by-entity basis, or is 
> there
> the possibility of a "standard" (de facto or officially sanctioned) 
> way of
> marking such traffic?

I think that it needs to be worked out on an entity-by-entity basis,
not because of lack of "standards", but because a lot of entities
aren't (today) willing to offer inter-domain QoS as a service at any 
price.
And in an IETF context, feature deployment is a local policy matter for
the network operator, not something that IETF would ever try to dictate.

If an entity doesn't offer inter-domain QoS as a service at all,
then it doesn't care how packets are labeled.

A digression.  There once was an ISP that offered a "Gold" service
to its customers in place of the "regular" service.  (I did not work
for this ISP.)  The implementation was reportedly as follows:
	1) Gold customers paid 50% more than regular customers for the
		same sized pipe.
	2) If a Gold customer felt during any billing period that they had
		not received good treatment, then they could call up the ISP
		and ask for a reduction down to the "regular" price for that
		billing period.

Most gold customers happily paid the extra money every month.  They
were happy because their packets did not get dropped.  The regular
customers were happy because their packets did not get dropped.  And
the ISP's shareholders were happy because the overall margins increased.

> And, is there a way to pass that "identification as emergency traffic" 
> from
> the access links at the originating node to the access links at the
> terminating node (if the terminating node is in another domain)? (From 
> what
> I have read here, and ToS bits based marking would be lost in the
> intervening ISP.)

I don't know of any practical-to-deploy way to do that across arbitrary 
ISP
boundaries -- only because whether ToS bits are zeroed or not is 
properly a local
operator policy decision (and deployment policies are not within IETF's 
scope).
It is not a technical problem, just a contractual one (where the 
contract
would say bit pattern FOO is not zeroed by the ISP).  One might be able 
to work out
contracts with specific ISPs or some kind of multi-lateral contract to 
not zero
IP ToS bits, at least on particular links.   Any road, that necessarily 
would be
a business issue or a contractual issue, rather than a technical issue 
or standards
issue.

An important consideration is that during times of stress, more than a 
usual
amount of path change over time is likely (as the routing protocols 
route
traffic around any damage).  So if one wanted to have high probability 
of
ToS bit markings not being zeroed end-to-end, one would need to have 
contracts
with a whole lot of operators.

I think this takes us back to the perspective where the best approach 
would
be for NCS to contract with operators to deliver packets within some 
sort of
error bounds.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 17:33:48 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15765
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 17:33:48 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2MaCZ06764
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 17:36:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2MaCv06761
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 17:36:12 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15745
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 17:33:17 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2MZ4v06720;
	Mon, 2 Dec 2002 17:35:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2MYVv06650
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 17:34:31 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15652
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 17:31:37 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 2 Dec 2002 17:33:58 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120217341530968
 for <ieprep@ietf.org>; Mon, 02 Dec 2002 17:34:15 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YDVM5KFA>; Mon, 2 Dec 2002 17:34:02 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E779DA@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
Date: Mon, 2 Dec 2002 17:34:12 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain
Subject: [Ieprep] SIP requirements draft after WG last call
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

"Requirements for Resource Priority Mechanisms for the Session Initiation
Protocol" draft has completed WG last call.  
  
Henning will submit a new version today that reflects comments.  Please
review and direct any last minute issues to Henning (within 24 hours).

The chairs will then ask the ADs to publish the draft as an informational
RFC and also ask that it be submitted for consideration by the SIPPING WG.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 17:39:38 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15975
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 17:39:38 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2Mg2k07584
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 17:42:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Mg2v07581
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 17:42:02 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15932
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 17:39:06 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Mc0v07416;
	Mon, 2 Dec 2002 17:38:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2Mb6v06982
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 17:37:06 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15803
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 17:34:12 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id CF11367106; Mon,  2 Dec 2002 14:02:23 -0500 (EST)
Date: Mon, 2 Dec 2002 17:36:55 -0500
Subject: Re: [Ieprep] Re: Looking for governmental feedback as well
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: "Nguyen, An" <nguyena@ncs.gov>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <7F18415E4D63CB45BB9B3A591F68D12D5B3DA4@emshqs1.ncr.disa.mil>
Message-Id: <8FB46E2C-0646-11D7-9562-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 16:54 America/Montreal, Nguyen, An wrote:
> I still think congestion in access links needs to be resolved.

I agree that potential congestion in access links is an issue, but we 
have
also discussed ways to address them, such as:
	- deploying CoS queuing in routers connected to access links
		(e.g. Sprint's CoS service offering)
	- deploying intra-domain IP ToS bit markings to ensure that
		a given site's access router has lower drop probability
		for preferred traffic that is destined for the ISP uplink
		from that site
	- deploying link-layer ToS mechanisms (e.g. 802.1p, others)

Mechanisms that live entirely inside a router (e.g. queuing disciplines)
is outside IETF scope.  DiffServ is already on the IETF standards-track
and covers use of the IP ToS bits.  Link-layers are outside IETF scope
and some (e.g. Ethernet) already have the standards specified (e.g. 
802.1p).

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 18:16:40 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA16934
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 18:16:40 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2NJ4N09715
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 18:19:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2NJ4v09712
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 18:19:04 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA16931
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 18:16:09 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2NI2v09640;
	Mon, 2 Dec 2002 18:18:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2NHnv09591
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 18:17:49 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA16905
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 18:14:54 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB2NBVo23451;
	Mon, 2 Dec 2002 15:11:31 -0800 (PST)
Date: Mon, 2 Dec 2002 15:11:31 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: RJ Atkinson <rja@extremenetworks.com>
cc: Mpierce1@aol.com, ieprep <ieprep@ietf.org>
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
In-Reply-To: <7248D878-0611-11D7-960D-00039357A82A@extremenetworks.com>
Message-ID: <Pine.GSO.4.44.0212021504210.23322-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Mon, 2 Dec 2002, RJ Atkinson wrote:
    > Bill's SIP server is not on the access link (as I understand it),
    > so would not be subject to the emergency -- it would be able to process
    > calls normally without any MLPP.  (Bill please clarify these
    > points.)

Currently there's just one server in the system, which we don't regard as
being a production condition, of course.  That server is on a pair of
OC12s, 1ms-2ms off the PAIX, and that link typically runs about about
1%-1.5% utilization, which is fairly typical for big corporate networks.

When we switch to the hardened production servers, of which there are
currently eighteen, we'll have considerably more capacity...  Each of
those is actually within an exchange, and has order of 2gb/sec-5gb/sec of
capacity, typically running at about 0.1% utilization.

    > It is entirely possible that the Nairobi VoIP user was in fact
    > having no problems using his link during that situation.  As of this
    > second, we don't know (and Bill should ask).

He shares that bandwidth with so many other people that the "fairness"
doesn't really impact one flow versus another out of his particular
tail-circuit.  So no, he doesn't notice any impact on his other
performance when he's using his phone.  Or was that not the question?

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 18:17:13 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA16962
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 18:17:13 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB2NJb909740
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 18:19:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2NJbv09737
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 18:19:37 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA16948
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 18:16:42 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2NH4v09532;
	Mon, 2 Dec 2002 18:17:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB2NGMv09507
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 18:16:22 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA16852
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 18:13:27 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 2 Dec 2002 18:15:41 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120218155816444
 ; Mon, 02 Dec 2002 18:15:58 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YDVQTMMC>; Mon, 2 Dec 2002 18:17:27 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E779DC@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson '" <rja@extremenetworks.com>,
        "'Gunn, Janet '" <Janet.Gunn@dyncorp.com>
Cc: "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
Date: Mon, 2 Dec 2002 18:15:57 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

On Monday, Dec 2, 2002, at 16:37 America/Montreal, Gunn, Janet wrote:
> Is there a consistent way to identify emergency traffic across several
> different "entities that are addressing congestion on the access 
> links"?
> Does this have to be worked out on an entity-by-entity basis, or is 
> there
> the possibility of a "standard" (de facto or officially sanctioned) 
> way of
> marking such traffic?

Ran wrote:
>I think that it needs to be worked out on an entity-by-entity basis,
>not because of lack of "standards", but because a lot of entities
>aren't (today) willing to offer inter-domain QoS as a service at any 
>price. And in an IETF context, feature deployment is a local policy 
>matter for the network operator, not something that IETF would ever try 
>to dictate.

I think Janet's question is excellent.  While it is true that "entities that
are addressing congestion on the access link" may use different methods,
this doesn't stop the hope that in many cases a common configuration of
*user software* is possible.  It doesn't have to imply interprovider QoS
either.

Nor does it have to be IETF "dictates", why not simply a BCP telling of this
common configuration.  (A common configuration would allow the same user
software to be used on a variety of networks, should default to best effort
if special stuff is not supported, and provide for enhanced interoperability
among emergency workers all using this common configuration.)  But these
types of ideas need to be made concrete.  Best practices for codecs may fall
into this category. Rohan had a creative idea that he posted last week.  His
idea falls into this category too.

Kimberly

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 19:29:46 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA19293
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 19:29:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB30WCP14303
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 19:32:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB30WCv14300
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 19:32:12 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA19285
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 19:29:15 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB30V8v14251;
	Mon, 2 Dec 2002 19:31:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB30UHv14204
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 19:30:17 -0500
Received: from [140.32.132.66] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA19252
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 19:27:21 -0500 (EST)
Received: from mail.nps.navy.mil by [140.32.132.66]
          via smtpd (for [132.151.6.1]) with ESMTP; Mon, 2 Dec 2002 16:25:58 -0800
Received: from ziggy ([131.120.179.220])
	by capella.nps.navy.mil (8.12.2/8.12.2) with ESMTP id gB30QcbR018486;
	Mon, 2 Dec 2002 16:26:38 -0800 (PST)
Subject: RE: [Ieprep] Accounting
From: Rex Buddenberg <budden@nps.navy.mil>
To: Steve Silverman <steves@shentel.net>
Cc: RJ Atkinson <rja@extremenetworks.com>, ieprep <ieprep@ietf.org>
In-Reply-To: <KMEJKAPHBBCMCDDGMBLOKECDCBAA.steves@shentel.net>
References: <KMEJKAPHBBCMCDDGMBLOKECDCBAA.steves@shentel.net>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) 
Date: 02 Dec 2002 16:22:31 -0800
Message-Id: <1038874954.1219.88.camel@ziggy>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

I'm firmly convinced that the reason World War Three never happened is
that nobody could properly authenticate the mission order.  

On Mon, 2002-12-02 at 12:53, Steve Silverman wrote:
> I have to point out that if a call purports to be an emergency call,
> it should be transmitted with or without
> authentication.  
> >
> 
> 


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 19:37:37 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA19475
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 19:37:37 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB30e3X15226
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 19:40:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB30e3v15223
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 19:40:03 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA19451
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 19:37:05 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB30d1v15098;
	Mon, 2 Dec 2002 19:39:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB30cfv15074
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 19:38:41 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA19419
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 19:35:45 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.126])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 2E9E47A0A; Mon,  2 Dec 2002 16:38:30 -0800 (PST)
Date: Mon, 2 Dec 2002 19:38:29 -0500
Subject: Re: [Ieprep] identification of packets
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep '" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E779DC@mcl-its-exs01.mail.saic.com>
Message-Id: <8AFD594F-0657-11D7-960F-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 18:15 America/Montreal, King, Kimberly S. 
wrote:
> I think Janet's question is excellent.

Agreed.

> While it is true that "entities that
> are addressing congestion on the access link" may use different 
> methods,
> this doesn't stop the hope that in many cases a common configuration of
> *user software* is possible.  It doesn't have to imply interprovider 
> QoS
> either.

Hmm.

What do you mean by "common configuration of user software" ?

One interpretation could be "common CLI syntax".  The IESG recently
chose not to charter an XML Configuration WG that might have worked
on common syntax, but for now the syntax for configuring OSPF (or
anything else) varies significantly from one vendor to another.  So
I'm thinking you mean something different maybe.

> Nor does it have to be IETF "dictates", why not simply a BCP telling 
> of this
> common configuration.  (A common configuration would allow the same 
> user
> software to be used on a variety of networks, should default to best 
> effort
> if special stuff is not supported, and provide for enhanced 
> interoperability
> among emergency workers all using this common configuration.)  But 
> these
> types of ideas need to be made concrete.  Best practices for codecs 
> may fall
> into this category. Rohan had a creative idea that he posted last 
> week.  His
> idea falls into this category too.

Please try to restate the above paragraph.  I'm sure I am not following
the intended meaning of the above paragraph.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 20:06:04 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20194
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 20:06:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB318Uj16862
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 20:08:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB318Uv16859
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 20:08:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20176
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 20:05:33 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3177v16387;
	Mon, 2 Dec 2002 20:07:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB316ev16158
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 20:06:40 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20118
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 20:03:43 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.126])
	by smtp1.extremenetworks.com (Postfix) with ESMTP id 5C9797A0A
	for <ieprep@ietf.org>; Mon,  2 Dec 2002 17:06:29 -0800 (PST)
Date: Mon, 2 Dec 2002 20:06:28 -0500
Mime-Version: 1.0 (Apple Message framework v548)
Content-Type: text/plain; charset=US-ASCII; format=flowed
From: RJ Atkinson <rja@extremenetworks.com>
To: ieprep <ieprep@ietf.org>
Content-Transfer-Encoding: 7bit
Message-Id: <73F7FEF4-065B-11D7-960F-00039357A82A@extremenetworks.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] correction on IEprep in JP
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


I'm cautiously optimistic that a writeup of the Japanese
deployment of IEprep for IAA by the IAA folks will happen
and will appear here once it has happened.  That should help
level set all regarding the situation in Japan.

I need to make a correction due to increased clarity in
recent email from Japan.  IEprep deployment in Japan
is currently being run by WIDE Project rather than run
by the commercial ISPs, according to my contacts.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 20:13:05 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20469
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 20:13:05 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB31FVp17254
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 20:15:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB31FVv17251
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 20:15:31 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20453
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 20:12:34 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB31D1v17118;
	Mon, 2 Dec 2002 20:13:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB31C8v17048
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 20:12:08 -0500
Received: from newdev.harvard.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20336
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 20:09:11 -0500 (EST)
Received: (from sob@localhost)
	by newdev.harvard.edu (8.12.2/8.12.2) id gB31BMlA001187;
	Mon, 2 Dec 2002 20:11:22 -0500 (EST)
Date: Mon, 2 Dec 2002 20:11:22 -0500 (EST)
From: Scott  Bradner <sob@harvard.edu>
Message-Id: <200212030111.gB31BMlA001187@newdev.harvard.edu>
To: Mpierce1@aol.com, rja@extremenetworks.com, sob@harvard.edu, woody@pch.net
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Cc: ieprep@ietf.org
In-Reply-To: <1ac.ce9e43a.2b1d2c94@aol.com>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

> I meant that whatever entity was
> setting up calls (SIP proxy) would have some knowledge of how many it could 
> attempt to setup simultaneously, at least based on the provisioned capacity. 

1/ as has been pointed out by others, there is no way to know that
there is any relationship between a particular link and any particular
SIP proxy - many proxies, which have no idea that any other proxies exist, 
can set up calls that wind up going over the same link

2/ > If you're telling me that no such CAC is possible in a SIP proxy

even if it were possible (and I do not know if a vendor has that
sort of feature in their product where the product is not a 
gateway to the pstn (where the number of PSTN links is a hard limit)) 
because of #1 it would not help all that much in many of not most cases 

> I have learned, from some other comments, that this lack of being able to 
> respond to congestion (by routing changes or blocking calls) is a significant 
> problem not only for priority traffic.

the IETF has looked at developing routing protocls that respond to
congestion in the past but there is no such routing protocol
in general use in ISPs (cisco's IGRP did have some features to
try to deal with this but I do not know how many enterprises turned that on

Scott
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 20:27:42 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20859
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 20:27:42 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB31U8418086
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 20:30:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB31U8v18083
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 20:30:08 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20848
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 20:27:11 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB31Q1v17880;
	Mon, 2 Dec 2002 20:26:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB31PDv17845
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 20:25:13 -0500
Received: from smtp1.extremenetworks.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20735
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 20:22:16 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.126])
	by smtp1.extremenetworks.com (Postfix) with ESMTP
	id 0AB247A0A; Mon,  2 Dec 2002 17:25:02 -0800 (PST)
Date: Mon, 2 Dec 2002 20:25:01 -0500
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Scott Bradner <sob@harvard.edu>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <200212030111.gB31BMlA001187@newdev.harvard.edu>
Message-Id: <0B3A6089-065E-11D7-960F-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 2, 2002, at 20:11 America/Montreal, Scott Bradner wrote:
>> I have learned, from some other comments, that this lack of being 
>> able to
>> respond to congestion (by routing changes or blocking calls) is a 
>> significant
>> problem not only for priority traffic.
>
> the IETF has looked at developing routing protocls that respond to
> congestion in the past but there is no such routing protocol
> in general use in ISPs

Both IS-IS and OSPF support equal-cost multi-path (ECMP) routing, which
lets a router use a pair of physical links as if it were a single link.
This is widely implemented by many vendors.  Enough customers ask for
this that I think it must have non-trivial deployment.  It is not a
"routing protocol that responds to congestion", but might be relevant.

I agree with Scott's comments.  I'd go farther and say that radical
changes to BGP are unlikely to get past the IDR WG and that ISPs
are unlikely to make wild changes to their IGPs.  (Stable networks
are generally run by risk averse folks).

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  2 20:27:43 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20863
	for <ieprep-archive@odin.ietf.org>; Mon, 2 Dec 2002 20:27:43 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB31U9i18101
	for ieprep-archive@odin.ietf.org; Mon, 2 Dec 2002 20:30:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB31U9v18098
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 2 Dec 2002 20:30:09 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20850
	for <ieprep-web-archive@ietf.org>; Mon, 2 Dec 2002 20:27:12 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB31Q3v17896;
	Mon, 2 Dec 2002 20:26:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB31Pnv17853
	for <ieprep@optimus.ietf.org>; Mon, 2 Dec 2002 20:25:49 -0500
Received: from newdev.harvard.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA20739
	for <ieprep@ietf.org>; Mon, 2 Dec 2002 20:22:52 -0500 (EST)
Received: (from sob@localhost)
	by newdev.harvard.edu (8.12.2/8.12.2) id gB31P4tx001282;
	Mon, 2 Dec 2002 20:25:04 -0500 (EST)
Date: Mon, 2 Dec 2002 20:25:04 -0500 (EST)
From: Scott  Bradner <sob@harvard.edu>
Message-Id: <200212030125.gB31P4tx001282@newdev.harvard.edu>
To: budden@nps.navy.mil, steves@shentel.net
Subject: RE: [Ieprep] Accounting
Cc: ieprep@ietf.org, rja@extremenetworks.com
In-Reply-To: <1038864835.1224.70.camel@ziggy>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

> As a data point, the Common Alerting Protocol working group is working

which working group?  (I do not see a "Common Alerting Protocol" working
group in the IETF)

Scott
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 08:11:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA17890
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 08:11:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3DDUu06297
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 08:13:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3DDUv06294
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 08:13:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA17872
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 08:10:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3D9Sv05956;
	Tue, 3 Dec 2002 08:09:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3D34v05131
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 08:03:04 -0500
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA17224;
	Tue, 3 Dec 2002 08:00:12 -0500 (EST)
Message-Id: <200212031300.IAA17224@ietf.org>
Mime-Version: 1.0
Content-Type: Multipart/Mixed; Boundary="NextPart"
To: IETF-Announce: ;
Cc: ieprep@ietf.org
From: Internet-Drafts@ietf.org
Reply-to: Internet-Drafts@ietf.org
Date: Tue, 03 Dec 2002 08:00:12 -0500
Subject: [Ieprep] I-D ACTION:draft-ietf-ieprep-sip-reqs-02.txt
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

--NextPart

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

	Title		: Requirements for Resource Priority Mechanisms for the 
                          Session Initiation Protocol
	Author(s)	: H. Schulzrinne
	Filename	: draft-ietf-ieprep-sip-reqs-02.txt
	Pages		: 16
	Date		: 2002-12-2
	
This document summarizes requirements for prioritizing access to
circuit-switched network, end system and proxy resources for
emergency preparedness communications using the Session Initiation
Protocol (SIP).

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-ieprep-sip-reqs-02.txt

To remove yourself from the IETF Announcement list, send a message to 
ietf-announce-request with the word unsubscribe in the body of the message.

Internet-Drafts are also available by anonymous FTP. Login with the username
"anonymous" and a password of your e-mail address. After logging in,
type "cd internet-drafts" and then
	"get draft-ietf-ieprep-sip-reqs-02.txt".

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


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

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-ieprep-sip-reqs-02.txt".
	
NOTE:	The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.
		
		
Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

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

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

Content-Type: text/plain
Content-ID:	<2002-12-2140109.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-ieprep-sip-reqs-02.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-ieprep-sip-reqs-02.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<2002-12-2140109.I-D@ietf.org>

--OtherAccess--

--NextPart--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 08:56:47 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20093
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 08:56:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3Dx8f09635
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 08:59:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Dx7v09632
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 08:59:07 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20072
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 08:56:15 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Dw2v09602;
	Tue, 3 Dec 2002 08:58:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Dvxv09580
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 08:57:59 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA20010
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 08:55:07 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Tue, 3 Dec 2002 08:57:26 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120308574331284
 ; Tue, 03 Dec 2002 08:57:43 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YDVM6NV7>; Tue, 3 Dec 2002 08:57:30 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E779DE@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Scott Bradner '" <sob@harvard.edu>,
        "'budden@nps.navy.mil '" <budden@nps.navy.mil>,
        "'steves@shentel.net '" <steves@shentel.net>
Cc: "'ieprep@ietf.org '" <ieprep@ietf.org>,
        "'rja@extremenetworks.com '" <rja@extremenetworks.com>
Date: Tue, 3 Dec 2002 08:57:41 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] Common Alerting Protocol
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

>which working group?  (I do not see a "Common Alerting Protocol" 
>working group in the IETF)

Can't get anything by you :-)

They are talking about a prototype http://www.incident.com/cap/ early
warning system.  The "working group" seems to indicate they want to use the
IETF RFC type of process although they are certainly not part of the IETF.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 09:03:47 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA20599
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 09:03:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3E68Z10125
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 09:06:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3E68v10122
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 09:06:08 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA20562
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 09:03:16 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3E51v10055;
	Tue, 3 Dec 2002 09:05:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3E4qv10002
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 09:04:52 -0500
Received: from cs.columbia.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA20474
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 09:02:00 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB3E4nl8007624
	(version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NOT);
	Tue, 3 Dec 2002 09:04:50 -0500 (EST)
Received: from cs.columbia.edu (tallgrass.netlab.uky.edu [204.198.76.66])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB3E4mdG029597
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Tue, 3 Dec 2002 09:04:49 -0500 (EST)
Message-ID: <3DECB97F.6050700@cs.columbia.edu>
Date: Tue, 03 Dec 2002 09:02:39 -0500
From: Henning Schulzrinne <hgs@cs.columbia.edu>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.2b) Gecko/20021016
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: "King, Kimberly S." <KIMBERLY.S.KING@saic.com>
CC: "'ieprep@ietf.org '" <ieprep@ietf.org>
Subject: Re: [Ieprep] Common Alerting Protocol
References: <B8030EB94AF1D51196D70002A589D64207E779DE@mcl-its-exs01.mail.saic.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E779DE@mcl-its-exs01.mail.saic.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

For what it's worth, we have an implementation of the CAP data format on 
top of SIP IM in our SIP UA. Thus, there's some connection to IETF 
protocols :-)

King, Kimberly S. wrote:

> >which working group?  (I do not see a "Common Alerting Protocol"
> >working group in the IETF)
>
>
> Can't get anything by you :-)
>
> They are talking about a prototype http://www.incident.com/cap/ early
> warning system.  The "working group" seems to indicate they want to 
> use the
> IETF RFC type of process although they are certainly not part of the IETF.
>
> Kimberly
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 10:29:19 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA26236
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 10:29:19 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3FVfA17811
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 10:31:41 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3FVfv17808
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 10:31:41 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA26211
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 10:28:47 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3FU6v17587;
	Tue, 3 Dec 2002 10:30:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3FT2v17516
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 10:29:02 -0500
Received: from imo-m06.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA26027
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 10:26:08 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m06.mx.aol.com (mail_out_v34.13.) id j.181.12dd876c (657);
	Tue, 3 Dec 2002 10:28:51 -0500 (EST)
Message-ID: <181.12dd876c.2b1e27b3@aol.com>
Date: Tue, 3 Dec 2002 10:28:51 EST
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
To: rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_181.12dd876c.2b1e27b3_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_181.12dd876c.2b1e27b3_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/2/2002 4:20:23 PM Eastern Standard Time, 
rja@extremenetworks.com writes:


On Monday, Dec 2, 2002, at 12:54 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] I understand fully that the IP-layer is unaware of "calls" or 
> "voice packets". That wasn't the point. If some higher layer function 
> doesn't limit the setup of new calls (2. above) when it knows that 
> there are already too many being stuffed through a finite size link, 
> then the IP-layer would be overwhelmed with packets which it can't 
> handle (1. above).

[RJA] The IP-layer can be overwhelmed anyway since IP does not have any 
equivalent
to "call setup".  Even if SIP were informed (which it isn't) and had some way
to control the voice calls (which it doesn't since anyone on the Internet can
use any cooperating SIP server, not just the local one), other unrelated 
applications
like mail or ftp or http don't use SIP (and won't be using it in future).



[MAP] So this says that congestion can occur and there is no way for the 
application (SIP) to deal with it, like by restricting new call setups, so 
the IP-layer is left completely on its own to deal with it, and the only way 
it can is by discarding packets.


[RJA] Now one could use IP ToS bits to mark the packets from inside a domain 
for
the uplink router to prioritise, if the uplink router happened to have that
policy configured in it by the router's operator.  One could also put various
other policies into the router based on some combinations of (address, 
protocol, port),
if one did not want to use the IP ToS bits.  If the link is Ethernet, one 
could
use 802.1p.  If the link is something else, then one could use other 
link-specific
mechanisms (e.g. ATM has PSTN-like calls).



[MAP] I think we've agreed that packet "prioritization" of one voice packet 
before another is not feasible or a good idea, so I don't think you mean 
that. I can't imagine that handling of packets for priority calls could be 
done by putting specific addresses in the routers. (Sounds like "state" to me 
anyway.)


> [MAP] So I take that as an agreement with what I said, that in the 
> case of congestion, a lot of packets would be lost.

[RJA] No.  The number of packets lost would vary with (offered load, link 
capacity,
traffic pattern, packet memory depth on each side of the link).  In many cases
there is minimal or no congestion, because the router buffer is deep enough
to handle a trans-oceanic RTT, which in turn gives TCP or SCTP enough time
to reduce the transmission rate on existing sessions, which then leaves
enough room for the new traffic to be carried on the link.



[MAP] Your answer relates to elastic traffic, e.g., file transfers, where the 
buffer can be made as "deep" as you have memory available. The primary 
subject here is inelastic traffic (voice) in which there are relatively 
strict limits on the buffer length.


> [MAP] I'll repeat the question which I don't believe I've gotten any 
> concrete answer to: In the case above of the 10 callers from Nairobi, 
> if one of them was authorized to make emergency calls and the other 9 
> weren't, what specific standardized mechanism (not products) could be 
> used to provide the required service across the congested 384 kbit/s 
> satellite link? (The required service is that their call gets set up 
> and they can talk.) If you don't believe that 10 users would cause 
> congestion, then make it 20 or however many of the several thousand it 
> would take.
>
> I would find it hard to believe that there are no standardized 
> solutions to this. You're right, maybe there is no need for this WG, 
> but I haven't heard the solution yet.

[RJA] One could use IP ToS bits, link-specific ToS bits (802.1p in the case 
of Ethernet),
or some combination of (address/protocol/port) to tell the access router which
traffic to prefer from the site connected to that access router.  Preference 
can be
implemented using the CoS mechanism that Sprint is using, various flavours of 
which
are available from many vendors.  Whether that policy (either to obey ToS 
bits or to sort
based on the address/protocol/port) is configured into that router is a 
question
for whoever  owns/operates that access router.

If the offered load of preferred traffic exceeds link capacity, then packets
still will be dropped.

[MAP] Our goal is to point to a specific set of standards (or proposed) to 
describe how to provide preferential treatment. Can you point me to some 
specific ones? Os is it naive to think that it is as simple as referrencing 
EF or some such RFC?

Mike Pierce
Artel


--part1_181.12dd876c.2b1e27b3_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/2/2002 4:20:23 PM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR>On Monday, Dec 2, 2002, at 12:54 America/Montreal, Mpierce1@aol.com 
<BR>wrote:
<BR>&gt; [MAP] I understand fully that the IP-layer is unaware of "calls" or 
<BR>&gt; "voice packets". That wasn't the point. If some higher layer function 
<BR>&gt; doesn't limit the setup of new calls (2. above) when it knows that 
<BR>&gt; there are already too many being stuffed through a finite size link, 
<BR>&gt; then the IP-layer would be overwhelmed with packets which it can't 
<BR>&gt; handle (1. above).
<BR>
<BR>[RJA] The IP-layer can be overwhelmed anyway since IP does not have any equivalent
<BR>to "call setup". &nbsp;Even if SIP were informed (which it isn't) and had some way
<BR>to control the voice calls (which it doesn't since anyone on the Internet can
<BR>use any cooperating SIP server, not just the local one), other unrelated applications
<BR>like mail or ftp or http don't use SIP (and won't be using it in future).
<BR>
<BR>
<BR>
<BR>[MAP] So this says that congestion can occur and there is no way for the application (SIP) to deal with it, like by restricting new call setups, so the IP-layer is left completely on its own to deal with it, and the only way it can is by discarding packets.
<BR>
<BR>
<BR>[RJA] Now one could use IP ToS bits to mark the packets from inside a domain for
<BR>the uplink router to prioritise, if the uplink router happened to have that
<BR>policy configured in it by the router's operator. &nbsp;One could also put various
<BR>other policies into the router based on some combinations of (address, protocol, port),
<BR>if one did not want to use the IP ToS bits. &nbsp;If the link is Ethernet, one could
<BR>use 802.1p. &nbsp;If the link is something else, then one could use other link-specific
<BR>mechanisms (e.g. ATM has PSTN-like calls).
<BR>
<BR>
<BR>
<BR>[MAP] I think we've agreed that packet "prioritization" of one voice packet before another is not feasible or a good idea, so I don't think you mean that. I can't imagine that handling of packets for priority calls could be done by putting specific addresses in the routers. (Sounds like "state" to me anyway.)
<BR>
<BR>
<BR>&gt; [MAP] So I take that as an agreement with what I said, that in the 
<BR>&gt; case of congestion, a lot of packets would be lost.
<BR>
<BR>[RJA] No. &nbsp;The number of packets lost would vary with (offered load, link capacity,
<BR>traffic pattern, packet memory depth on each side of the link). &nbsp;In many cases
<BR>there is minimal or no congestion, because the router buffer is deep enough
<BR>to handle a trans-oceanic RTT, which in turn gives TCP or SCTP enough time
<BR>to reduce the transmission rate on existing sessions, which then leaves
<BR>enough room for the new traffic to be carried on the link.
<BR>
<BR>
<BR>
<BR>[MAP] Your answer relates to elastic traffic, e.g., file transfers, where the buffer can be made as "deep" as you have memory available. The primary subject here is inelastic traffic (voice) in which there are relatively strict limits on the buffer length.
<BR>
<BR>
<BR>&gt; [MAP] I'll repeat the question which I don't believe I've gotten any 
<BR>&gt; concrete answer to: In the case above of the 10 callers from Nairobi, 
<BR>&gt; if one of them was authorized to make emergency calls and the other 9 
<BR>&gt; weren't, what specific standardized mechanism (not products) could be 
<BR>&gt; used to provide the required service across the congested 384 kbit/s 
<BR>&gt; satellite link? (The required service is that their call gets set up 
<BR>&gt; and they can talk.) If you don't believe that 10 users would cause 
<BR>&gt; congestion, then make it 20 or however many of the several thousand it 
<BR>&gt; would take.
<BR>&gt;
<BR>&gt; I would find it hard to believe that there are no standardized 
<BR>&gt; solutions to this. You're right, maybe there is no need for this WG, 
<BR>&gt; but I haven't heard the solution yet.
<BR>
<BR>[RJA] One could use IP ToS bits, link-specific ToS bits (802.1p in the case of Ethernet),
<BR>or some combination of (address/protocol/port) to tell the access router which
<BR>traffic to prefer from the site connected to that access router. &nbsp;Preference can be
<BR>implemented using the CoS mechanism that Sprint is using, various flavours of which
<BR>are available from many vendors. &nbsp;Whether that policy (either to obey ToS bits or to sort
<BR>based on the address/protocol/port) is configured into that router is a question
<BR>for whoever &nbsp;owns/operates that access router.
<BR>
<BR>If the offered load of preferred traffic exceeds link capacity, then packets
<BR>still will be dropped.
<BR>
<BR>[MAP] Our goal is to point to a specific set of standards (or proposed) to describe how to provide preferential treatment. Can you point me to some specific ones? Os is it naive to think that it is as simple as referrencing EF or some such RFC?
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_181.12dd876c.2b1e27b3_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 10:29:20 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA26239
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 10:29:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3FVgU17826
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 10:31:42 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3FVgv17823
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 10:31:42 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA26215
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 10:28:48 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3FU9v17605;
	Tue, 3 Dec 2002 10:30:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3FT5v17520
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 10:29:05 -0500
Received: from imo-d04.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA26030
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 10:26:12 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d04.mx.aol.com (mail_out_v34.13.) id a.177.12d0de1f (657);
	Tue, 3 Dec 2002 10:28:29 -0500 (EST)
Message-ID: <177.12d0de1f.2b1e279d@aol.com>
Date: Tue, 3 Dec 2002 10:28:29 EST
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
To: KIMBERLY.S.KING@saic.com, rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_177.12d0de1f.2b1e279d_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_177.12d0de1f.2b1e279d_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/2/2002 4:38:56 PM Eastern Standard Time, 
KIMBERLY.S.KING@saic.com writes:


> On Monday, Dec 2, 2002, at 10:30 America/Montreal, Mpierce1@aol.com 
> wrote: 
> > [MAP] So take this simple case and apply an emergency. (I'd like to 
> > see the performance last week right after the bombing in Kenya!) 
> > 
> > If 10 or 20 or whatever number of those "several thousands" of end 
> > users tried to make a VoIP call via that single 384K link, what would 
> > have happened? It is my understanding that, at some point, the 
> > overload would have caused either: 
> > 
> > 1. all calls to be of unacceptable quality 
> > 2. new call attempts would have been rejected 
> 
> [MAP] I'll repeat the question which I don't believe I've gotten any
> concrete answer to: In the case above of the 10 callers from Nairobi, if
> one of them was authorized to make emergency calls and the other 9
> weren't, what specific standardized mechanism (not products) could be
> used to provide the required service across the congested 384 kbit/s
> satellite link? (The required service is that their call gets set up and
> they can talk.) If you don't believe that 10 users would cause
> congestion, then make it 20 or however many of the several thousand it
> would take. 
> 
> Mike,
> 
> The answers to your question depends upon the assumptions you make (i.e.,
> problem context).  Do you assume your 10 or however many users are just
> random folks who happen to be using the same access link owned by an
> arbitrary ISP?  Then your answer is very likely to be "the call quality for
> all users is going to be poor".  



[MAP] The case began with that situation, and, as you said, all users get 
poor quality (probably too poor to converse at some point of congestion). The 
case then moved on to the one of 9 random folks and 1 with authorized 
emergency access.

> 
> Let's us examine why...well, we could try to identify emergency users with
> SIP signaling and try to use "call admission control" in the telco sense.
> Fine except your 10 users might use 10 different SIP service providers
> completely unconnected to the ISP who happens to be providing the access
> link.  Those SIP providers certainly have no way of knowing what other SIP
> service providers are doing nor do they track bandwidth utilization across
> some random link (they have no way to do this) hence they can't provide a
> "call admission control".  So, you can then assume all 10 users are allowed
> to setup calls.  Your access link provider is quite likely to be just moving
> 



[MAP] Yes, I understand that 10 independent SIP providers can't know about 
the link loading in order to decide whether or not to allow a new call to be 
set up (which is going to "fail" anyway due to bad audio). This is a major 
shortcoming with SIP and VoIP and IP routing and until it is recognized, 
admitted, and fixed, it can not be used for real telephone service (much less 
emergency services).
> 
> So, with these set of assumptions, the answer is the call quality will be
> 



[MAP] And, as we know from the telephony world, many of those who hang up 
will immediately redial.

> 
> But, under a different set of assumptions, the answer is different.  Assume
> all 10 users have a local soft PBX that provides a "call admission control"
> for all off net calls.  It also tracks who's got priority and can even
> preempt lower priority calls.  Then 10 users try to setup but only the
> emergency worker and the max number the access link can support are allowed
> to complete.  This could actually support Assured Service.  
> 



[MAP] Of course, I would expect any "soft PBX" to think in terms of circuit 
swithing, i.e., it knows that its own access link can support a certain 
number of calls and would reject any more. (Of course, this doesn't help if 
the congestion is at a further point down the path.)

> 



[MAP] Absolutely right. And it would be nice if we could exchange some good 
diagrams af what we are talking about. I for one, have a hard time 
visualizing what someone else is thinking based on their verbal description. 
(I don't know exactly what the configuration was for the subject call from 
Nairobi with thousands of potential competing users. I presume it was a 
router between the thousands of users and the 384 kbit/s sataillite link, so 
it couldn't do CAC.)

Mike Pierce
Artel


--part1_177.12d0de1f.2b1e279d_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/2/2002 4:38:56 PM Eastern Standard Time, KIMBERLY.S.KING@saic.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">On Monday, Dec 2, 2002, at 10:30 America/Montreal, Mpierce1@aol.com 
<BR>wrote: 
<BR>&gt; [MAP] So take this simple case and apply an emergency. (I'd like to 
<BR>&gt; see the performance last week right after the bombing in Kenya!) 
<BR>&gt; 
<BR>&gt; If 10 or 20 or whatever number of those "several thousands" of end 
<BR>&gt; users tried to make a VoIP call via that single 384K link, what would 
<BR>&gt; have happened? It is my understanding that, at some point, the 
<BR>&gt; overload would have caused either: 
<BR>&gt; 
<BR>&gt; 1. all calls to be of unacceptable quality 
<BR>&gt; 2. new call attempts would have been rejected 
<BR>
<BR>[MAP] I'll repeat the question which I don't believe I've gotten any
<BR>concrete answer to: In the case above of the 10 callers from Nairobi, if
<BR>one of them was authorized to make emergency calls and the other 9
<BR>weren't, what specific standardized mechanism (not products) could be
<BR>used to provide the required service across the congested 384 kbit/s
<BR>satellite link? (The required service is that their call gets set up and
<BR>they can talk.) If you don't believe that 10 users would cause
<BR>congestion, then make it 20 or however many of the several thousand it
<BR>would take. 
<BR>
<BR>Mike,
<BR>
<BR>The answers to your question depends upon the assumptions you make (i.e.,
<BR>problem context). &nbsp;Do you assume your 10 or however many users are just
<BR>random folks who happen to be using the same access link owned by an
<BR>arbitrary ISP? &nbsp;Then your answer is very likely to be "the call quality for
<BR>all users is going to be poor". &nbsp;</BLOCKQUOTE>
<BR>
<BR>
<BR>
<BR>[MAP] The case began with that situation, and, as you said, all users get poor quality (probably too poor to converse at some point of congestion). The case then moved on to the one of 9 random folks and 1 with authorized emergency access.
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>Let's us examine why...well, we could try to identify emergency users with
<BR>SIP signaling and try to use "call admission control" in the telco sense.
<BR>Fine except your 10 users might use 10 different SIP service providers
<BR>completely unconnected to the ISP who happens to be providing the access
<BR>link. &nbsp;Those SIP providers certainly have no way of knowing what other SIP
<BR>service providers are doing nor do they track bandwidth utilization across
<BR>some random link (they have no way to do this) hence they can't provide a
<BR>"call admission control". &nbsp;So, you can then assume all 10 users are allowed
<BR>to setup calls. &nbsp;Your access link provider is quite likely to be just moving
<BR>packets.</BLOCKQUOTE>
<BR>
<BR>
<BR>
<BR>[MAP] Yes, I understand that 10 independent SIP providers can't know about the link loading in order to decide whether or not to allow a new call to be set up (which is going to "fail" anyway due to bad audio). This is a major shortcoming with SIP and VoIP and IP routing and until it is recognized, admitted, and fixed, it can not be used for real telephone service (much less emergency services).
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>So, with these set of assumptions, the answer is the call quality will be
<BR>poor for all until some people hang up.</BLOCKQUOTE>
<BR>
<BR>
<BR>
<BR>[MAP] And, as we know from the telephony world, many of those who hang up will immediately redial.
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>But, under a different set of assumptions, the answer is different. &nbsp;Assume
<BR>all 10 users have a local soft PBX that provides a "call admission control"
<BR>for all off net calls. &nbsp;It also tracks who's got priority and can even
<BR>preempt lower priority calls. &nbsp;Then 10 users try to setup but only the
<BR>emergency worker and the max number the access link can support are allowed
<BR>to complete. &nbsp;This could actually support Assured Service. &nbsp;
<BR></BLOCKQUOTE>
<BR>
<BR>
<BR>
<BR>[MAP] Of course, I would expect any "soft PBX" to think in terms of circuit swithing, i.e., it knows that its own access link can support a certain number of calls and would reject any more. (Of course, this doesn't help if the congestion is at a further point down the path.)
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">See how important is it to define the context?</BLOCKQUOTE>
<BR>
<BR>
<BR>
<BR>[MAP] Absolutely right. And it would be nice if we could exchange some good diagrams af what we are talking about. I for one, have a hard time visualizing what someone else is thinking based on their verbal description. (I don't know exactly what the configuration was for the subject call from Nairobi with thousands of potential competing users. I presume it was a router between the thousands of users and the 384 kbit/s sataillite link, so it couldn't do CAC.)
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_177.12d0de1f.2b1e279d_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 10:54:02 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA28319
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 10:54:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3FuPY20416
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 10:56:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3FuPv20413
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 10:56:25 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA28284
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 10:53:31 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Fq9v20008;
	Tue, 3 Dec 2002 10:52:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3FpJv19944
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 10:51:19 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA27789
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 10:48:26 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Tue, 3 Dec 2002 09:48:02 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120309482028010
 ; Tue, 03 Dec 2002 09:48:20 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YDVQTR6R>; Tue, 3 Dec 2002 09:49:49 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E779E3@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
Cc: "'sob@harvard.edu'" <sob@harvard.edu>
Date: Tue, 3 Dec 2002 09:48:18 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain
Subject: [Ieprep] IEPREP Topology draft after WG last call
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

"IEPREP Topology Scenarios" draft has completed WG last call.  
  
James will submit a new version today that reflects the comments he received
from the WG.  Please review and direct any last minute issues to him (within
24 hours).

The chairs will then ask the AD to publish the draft as an informational
RFC.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 11:21:24 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA00399
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 11:21:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3GNjN23121
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 11:23:45 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3GNjv23118
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 11:23:45 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA00376
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 11:20:52 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3GM4v23021;
	Tue, 3 Dec 2002 11:22:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3GJ2v22704
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 11:19:02 -0500
Received: from [140.32.132.66] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA00111
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 11:16:10 -0500 (EST)
Received: from mail.nps.navy.mil by [140.32.132.66]
          via smtpd (for [132.151.6.1]) with ESMTP; Tue, 3 Dec 2002 08:14:45 -0800
Received: from ziggy ([131.120.179.220])
	by capella.nps.navy.mil (8.12.2/8.12.2) with ESMTP id gB3GFQbR016313;
	Tue, 3 Dec 2002 08:15:26 -0800 (PST)
Subject: RE: [Ieprep] Accounting
From: Rex Buddenberg <budden@nps.navy.mil>
To: Scott Bradner <sob@harvard.edu>
Cc: steves@shentel.net, ieprep@ietf.org, rja@extremenetworks.com
In-Reply-To: <200212030125.gB31P4tx001282@newdev.harvard.edu>
References: <200212030125.gB31P4tx001282@newdev.harvard.edu>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) 
Date: 03 Dec 2002 08:11:15 -0800
Message-Id: <1038931876.1224.90.camel@ziggy>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

On Mon, 2002-12-02 at 17:25, Scott Bradner wrote:
> > As a data point, the Common Alerting Protocol working group is working
> 
> which working group?  (I do not see a "Common Alerting Protocol" working
> group in the IETF)
> 
> Scott


CAP-list@incident.com
http://www.incident.com/mailman/listinfo
"CAP-list" is an email list for discussion about concepts, requirements
and specifications for a Common Alerting Protocol.  It is not for
product announcements, advertising or advocacy.  For more information
about the Common Alerting Protocol, visit
<http://www.incident.com/cap/>.


... not part of IETF

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 12:53:40 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA06209
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 12:53:40 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3Hu2u31843
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 12:56:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Hu2v31840
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 12:56:02 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA06171
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 12:53:08 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Hq6v31558;
	Tue, 3 Dec 2002 12:52:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3HpQv31531
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 12:51:26 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA05923
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 12:48:32 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Tue, 3 Dec 2002 12:50:42 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120312505800006
 ; Tue, 03 Dec 2002 12:50:58 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YDVQTXAQ>; Tue, 3 Dec 2002 12:52:27 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E779E6@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson '" <rja@extremenetworks.com>,
        "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>
Cc: "''ieprep ' '" <ieprep@ietf.org>, "'rohan@cisco.com'" <rohan@cisco.com>
Subject: RE: [Ieprep] identification of packets
Date: Tue, 3 Dec 2002 12:50:55 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gB3HpQv31532
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

I said:
> While it is true that "entities that
> are addressing congestion on the access link" may use different 
> methods,
> this doesn't stop the hope that in many cases a common configuration
of
> *user software* is possible.  It doesn't have to imply interprovider 
> QoS
> either.

Ran said:
>Hmm.

>What do you mean by "common configuration of user software" ?

I'm an intuitionist, I was thinking in that way at the time.

Anyway, I was thinking about old conversations about the differences between
the PSTN and the Internet and how it is endsystems (lack of common
configuration) that stalls working systems on the Internet.  I was also
thinking about a conversation I had with Rohan Mahy at the last IETF about
an idea that he has.  Here the common configuration would be some user
software, configured in a particular way to work with a select number of
ISPs.  

(Sorry Rohan if I'm about to misrepresent your idea...)

Scenario:  Emergency Services Organization contracts with several major ISPs
to provide the following service.  Critical emergency workers are given some
type of identification/authorization single use token.  When a disaster
occurs, then the emergency worker finds an Internet connection such as an
Internet café.  

Suppose the service quality isn't as good as desired (due to dreaded
congestion).  The emergency worker then sends the authentication information
out to the network (e.g., a modification of RSVP that is intercepted by the
first hop access router).  

If the access router is part of an ISP providing this service, this
authentication token triggers a special configuration on the access router.
Namely, this triggers a configuration method existing on this access router
that then provides additional resources for the connection from the given
location (e.g., increases the token bucket size or some other form of rough
additional resource).  This behavior isn't necessarily based upon flows from
the particular user but rather for the entire connection, say for some fixed
time period (e.g., 5 days).  

The WG can investigate if more complicated behavior is reasonable such as
the emergency workers flows be marked by particular DiffServ codepoints
however the WG is strongly suggested follow the KISS principal.  Remember
that even the roughest of preferential treatment can potentially increase
the probably of successful communications.  Also keep in mind the tradeoff
between complexity and security, and complexity and cost (i.e., complicated
deployments are often expensive deployments).  Possibly, additional
resources could be provided at both directions at the access router (i.e.,
extra resources in accepting packets from that connection and also resources
for packets destine back to the emergency worker).  If the application also
involves another users such as with VoIP, is also possible that the egress
router, should it also be in one of the contracted ISPs would also provide
some extra measure of QoS.  Note that there will be security concerns
including replay attacks.  
	
So, I think we should toss around this idea with the IP experts and see if
they can figure out if such a scheme is viable, and what protocols might
need some modification to support such a feature.  (For example, RSVP may be
an ideal choice.)  

If this all seems positive, I'll try to round up a design team to work on
the problem.  Members of the design team are responsible for choices made
along the way however their progress and background on why they make the
choices they make will be presented to the working group.  This isn't a case
where rough consensus is abandoned.  Any requirements or BCPs produced by
the WG are still subject to WG rough consensus.

Thoughts?

Kimberly


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 13:22:57 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA07927
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 13:22:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3IPKH01906
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 13:25:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3IPKv01903
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 13:25:20 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA07868
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 13:22:26 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3IL3v01688;
	Tue, 3 Dec 2002 13:21:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3IKEv01645
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 13:20:14 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA07550
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 13:17:19 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id BE6D867104; Tue,  3 Dec 2002 09:45:44 -0500 (EST)
Date: Tue, 3 Dec 2002 13:20:08 -0500
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep@ietf.org
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <181.12dd876c.2b1e27b3@aol.com>
Message-Id: <DA9FF9A2-06EB-11D7-A694-00039357A82A@extremenetworks.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gB3IKEv01646
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


On Tuesday, Dec 3, 2002, at 10:28 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] So this says that congestion can occur and there is no way for 
> the application (SIP) to deal with it, like by restricting new call 
> setups, so the IP-layer is left completely on its own to deal with it, 
> and the only way it can is by discarding packets.

Hmm.  Almost, but not quite right.  The application end-points, that is 
the
application software on either end of the actual VoIP session, *can* 
and *do*
know that they are experiencing X% packet loss or that jitter is 
increasing
from W to Z, or that latency has increased from A to B, or any 
combination
of the above.  Having that knowledge, which is in the VoIP endpoints 
not in the
SIP server, the endpoints could react in any of several ways.  An 
example
that seems most obvious to me would be to switch from G.711 to a more 
robust
CODEC that consumes less bandwidth (e.g. some form of CELP).

Now I'm not an application-layer guy very often, so I don't know the 
details
of SIP or the details of how the voice is being sent (presumably over 
RTP/UDP/IP
somehow).  However, it is clear that protocol engineering could be done 
to add
this kind of capability (e.g. codec negotiation) if folks in that part 
of the
application community cared to do it.

> [RJA] Now one could use IP ToS bits to mark the packets from inside a 
> domain for
> the uplink router to prioritise, if the uplink router happened to have 
> that
> policy configured in it by the router's operator.  One could also put 
> various
> other policies into the router based on some combinations of (address, 
> protocol, port),
> if one did not want to use the IP ToS bits.  If the link is Ethernet, 
> one could
> use 802.1p.  If the link is something else, then one could use other 
> link-specific
> mechanisms (e.g. ATM has PSTN-like calls).
>
> [MAP] I think we've agreed that packet "prioritization" of one voice 
> packet before another is not feasible or a good idea, so I don't think 
> you mean that. I can't imagine that handling of packets for priority 
> calls could be done by putting specific addresses in the routers. 
> (Sounds like "state" to me anyway.)

In the text from me that you've quoted just above, one concept could be 
that
IEprep traffic (whether VoIP or other) *could* be marked using IP ToS 
bits
(either at origin, by first-hop router inside the domain, or by some 
other
router inside the domain) and (subject to local network operator's 
policy)
the IP ToS bits in each packet could be examined by the domain's uplink 
router
to give IEprep packets lower drop probability using some 
(policy-dependent)
queuing configuration inside the router (e.g. Sprint's CoS stuff, which
multiple router vendors support in COTS product today).

An advantage to this approach over a NSIS/RSVP-like approach is that one
has a relatively small amount of total state for the QoS, pre-configured
into the router, rather than adding router state for each additional
IP-layer flow that happens to be "preferred".  Basically it adds a small
amount of constant state to handle an unlimited number of concurrent 
flows,
rather than adding state for each additional flow that might be present.
The issue with state in routers is always how much one has.  This is an 
approach
to keep the total state related to QoS small regardless of traffic load 
or mix.

> > [MAP] So I take that as an agreement with what I said, that in the
> > case of congestion, a lot of packets would be lost.
>
> [RJA] No.  The number of packets lost would vary with (offered load, 
> link capacity,
> traffic pattern, packet memory depth on each side of the link).  In 
> many cases
> there is minimal or no congestion, because the router buffer is deep 
> enough
> to handle a trans-oceanic RTT, which in turn gives TCP or SCTP enough 
> time
> to reduce the transmission rate on existing sessions, which then leaves
> enough room for the new traffic to be carried on the link.
>
> [MAP] Your answer relates to elastic traffic, e.g., file transfers, 
> where the buffer can be made as "deep" as you have memory available. 
> The primary subject here is inelastic traffic (voice) in which there 
> are relatively strict limits on the buffer length.

Routers have no idea whether a packet is voice or file transfer.  The 
amount
of packet memory in the router is a property of what humans have 
physically
inserted.  So my comment quoted above is independent of what kind of 
application
happens to be running on the packets being buffered.

Variable latency is a fact of life in any packet network.  All 
Internet-based
applications, including voice, have to be reasonably tolerant of some 
amount
of jitter or they won't work at all.

An IAB comment by Sally Floyd at the IEprep meeting in Atlanta bears 
repeating.
Because the current VoIP stuff does not include congestion avoidance, 
the tendency
will be for the VoIP traffic to "crowd out" the congestion avoiding 
traffic
(e.g. anything running over TCP or SCTP).  This is a major problem.  
Somehow
the community needs to add congestion avoidance into the VoIP 
applications
protocols.

> [MAP] Our goal is to point to a specific set of standards (or 
> proposed) to describe how to provide preferential treatment. Can you 
> point me to some specific ones? Os is it naive to think that it is as 
> simple as referrencing EF or some such RFC?

If one wants to talk about standards-track QoS mechanisms, there are a 
wide range of
RFCs to choose from.  They each have different properties.  Each is 
suited for somewhat
different operating conditions.  There is no "one size fits all" RFC.

It is, however, important to keep in mind that IETF avoids recommending 
or dictating
that an operator deploy particular standards.  Whether an ISP chooses 
to deploy
anything (e.g. OSPF routing protocol vs IS-IS routing protocol vs 
some-proprietary
routing protocol) is *entirely* a decision of that ISP.

For many reasons, including some discussed by the IAB in RFC-2990, 
Sections 3.10
through 3.12, the deployment of inter-domain QoS is impractical and 
improbable
anytime in the near future.  So if one wants to offer a robust service 
over a
multi-domain environment, one would be well advised to focus on 
ensuring that
one uses robust applications (e.g. very thoughtful codec selection, not 
limited
just to ITU specs, in the case of VoIP applications).

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 13:28:34 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA08330
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 13:28:33 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3IUu002491
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 13:30:56 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3IUuv02488
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 13:30:56 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA08294
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 13:28:02 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3IR2v02075;
	Tue, 3 Dec 2002 13:27:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3IQXv01962
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 13:26:33 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA07985
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 13:23:39 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id EDAF567104; Tue,  3 Dec 2002 09:52:03 -0500 (EST)
Date: Tue, 3 Dec 2002 13:26:27 -0500
Subject: Re: [Ieprep] Architectural considerations
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <177.12d0de1f.2b1e279d@aol.com>
Message-Id: <BC9FFBF0-06EC-11D7-A694-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 3, 2002, at 10:28 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] Yes, I understand that 10 independent SIP providers can't know 
> about the link loading in order to decide whether or not to allow a 
> new call to be set up (which is going to "fail" anyway due to bad 
> audio). This is a major shortcoming with SIP and VoIP and IP routing 
> and until it is recognized, admitted, and fixed, it can not be used 
> for real telephone service (much less emergency services).

I consider this an architectural feature.

And I think the main fix is to design and deploy applications that are 
robust in the
real deployed Internet.  Since some have been using VoIP for some years 
now in a
production environment that includes voice/IP/HF Radio (with channel 
fading and
other joys), it seems clear to me that this is entirely possible.  It 
might mean
using codecs other than G.711, and other than ITU-T blessed codecs, 
however,
to give an example.

The Internet has a fundamentally different architecture than the PSTN.  
To run
voice over the Internet, one should adapt the voice application to the 
Internet
architecture rather than insisting that the Internet adopt the PSTN's 
broken
(e.g. no dynamic routing of traffic around fibre cuts) architecture.

I view someone coming to IETF and demanding that the Internet adopt the 
PSTN
architecture to be just as silly as it would be for me to go to ITU-T 
and
demand that the deployed PSTN abandon its design and use the Internet 
architecture.
So far, I've avoided doing the latter by avoiding ITU-T meetings/lists.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 14:29:41 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA11886
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 14:29:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3JW5n07497
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 14:32:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3JW5v07494
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 14:32:05 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA11846
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 14:29:10 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3JS7v07279;
	Tue, 3 Dec 2002 14:28:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3JRev07233
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 14:27:40 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA11697
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 14:24:45 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id ED62D67104; Tue,  3 Dec 2002 10:53:07 -0500 (EST)
Date: Tue, 3 Dec 2002 14:27:30 -0500
Subject: Re: [Ieprep] identification of packets
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "''ieprep ' '" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E779E6@mcl-its-exs01.mail.saic.com>
Message-Id: <43A8AF80-06F5-11D7-A694-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 3, 2002, at 12:50 America/Montreal, King, Kimberly S. 
wrote:
> The WG can investigate if more complicated behavior is reasonable such 
> as
> the emergency workers flows be marked by particular DiffServ codepoints
> however the WG is strongly suggested follow the KISS principal.  
> Remember
> that even the roughest of preferential treatment can potentially 
> increase
> the probably of successful communications.

	Deploying any form of QoS can also *reduce* the probability of
successful communications, by increasing the complexity of the network.

> Also keep in mind the tradeoff
> between complexity and security, and complexity and cost (i.e., 
> complicated
> deployments are often expensive deployments).

And the tradeoffs between:
	complexity and network availability
	complexity and mean-time-to-repair
	complexity and mean-time-between-failure

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 15:24:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA16011
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 15:24:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3KQYl12243
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 15:26:34 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3KQYv12240
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 15:26:34 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15956
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 15:23:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3KMJv11946;
	Tue, 3 Dec 2002 15:22:19 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3KJsv11717
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 15:19:54 -0500
Received: from imo-r07.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15366
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 15:16:58 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r07.mx.aol.com (mail_out_v34.13.) id j.1a8.cf7fd2e (4206);
	Tue, 3 Dec 2002 15:19:41 -0500 (EST)
Message-ID: <1a8.cf7fd2e.2b1e6bdc@aol.com>
Date: Tue, 3 Dec 2002 15:19:40 EST
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
To: rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_1a8.cf7fd2e.2b1e6bdc_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_1a8.cf7fd2e.2b1e6bdc_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/3/2002 1:21:13 PM Eastern Standard Time, 
rja@extremenetworks.com writes:


> On Tuesday, Dec 3, 2002, at 10:28 America/Montreal, Mpierce1@aol.com 
> wrote:
> > [MAP] So this says that congestion can occur and there is no way for 
> > the application (SIP) to deal with it, like by restricting new call 
> > setups, so the IP-layer is left completely on its own to deal with it, 
> > and the only way it can is by discarding packets.
> 
> Hmm.  Almost, but not quite right.  The application end-points, that is the
> application software on either end of the actual VoIP session, *can* and 
> *do*
> know that they are experiencing X% packet loss or that jitter is increasing
> from W to Z, or that latency has increased from A to B, or any combination
> of the above.  Having that knowledge, which is in the VoIP endpoints not in 
> the
> SIP server, the endpoints could react in any of several ways.  An example
> that seems most obvious to me would be to switch from G.711 to a more robust
> CODEC that consumes less bandwidth (e.g. some form of CELP).
> 


[MAP]
This is an interesting one. There's been a lot of talk about this idea in 
other places too. But let's examine it in reality.

If my application is handling my phone call, first, I don't know what 
functionality tells it about packet loss or latency or jitter at the other 
end, unless we add specific measuring protocols (which I thing are being 
worked on). If my application had a way to know about excess packet loss at 
the other end (presumably caused by there being too many other people trying 
to talk over the same link), what could it do about it? It could switch to a 
lower speed codec, thereby giving everyone a slighty lower packet loss ratio 
(including myself), but I would surely get worse quality from the lower speed 
codec. (If it were a better codec, I would have selected it in the first 
place.)

Or my application could switch to a higher bandwidth utilitzation, like 
adding more forward error correction to make up for the lost packets or to a 
higher rate codec that is less suseptable to packet loss, thereby giving 
everyone a slightly higher packet loss ratio (including myself), but I 
probably get better quality (at least as far as packet loss is concerned).

So which would you choose? Right! And everyone else would do the same (unless 
you're charged according to bandwidth) and the congestion and packet loss 
goes up and the net gain is negative.

This plan works if there is an intelligent entity in the middle that forces 
all users to switch to a lower rate codec (and accept lower quality). This is 
clearly not the answer in the Internet model.

Mike Pierce
Artel




--part1_1a8.cf7fd2e.2b1e6bdc_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/3/2002 1:21:13 PM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">On Tuesday, Dec 3, 2002, at 10:28 America/Montreal, Mpierce1@aol.com 
<BR>wrote:
<BR>&gt; [MAP] So this says that congestion can occur and there is no way for 
<BR>&gt; the application (SIP) to deal with it, like by restricting new call 
<BR>&gt; setups, so the IP-layer is left completely on its own to deal with it, 
<BR>&gt; and the only way it can is by discarding packets.
<BR>
<BR>Hmm. &nbsp;Almost, but not quite right. &nbsp;The application end-points, that is the
<BR>application software on either end of the actual VoIP session, *can* and *do*
<BR>know that they are experiencing X% packet loss or that jitter is increasing
<BR>from W to Z, or that latency has increased from A to B, or any combination
<BR>of the above. &nbsp;Having that knowledge, which is in the VoIP endpoints not in the
<BR>SIP server, the endpoints could react in any of several ways. &nbsp;An example
<BR>that seems most obvious to me would be to switch from G.711 to a more robust
<BR>CODEC that consumes less bandwidth (e.g. some form of CELP).
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>[MAP]
<BR>This is an interesting one. There's been a lot of talk about this idea in other places too. But let's examine it in reality.
<BR>
<BR>If my application is handling my phone call, first, I don't know what functionality tells it about packet loss or latency or jitter at the other end, unless we add specific measuring protocols (which I thing are being worked on). If my application had a way to know about excess packet loss at the other end (presumably caused by there being too many other people trying to talk over the same link), what could it do about it? It could switch to a lower speed codec, thereby giving everyone a slighty lower packet loss ratio (including myself), but I would surely get worse quality from the lower speed codec. (If it were a better codec, I would have selected it in the first place.)
<BR>
<BR>Or my application could switch to a higher bandwidth utilitzation, like adding more forward error correction to make up for the lost packets or to a higher rate codec that is less suseptable to packet loss, thereby giving everyone a slightly higher packet loss ratio (including myself), but I probably get better quality (at least as far as packet loss is concerned).
<BR>
<BR>So which would you choose? Right! And everyone else would do the same (unless you're charged according to bandwidth) and the congestion and packet loss goes up and the net gain is negative.
<BR>
<BR>This plan works if there is an intelligent entity in the middle that forces all users to switch to a lower rate codec (and accept lower quality). This is clearly not the answer in the Internet model.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR>
<BR>
<BR></FONT></HTML>

--part1_1a8.cf7fd2e.2b1e6bdc_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 15:24:12 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA16024
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 15:24:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3KQak12258
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 15:26:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3KQav12255
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 15:26:36 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15960
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 15:23:41 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3KMEv11925;
	Tue, 3 Dec 2002 15:22:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3KJrv11713
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 15:19:53 -0500
Received: from imo-r09.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA15363
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 15:16:57 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r09.mx.aol.com (mail_out_v34.13.) id j.20.3a387e9 (4206);
	Tue, 3 Dec 2002 15:19:42 -0500 (EST)
Message-ID: <20.3a387e9.2b1e6bde@aol.com>
Date: Tue, 3 Dec 2002 15:19:42 EST
Subject: Re: [Ieprep] Architectural considerations
To: rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_20.3a387e9.2b1e6bde_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_20.3a387e9.2b1e6bde_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/3/2002 1:26:55 PM Eastern Standard Time, 
rja@extremenetworks.com writes:


> The Internet has a fundamentally different architecture than the PSTN. To run
> voice over the Internet, one should adapt the voice application to the 
> Internet
> architecture rather than insisting that the Internet adopt the PSTN's broken
> (e.g. no dynamic routing of traffic around fibre cuts) architecture.
> 

[MAP] I'm always confused by statements like that last one that the PSTN 
can't do dynamic routing of traffic around fibre cuts. There are many lower 
layer schemes (SONET, etc) that do just that. I'm sure that a lot of the 
reliability of the Internet depends on such "rerouting" at the physical 
layer. At the "application" layer, every telephone switch recognizes when 
trunks to another switch are out of service and routes around them. Just like 
it routes around trunk groups that are congested (which the Internet 
apparently can't do).

I'd really be curious to understand what it is that you think the current 
PSTN can't do regarding routing that is a result of a "broken architecture".

MIke Pierce
Artel



--part1_20.3a387e9.2b1e6bde_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/3/2002 1:26:55 PM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">The Internet has a fundamentally different architecture than the PSTN. To run
<BR>voice over the Internet, one should adapt the voice application to the Internet
<BR>architecture rather than insisting that the Internet adopt the PSTN's broken
<BR>(e.g. no dynamic routing of traffic around fibre cuts) architecture.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>[MAP] I'm always confused by statements like that last one that the PSTN can't do dynamic routing of traffic around fibre cuts. There are many lower layer schemes (SONET, etc) that do just that. I'm sure that a lot of the reliability of the Internet depends on such "rerouting" at the physical layer. At the "application" layer, every telephone switch recognizes when trunks to another switch are out of service and routes around them. Just like it routes around trunk groups that are congested (which the Internet apparently can't do).
<BR>
<BR>I'd really be curious to understand what it is that you think the current PSTN can't do regarding routing that is a result of a "broken architecture".
<BR>
<BR>MIke Pierce
<BR>Artel
<BR>
<BR></FONT></HTML>

--part1_20.3a387e9.2b1e6bde_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 15:41:44 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA17067
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 15:41:44 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3Ki9Y13786
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 15:44:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Ki8v13783
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 15:44:08 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA17048
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 15:41:12 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Ke3v13636;
	Tue, 3 Dec 2002 15:40:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3KdSv13576
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 15:39:28 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA16818
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 15:36:33 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id CA7FA67104; Tue,  3 Dec 2002 12:04:59 -0500 (EST)
Date: Tue, 3 Dec 2002 15:39:21 -0500
Subject: Re: [Ieprep] Architectural considerations
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep@ietf.org
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <20.3a387e9.2b1e6bde@aol.com>
Message-Id: <4D843B43-06FF-11D7-A694-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 3, 2002, at 15:19 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] I'm always confused by statements like that last one that the 
> PSTN can't do dynamic routing of traffic around fibre cuts. There are 
> many lower layer schemes (SONET, etc) that do just that.

SONET rings are limited to 16 nodes, which is not the same capability 
as IP
has to route around arbitrary fibre cuts whether or not the physical 
topology
is a ring and over diverse layer-2 technologies mixed together 
arbitrarily.

Apologies if I wasn't sufficiently clear.

> I'm sure that a lot of the reliability of the Internet depends on such
> "rerouting" at the physical layer.

s/physical layer/Internet Layer/

	Now there can also be layer-2 fault recovery, such as Ethernet APS or
SONET APS, or FDDI rings, but that's independent and additional bonus
beyond what IP does regardless of the choice of underlying layer-2 
technology.

> At the "application" layer, every telephone switch recognizes when 
> trunks to another switch are out of service and routes around them. 
> Just like it routes around trunk groups that are congested (which the 
> Internet apparently can't do).

We do route around stuff, but we have a layered architecture with 
different
layering than the PSTN.  So the layers at which re-routing can happen 
are
different.  This yields different behaviours and properties between the 
two
kinds of networks.  So the layer in our system that does routing is 
unaware
of which packets are voice and which aren't.  In the PSTN, everything 
is a logical
circuit even if it is carrying packets instead of voice.

The main point is:
	Different architectures have different properties and require 
correspondingly
different approaches to meet a given objective.  Trying to apply 
Internet solutions
to PSTN issues is just as unlikely to work as applying PSTN solutions 
to Internet
issues.  The applications need to adapt to the actual network 
technology underneath.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 16:56:40 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20830
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 16:56:40 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3Lx5718734
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 16:59:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Lx5v18731
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 16:59:05 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20792
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 16:56:09 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3LwBv18613;
	Tue, 3 Dec 2002 16:58:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Lvjv18487
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 16:57:45 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20725
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 16:54:48 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 33E1967104; Tue,  3 Dec 2002 13:23:16 -0500 (EST)
Date: Tue, 3 Dec 2002 16:57:37 -0500
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others 
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Colin Perkins <csp@isi.edu>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <200212032049.gB3KnlN21049@chiron.nge.isi.edu>
Message-Id: <3C9B612C-070A-11D7-A694-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 3, 2002, at 15:49 America/Montreal, Colin Perkins wrote:
> The hope is that we can apply clue to VoIP implementations,

Colin,

I share that hope. :-)

> although that may not be realistic without policing. Certainly the RFCs
> that describe FEC use with RTP talk about these issues.

For folks like me who aren't up on RTP, could you please list some
assigned-reading by listing the most relevant RFCs on this thread
(including the current incarnation of the base RTP spec) ??

Thanks,

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 17:39:53 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23006
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 17:39:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3MgHi22755
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 17:42:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3MgHv22752
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 17:42:17 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA22970
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 17:39:22 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Mc7v22534;
	Tue, 3 Dec 2002 17:38:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Mbrv22516
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 17:37:53 -0500
Received: from imo-m08.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA22897
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 17:34:58 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m08.mx.aol.com (mail_out_v34.13.) id k.14b.1849b180 (3699);
	Tue, 3 Dec 2002 17:37:38 -0500 (EST)
Message-ID: <14b.1849b180.2b1e8c32@aol.com>
Date: Tue, 3 Dec 2002 17:37:38 EST
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others 
To: csp@isi.edu
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_14b.1849b180.2b1e8c32_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_14b.1849b180.2b1e8c32_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/3/2002 3:52:30 PM Eastern Standard Time, csp@isi.edu 
writes:


> If you implement RTP correctly, the RTP control protocol provides feedback
> on the packet loss, latency and jitter at the other end [RFC 1889].
> 

But:

In a message dated 12/2/2002 1:42:11 PM Eastern Standard Time, 
sob@harvard.edu writes:


> > 2. new call attempts would have been rejected
> 
> no- the congested link would not cause the SIP proxies to reject calls
> n


[MAP] Guess I'm missing something, since the RTP functionality mentioned by 
Colin is the type of thing that I thought would allow a call control 
application to know that it should stop trying to set up more calls. Scott 
(and others) say there is no way to do it.

Mike Pierce
Artel



--part1_14b.1849b180.2b1e8c32_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/3/2002 3:52:30 PM Eastern Standard Time, csp@isi.edu writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">If you implement RTP correctly, the RTP control protocol provides feedback
<BR>on the packet loss, latency and jitter at the other end [RFC 1889].
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>But:
<BR>
<BR>In a message dated 12/2/2002 1:42:11 PM Eastern Standard Time, sob@harvard.edu writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">&gt; 2. new call attempts would have been rejected
<BR>
<BR>no- the congested link would not cause the SIP proxies to reject calls
<BR>since they have no knowledge of the congestio</BLOCKQUOTE>n</FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>
<BR>[MAP] Guess I'm missing something, since the RTP functionality mentioned by Colin is the type of thing that I thought would allow a call control application to know that it should stop trying to set up more calls. Scott (and others) say there is no way to do it.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR>
<BR></FONT></HTML>

--part1_14b.1849b180.2b1e8c32_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 17:42:41 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23159
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 17:42:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3Mj5T22903
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 17:45:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Mj5v22900
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 17:45:05 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23121
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 17:42:10 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Mf2v22711;
	Tue, 3 Dec 2002 17:41:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Meqv22691
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 17:40:52 -0500
Received: from cs.columbia.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA22954
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 17:37:56 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB3Mefl8017964
	(version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NOT);
	Tue, 3 Dec 2002 17:40:42 -0500 (EST)
Received: from cs.columbia.edu (tallgrass.netlab.uky.edu [204.198.76.66])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB3MeddG015291
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Tue, 3 Dec 2002 17:40:40 -0500 (EST)
Message-ID: <3DED3264.5010309@cs.columbia.edu>
Date: Tue, 03 Dec 2002 17:38:28 -0500
From: Henning Schulzrinne <hgs@cs.columbia.edu>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.2.1) Gecko/20021130
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: RJ Atkinson <rja@extremenetworks.com>
CC: Colin Perkins <csp@isi.edu>, ieprep <ieprep@ietf.org>
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
References: <3C9B612C-070A-11D7-A694-00039357A82A@extremenetworks.com>
In-Reply-To: <3C9B612C-070A-11D7-A694-00039357A82A@extremenetworks.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

See http://www.cs.columbia.edu/~hgs/rtp for a reading list.

RJ Atkinson wrote:
> 
> On Tuesday, Dec 3, 2002, at 15:49 America/Montreal, Colin Perkins wrote:
> 
>> The hope is that we can apply clue to VoIP implementations,
> 
> 
> Colin,
> 
> I share that hope. :-)
> 
>> although that may not be realistic without policing. Certainly the RFCs
>> that describe FEC use with RTP talk about these issues.
> 
> 
> For folks like me who aren't up on RTP, could you please list some
> assigned-reading by listing the most relevant RFCs on this thread
> (including the current incarnation of the base RTP spec) ??
> 
> Thanks,
> 
> Ran
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 17:52:39 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23562
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 17:52:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3Mt3Z23443
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 17:55:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Mt3v23440
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 17:55:03 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23516
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 17:52:07 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3MpDv23192;
	Tue, 3 Dec 2002 17:51:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3MoLv23147
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 17:50:21 -0500
Received: from newdev.harvard.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23282
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 17:47:26 -0500 (EST)
Received: (from sob@localhost)
	by newdev.harvard.edu (8.12.2/8.12.2) id gB3MnevD006765;
	Tue, 3 Dec 2002 17:49:40 -0500 (EST)
Date: Tue, 3 Dec 2002 17:49:40 -0500 (EST)
From: Scott  Bradner <sob@harvard.edu>
Message-Id: <200212032249.gB3MnevD006765@newdev.harvard.edu>
To: csp@isi.edu, Mpierce1@aol.com
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Cc: ieprep@ietf.org
In-Reply-To: <14b.1849b180.2b1e8c32@aol.com>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

> [MAP] Guess I'm missing something, since the RTP functionality mentioned by 
> Colin is the type of thing that I thought would allow a call control 
> application to know that it should stop trying to set up more calls. Scott 
> (and others) say there is no way to do it.

a client (i.e. single phone) that already has a call going to point
X could tell that there is a problem with getting to point X and might
not set up more calls to point X (if setting up more calls to the
same point was something that it wanted to do) but it would have
no way to know if calls to point Y would find any problems since
there is no way for the phone to know what in the path from
the phone to point X was causing the lossage and if that same 
thing (link I would guess) is also in the path to point Y

Scott
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 17:57:42 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23710
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 17:57:42 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3N06v23651
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 18:00:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3N06v23648
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 18:00:06 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23689
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 17:57:11 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Mu2v23508;
	Tue, 3 Dec 2002 17:56:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3MtXv23463
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 17:55:33 -0500
Received: from multicasttech.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23558
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 17:52:37 -0500 (EST)
Received: from dhcp-office-44.multicasttech.com (account marshall_eubanks [63.105.122.236] verified)
  by multicasttech.com (CommuniGate Pro SMTP 3.4.8)
  with ESMTP-TLS id 1527313; Tue, 03 Dec 2002 17:55:25 -0500
Date: Tue, 3 Dec 2002 17:55:24 -0500
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others 
Content-Type: multipart/alternative; boundary=Apple-Mail-2-519382607
Mime-Version: 1.0 (Apple Message framework v482)
Cc: csp@isi.edu, ieprep@ietf.org
To: Mpierce1@aol.com
From: Marshall Eubanks <tme@multicasttech.com>
In-Reply-To: <14b.1849b180.2b1e8c32@aol.com>
Message-Id: <4F04A5A2-0712-11D7-8480-0030656561EA@multicasttech.com>
X-Mailer: Apple Mail (2.482)
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--Apple-Mail-2-519382607
Content-Type: text/plain;
	charset=US-ASCII;
	format=flowed
Content-Transfer-Encoding: 7bit

Colin et al;

I thought that the RTP spec specifically does _not_ mandate that RTCP be
implemented to "implement RTP correctly". I distinctly remember arguing 
otherwise in AVT meetings and being shot down.

Regards
Marshall Eubanks

On Tuesday, December 3, 2002, at 05:37 PM, Mpierce1@aol.com wrote:

> In a message dated 12/3/2002 3:52:30 PM Eastern Standard Time, 
> csp@isi.edu writes:
>
>
> If you implement RTP correctly, the RTP control protocol provides 
> feedback
> on the packet loss, latency and jitter at the other end [RFC 1889].
>
>
>
> But:
>
> In a message dated 12/2/2002 1:42:11 PM Eastern Standard Time, 
> sob@harvard.edu writes:
>
>
> > 2. new call attempts would have been rejected
>
> no- the congested link would not cause the SIP proxies to reject calls
> since they have no knowledge of the congestio
>
> n
>
>
> [MAP] Guess I'm missing something, since the RTP functionality 
> mentioned by Colin is the type of thing that I thought would allow a 
> call control application to know that it should stop trying to set up 
> more calls. Scott (and others) say there is no way to do it.
>
> Mike Pierce
> Artel
>

--Apple-Mail-2-519382607
Content-Type: text/enriched;
	charset=US-ASCII
Content-Transfer-Encoding: 7bit

Colin et al;


I thought that the RTP spec specifically does _not_ mandate that RTCP
be

implemented to "implement RTP correctly". I distinctly remember
arguing otherwise in AVT meetings and being shot down.


Regards

Marshall Eubanks


On Tuesday, December 3, 2002, at 05:37 PM, Mpierce1@aol.com wrote:


<excerpt><fontfamily><param>Arial</param><smaller>In a message dated
12/3/2002 3:52:30 PM Eastern Standard Time, csp@isi.edu writes:



</smaller></fontfamily>If you implement RTP correctly, the RTP control
protocol provides feedback

on the packet loss, latency and jitter at the other end [RFC 1889].




<fontfamily><param>Arial</param><color><param>0000,0000,0000</param><smaller>But:


In a message dated 12/2/2002 1:42:11 PM Eastern Standard Time,
sob@harvard.edu writes:



> 2. new call attempts would have been rejected


no- the congested link would not cause the SIP proxies to reject calls

since they have no knowledge of the congestio


n



</smaller>[MAP] Guess I'm missing something, since the RTP
functionality mentioned by Colin is the type of thing that I thought
would allow a call control application to know that it should stop
trying to set up more calls. Scott (and others) say there is no way to
do it.


Mike Pierce

Artel


</color></fontfamily></excerpt>
--Apple-Mail-2-519382607--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 18:05:47 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA23975
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 18:05:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3N8CT24699
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 18:08:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3N8Cv24696
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 18:08:12 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA23969
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 18:05:16 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3N43v23854;
	Tue, 3 Dec 2002 18:04:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3N3tv23840
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 18:03:55 -0500
Received: from cs.columbia.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA23828
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 18:00:55 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB3N3fl8020165
	(version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NOT);
	Tue, 3 Dec 2002 18:03:41 -0500 (EST)
Received: from cs.columbia.edu (tallgrass.netlab.uky.edu [204.198.76.66])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB3N3ddG017514
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Tue, 3 Dec 2002 18:03:40 -0500 (EST)
Message-ID: <3DED37C8.4090808@cs.columbia.edu>
Date: Tue, 03 Dec 2002 18:01:28 -0500
From: Henning Schulzrinne <hgs@cs.columbia.edu>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.2.1) Gecko/20021130
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Marshall Eubanks <tme@multicasttech.com>
CC: Mpierce1@aol.com, csp@isi.edu, ieprep@ietf.org
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
References: <4F04A5A2-0712-11D7-8480-0030656561EA@multicasttech.com>
In-Reply-To: <4F04A5A2-0712-11D7-8480-0030656561EA@multicasttech.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

You would probably be shot down by the AVT chairs (with me lending cover 
from the microphone) if you claimed that RTCP is not needed for a 
compliant implementation. If the RTP spec says anywhere that RTCP is 
optional, please let me know and we'll fix this.

Marshall Eubanks wrote:
> Colin et al;
> 
> I thought that the RTP spec specifically does _not_ mandate that RTCP be
> implemented to "implement RTP correctly". I distinctly remember arguing 
> otherwise in AVT meetings and being shot down.
> 
> Regards
> Marshall Eubanks
> 

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 18:06:50 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA24020
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 18:06:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3N9Em24734
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 18:09:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3N9Ev24731
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 18:09:14 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA23989
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 18:06:19 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3N50v23898;
	Tue, 3 Dec 2002 18:05:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3N4Kv23864
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 18:04:20 -0500
Received: from chiron.nge.isi.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA23856
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 18:01:25 -0500 (EST)
Received: from chiron (csp@localhost)
	by chiron.nge.isi.edu (8.11.6/8.11.6) with ESMTP id gB3N4Do22302;
	Tue, 3 Dec 2002 18:04:13 -0500
Message-Id: <200212032304.gB3N4Do22302@chiron.nge.isi.edu>
To: Marshall Eubanks <tme@multicasttech.com>
cc: Mpierce1@aol.com, ieprep@ietf.org
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others 
In-Reply-To: Your message of "Tue, 03 Dec 2002 17:55:24 EST."
             <4F04A5A2-0712-11D7-8480-0030656561EA@multicasttech.com> 
Date: Tue, 03 Dec 2002 18:04:13 -0500
From: Colin Perkins <csp@isi.edu>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

--> Marshall Eubanks writes:
>Colin et al;
>
>I thought that the RTP spec specifically does _not_ mandate that RTCP be
>implemented to "implement RTP correctly". I distinctly remember arguing 
>otherwise in AVT meetings and being shot down.

Section 6 of draft-ietf-avt-rtp-new-11.txt says RTCP "SHOULD be used in all
environments, but particularly in the IP multicast environment" (although
the profile says that "the RTCP bandwidth for non-data-senders MAY be set
to zero when operating on unidirectional links or for sessions that don't
require feedback on the quality of reception").

Colin
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec  3 18:13:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA24295
	for <ieprep-archive@odin.ietf.org>; Tue, 3 Dec 2002 18:13:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB3NFZo24998
	for ieprep-archive@odin.ietf.org; Tue, 3 Dec 2002 18:15:35 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3NFZv24995
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 3 Dec 2002 18:15:35 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA24273
	for <ieprep-web-archive@ietf.org>; Tue, 3 Dec 2002 18:12:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3ND2v24867;
	Tue, 3 Dec 2002 18:13:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3NC5v24834
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 18:12:05 -0500
Received: from c015.snv.cp.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id SAA24100
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 18:09:09 -0500 (EST)
From: alan.duric@globalipsound.com
Received: (cpmta 27621 invoked from network); 3 Dec 2002 15:11:57 -0800
Received: from 209.228.35.115 (HELO mail.globalipsound.com.criticalpath.net)
  by smtp.globalipsound.com (209.228.35.122) with SMTP; 3 Dec 2002 15:11:57 -0800
X-Sent: 3 Dec 2002 23:11:57 GMT
Received: from [217.209.112.247] by mail.globalipsound.com with HTTP;
    Tue, 03 Dec 2002 15:11:55 -0800 (PST)
Content-Type: text/plain; charset=iso-8859-1
Content-Disposition: inline
MIME-Version: 1.0
To: Mpierce1@aol.com, ieprep@ietf.org
Cc: rja@extremenetworks.com
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
X-Sent-From: alan.duric@globalipsound.com
Date: Tue, 03 Dec 2002 15:11:55 -0800 (PST)
X-Mailer: Web Mail 5.1.1-3_sol28
Message-Id: <20021203151157.6509.h000.c015.wm@mail.globalipsound.com.criticalpath.net>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gB3NC5v24835
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

[RAN]
>the endpoints could react in any of several ways.  An
example
>that seems most obvious to me would be to switch from
G.711 to a more robust
>CODEC that consumes less bandwidth (e.g. some form of
CELP).

[Alan]
I could agree with the switching (see bellow) to lower
bit rate, but would disagree with switching to CELP,
which is bad since CELP is statefull, packets have
memory and due to this interpacket memory dependency
error propagates further, so by losing packet number N
You are losing properties of the packet N+1, and it
takes quite  some time for CELP to recover state,
resulting with degraded quality. See:
http://www.ietf.org/proceedings/02mar/slides/avt-6.pdf
http://www.cs.columbia.edu/~library/TR-repository/reports/reports-2001/cucs-016-01.pdf
moreover, I can provide You with a great deal of other
references. This behavior is not strange, if we bear on
mind that principally CELP codecs were not designed for
usage in packet networks rather for wireless or circuit
switched nets, where they supposed to cope with bit
errors instead of packet/frame loss.
 
[MAP]
>If my application is handling my phone call, first, I
don't know what functionality tells it about packet
loss or latency or jitter at the other end, 
unless we add specific measuring protocols (which I
thing are being worked on). If my application had a way
to know about excess packet loss at the other end
(presumably caused by there being too many other people
trying to talk over the same link), what could it do
about it? It could switch to a lower speed codec,
thereby giving everyone a slighty lower packet loss
ratio (including myself), but I would surely get worse
quality from the lower speed codec. (If it were a
better codec, I would have selected it in the first
place.)

[Alan]
This change in basic (with no impact of impaired net)
perceived quality is almost negligible for average
users for a number of low bitrate codecs..


[MAP]
>Or my application could switch to a higher bandwidth
utilitzation, like adding more forward error correction
to make up for the lost packets or to a higher rate
>codec that is less suseptable to packet loss, thereby
giving everyone a slightly higher packet loss ratio
(including myself), but I probably get better quality
(at >least as far as packet loss is concerned).

[Alan]
I would not agree with this solution since it will most
likely cause even greater packet loss if number of
other endpoints are doing the same thing, furthermore
calls that do not increase bit rate have better chances
to have smaller packet loss and delay values (since
number of routers on slow access link use WFQ as a
default and recommended mechanism)

[MAP]
>This plan works if there is an intelligent entity in
the middle that forces all users to switch to a lower
rate codec (and accept lower quality). This is clearly
not t>he answer in the Internet model.

[Alan]
I agree. Therefore  we should utilize what we have
available in terms of codec technology, and other
endpoint specific technologies that will improve
quality such as advanced jitter buffers, packet loss
concealment mechanisms etc. etc. …  and eventually come
up with some guidelines for that part.

Best regards,
Alan Duric
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec  4 12:38:17 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA24336
	for <ieprep-archive@odin.ietf.org>; Wed, 4 Dec 2002 12:38:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB4Hed406380
	for ieprep-archive@odin.ietf.org; Wed, 4 Dec 2002 12:40:39 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4Hedv06377
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 4 Dec 2002 12:40:39 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA24272
	for <ieprep-web-archive@ietf.org>; Wed, 4 Dec 2002 12:37:45 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4HdUv06242;
	Wed, 4 Dec 2002 12:39:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4Hccv06195
	for <ieprep@optimus.ietf.org>; Wed, 4 Dec 2002 12:38:38 -0500
Received: from sj-msg-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA24169
	for <ieprep@ietf.org>; Wed, 4 Dec 2002 12:35:44 -0500 (EST)
Received: from imop.cisco.com (IDENT:mirapoint@imop.cisco.com [171.69.11.44])
	by sj-msg-core-1.cisco.com (8.12.2/8.12.2) with ESMTP id gB4HcVK0019952;
	Wed, 4 Dec 2002 09:38:31 -0800 (PST)
Received: from cisco.com (ssh-sjc-1.cisco.com [171.68.225.134])
	by imop.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABK08491;
	Wed, 4 Dec 2002 09:35:25 -0800 (PST)
Date: Wed, 4 Dec 2002 09:38:45 -0800
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: Mpierce1@aol.com, ieprep@ietf.org
To: RJ Atkinson <rja@extremenetworks.com>
From: Rohan Mahy <rohan@cisco.com>
In-Reply-To: <DA9FF9A2-06EB-11D7-A694-00039357A82A@extremenetworks.com>
Message-Id: <3CC03986-07AF-11D7-9AE1-0003938AF740@cisco.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gB4Hccv06196
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

Hi,

Adding a bit of application-specific detail to Ran's comments...

On Tuesday, December 3, 2002, at 10:20 AM, RJ Atkinson wrote:
> On Tuesday, Dec 3, 2002, at 10:28 America/Montreal, Mpierce1@aol.com 
> wrote:
>> [MAP] So this says that congestion can occur and there is no way for 
>> the application (SIP) to deal with it, like by restricting new call 
>> setups, so the IP-layer is left completely on its own to deal with 
>> it, and the only way it can is by discarding packets.
>
> Hmm.  Almost, but not quite right.  The application end-points, that 
> is the
> application software on either end of the actual VoIP session, *can* 
> and *do*
> know that they are experiencing X% packet loss or that jitter is 
> increasing
> from W to Z, or that latency has increased from A to B, or any 
> combination
> of the above.  Having that knowledge, which is in the VoIP endpoints 
> not in the
> SIP server, the endpoints could react in any of several ways.  An 
> example
> that seems most obvious to me would be to switch from G.711 to a more 
> robust
> CODEC that consumes less bandwidth (e.g. some form of CELP).

A very good choice here is iLBC (Internet Low Bitrate Codec), a 
royalty-free low-bitrate codec designed for use on the Internet, and 
therefore very tolerant of loss, jitter, and delay.

> Now I'm not an application-layer guy very often, so I don't know the 
> details
> of SIP or the details of how the voice is being sent (presumably over 
> RTP/UDP/IP
> somehow).  However, it is clear that protocol engineering could be 
> done to add
> this kind of capability (e.g. codec negotiation) if folks in that part 
> of the
> application community cared to do it.

It's already there.

>> [RJA] Now one could use IP ToS bits to mark the packets from inside a 
>> domain for
>> the uplink router to prioritise, if the uplink router happened to 
>> have that
>> policy configured in it by the router's operator.  One could also put 
>> various
>> other policies into the router based on some combinations of 
>> (address, protocol, port),
>> if one did not want to use the IP ToS bits.  If the link is Ethernet, 
>> one could
>> use 802.1p.  If the link is something else, then one could use other 
>> link-specific
>> mechanisms (e.g. ATM has PSTN-like calls).
>>
>> [MAP] I think we've agreed that packet "prioritization" of one voice 
>> packet before another is not feasible or a good idea, so I don't 
>> think you mean that. I can't imagine that handling of packets for 
>> priority calls could be done by putting specific addresses in the 
>> routers. (Sounds like "state" to me anyway.)
>
> In the text from me that you've quoted just above, one concept could 
> be that
> IEprep traffic (whether VoIP or other) *could* be marked using IP ToS 
> bits
> (either at origin, by first-hop router inside the domain, or by some 
> other
> router inside the domain) and (subject to local network operator's 
> policy)
> the IP ToS bits in each packet could be examined by the domain's 
> uplink router
> to give IEprep packets lower drop probability using some 
> (policy-dependent)
> queuing configuration inside the router (e.g. Sprint's CoS stuff, which
> multiple router vendors support in COTS product today).
>
> An advantage to this approach over a NSIS/RSVP-like approach is that 
> one
> has a relatively small amount of total state for the QoS, 
> pre-configured
> into the router, rather than adding router state for each additional
> IP-layer flow that happens to be "preferred".  Basically it adds a 
> small
> amount of constant state to handle an unlimited number of concurrent 
> flows,
> rather than adding state for each additional flow that might be 
> present.
> The issue with state in routers is always how much one has.  This is 
> an approach
> to keep the total state related to QoS small regardless of traffic 
> load or mix.

>> > [MAP] So I take that as an agreement with what I said, that in the
>> > case of congestion, a lot of packets would be lost.
>>
>> [RJA] No.  The number of packets lost would vary with (offered load, 
>> link capacity,
>> traffic pattern, packet memory depth on each side of the link).  In 
>> many cases
>> there is minimal or no congestion, because the router buffer is deep 
>> enough
>> to handle a trans-oceanic RTT, which in turn gives TCP or SCTP enough 
>> time
>> to reduce the transmission rate on existing sessions, which then 
>> leaves
>> enough room for the new traffic to be carried on the link.
>>
>> [MAP] Your answer relates to elastic traffic, e.g., file transfers, 
>> where the buffer can be made as "deep" as you have memory available. 
>> The primary subject here is inelastic traffic (voice) in which there 
>> are relatively strict limits on the buffer length.
>
> Routers have no idea whether a packet is voice or file transfer.  The 
> amount
> of packet memory in the router is a property of what humans have 
> physically
> inserted.  So my comment quoted above is independent of what kind of 
> application
> happens to be running on the packets being buffered.
>
> Variable latency is a fact of life in any packet network.  All 
> Internet-based
> applications, including voice, have to be reasonably tolerant of some 
> amount
> of jitter or they won't work at all.
>
> An IAB comment by Sally Floyd at the IEprep meeting in Atlanta bears 
> repeating.
> Because the current VoIP stuff does not include congestion avoidance, 
> the tendency
> will be for the VoIP traffic to "crowd out" the congestion avoiding 
> traffic
> (e.g. anything running over TCP or SCTP).  This is a major problem.  
> Somehow
> the community needs to add congestion avoidance into the VoIP 
> applications
> protocols.

Actually there is work now to do this. The goal is to run RTP and RTCP 
over DCCP (Datagram Congestion Control Protocol).

thanks,
-rohan

>> [MAP] Our goal is to point to a specific set of standards (or 
>> proposed) to describe how to provide preferential treatment. Can you 
>> point me to some specific ones? Os is it naive to think that it is as 
>> simple as referrencing EF or some such RFC?
>
> If one wants to talk about standards-track QoS mechanisms, there are a 
> wide range of
> RFCs to choose from.  They each have different properties.  Each is 
> suited for somewhat
> different operating conditions.  There is no "one size fits all" RFC.
>
> It is, however, important to keep in mind that IETF avoids 
> recommending or dictating
> that an operator deploy particular standards.  Whether an ISP chooses 
> to deploy
> anything (e.g. OSPF routing protocol vs IS-IS routing protocol vs 
> some-proprietary
> routing protocol) is *entirely* a decision of that ISP.
>
> For many reasons, including some discussed by the IAB in RFC-2990, 
> Sections 3.10
> through 3.12, the deployment of inter-domain QoS is impractical and 
> improbable
> anytime in the near future.  So if one wants to offer a robust service 
> over a
> multi-domain environment, one would be well advised to focus on 
> ensuring that
> one uses robust applications (e.g. very thoughtful codec selection, 
> not limited
> just to ITU specs, in the case of VoIP applications).
>
> Ran
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec  4 13:02:39 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA25596
	for <ieprep-archive@odin.ietf.org>; Wed, 4 Dec 2002 13:02:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB4I52U07505
	for ieprep-archive@odin.ietf.org; Wed, 4 Dec 2002 13:05:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4I52v07502
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 4 Dec 2002 13:05:02 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA25579
	for <ieprep-web-archive@ietf.org>; Wed, 4 Dec 2002 13:02:08 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4I4Dv07471;
	Wed, 4 Dec 2002 13:04:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4I35v07412
	for <ieprep@optimus.ietf.org>; Wed, 4 Dec 2002 13:03:05 -0500
Received: from sj-msg-core-4.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA25487
	for <ieprep@ietf.org>; Wed, 4 Dec 2002 13:00:12 -0500 (EST)
Received: from imop.cisco.com (IDENT:mirapoint@imop.cisco.com [171.69.11.44])
	by sj-msg-core-4.cisco.com (8.12.2/8.12.2) with ESMTP id gB4I2tLW018851;
	Wed, 4 Dec 2002 10:02:55 -0800 (PST)
Received: from cisco.com (ssh-sjc-1.cisco.com [171.68.225.134])
	by imop.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABK08827;
	Wed, 4 Dec 2002 09:59:47 -0800 (PST)
Date: Wed, 4 Dec 2002 10:03:07 -0800
Subject: Re: [Ieprep] Architectural considerations
Content-Type: text/plain; charset=WINDOWS-1252; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: Mpierce1@aol.com, ieprep <ieprep@ietf.org>
To: RJ Atkinson <rja@extremenetworks.com>
From: Rohan Mahy <rohan@cisco.com>
In-Reply-To: <BC9FFBF0-06EC-11D7-A694-00039357A82A@extremenetworks.com>
Message-Id: <A481CDF2-07B2-11D7-9AE1-0003938AF740@cisco.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gB4I36v07413
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


On Tuesday, Dec 3, 2002, at 10:28 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] Yes, I understand that 10 independent SIP providers can’t know 
> about the link loading in order to decide whether or not to allow a 
> new call to be set up (which is going to “fail” anyway due to bad 
> audio). This is a major shortcoming with SIP and VoIP and IP routing 
> and until it is recognized, admitted, and fixed, it can not be used 
> for real telephone service (much less emergency services).

Mike,

I don't need a "SIP Provider" to get telephone services, I just need a 
gateway provider and an Internet connection. These can be unrelated if 
I so desire.  In my case, I am using cliq.com for internet services, 
and deltathree.com for gateway services. While you can certainly decide 
if a technology is suitable for *your* phone calls and only use 
providers that meet your criteria, nobody died and made you king.  
Fortunately for the rest of us, the market gets to decide when the 
Internet *can* be used for real telephone service. In some cases 
various governments will attempt to prevent the market from deciding 
and they will either give up (India) or merely fairly miserably 
(Panama).

thanks,
-rohan

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec  4 13:06:36 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA25869
	for <ieprep-archive@odin.ietf.org>; Wed, 4 Dec 2002 13:06:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB4I8x608303
	for ieprep-archive@odin.ietf.org; Wed, 4 Dec 2002 13:08:59 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4I8xv08300
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 4 Dec 2002 13:08:59 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA25850
	for <ieprep-web-archive@ietf.org>; Wed, 4 Dec 2002 13:06:04 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4I83v08263;
	Wed, 4 Dec 2002 13:08:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4I7Bv07972
	for <ieprep@optimus.ietf.org>; Wed, 4 Dec 2002 13:07:11 -0500
Received: from sj-msg-core-1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA25740
	for <ieprep@ietf.org>; Wed, 4 Dec 2002 13:04:17 -0500 (EST)
Received: from imop.cisco.com (IDENT:mirapoint@imop.cisco.com [171.69.11.44])
	by sj-msg-core-1.cisco.com (8.12.2/8.12.2) with ESMTP id gB4I74K0007926;
	Wed, 4 Dec 2002 10:07:04 -0800 (PST)
Received: from cisco.com (ssh-sjc-1.cisco.com [171.68.225.134])
	by imop.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABK08910;
	Wed, 4 Dec 2002 10:03:57 -0800 (PST)
Date: Wed, 4 Dec 2002 10:07:16 -0800
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Content-Type: multipart/alternative; boundary=Apple-Mail-31-588494679
Mime-Version: 1.0 (Apple Message framework v548)
Cc: rja@extremenetworks.com, ieprep@ietf.org
To: Mpierce1@aol.com
From: Rohan Mahy <rohan@cisco.com>
In-Reply-To: <1a8.cf7fd2e.2b1e6bdc@aol.com>
Message-Id: <3905CB50-07B3-11D7-9AE1-0003938AF740@cisco.com>
X-Mailer: Apple Mail (2.548)
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--Apple-Mail-31-588494679
Content-Type: text/plain;
	charset=ISO-8859-1;
	format=flowed
Content-Transfer-Encoding: quoted-printable


On Tuesday, December 3, 2002, at 12:19 PM, Mpierce1@aol.com wrote:

> In a message dated 12/3/2002 1:21:13 PM Eastern Standard Time,=20
> rja@extremenetworks.com writes:
>
>
> On Tuesday, Dec 3, 2002, at 10:28 America/Montreal, Mpierce1@aol.com
> wrote:
> > [MAP] So this says that congestion can occur and there is no way for
> > the application (SIP) to deal with it, like by restricting new call
> > setups, so the IP-layer is left completely on its own to deal with=20=

> it,
> > and the only way it can is by discarding packets.
>
> Hmm. =A0Almost, but not quite right. =A0The application end-points, =
that=20
> is the
> application software on either end of the actual VoIP session, *can*=20=

> and *do*
> know that they are experiencing X% packet loss or that jitter is=20
> increasing
> from W to Z, or that latency has increased from A to B, or any=20
> combination
> of the above. =A0Having that knowledge, which is in the VoIP endpoints=20=

> not in the
> SIP server, the endpoints could react in any of several ways. =A0An=20
> example
> that seems most obvious to me would be to switch from G.711 to a more=20=

> robust
> CODEC that consumes less bandwidth (e.g. some form of CELP).
>
>
>
>
> [MAP]
> This is an interesting one. There's been a lot of talk about this idea=20=

> in other places too. But let's examine it in reality.
>
> If my application is handling my phone call, first, I don't know what=20=

> functionality tells it about packet loss or latency or jitter at the=20=

> other end, unless we add specific measuring protocols (which I thing=20=

> are being worked on).

That thing is called RTCP and has been an integral part of RTP since=20
its inception.

> If my application had a way to know about excess packet loss at the=20
> other end (presumably caused by there being too many other people=20
> trying to talk over the same link), what could it do about it? It=20
> could switch to a lower speed codec, thereby giving everyone a slighty=20=

> lower packet loss ratio (including myself), but I would surely get=20
> worse quality from the lower speed codec. (If it were a better codec,=20=

> I would have selected it in the first place.)

In the case of iLBC, you would consume more CPU on your phone or=20
computer but your quality would improve (from loss concealment) more=20
than it would deteriorate (from compression).   so it actually benefits=20=

you to switch, and it becomes even better if other switch as well.

thanks,
-rohan

> Or my application could switch to a higher bandwidth utilitzation,=20
> like adding more forward error correction to make up for the lost=20
> packets or to a higher rate codec that is less suseptable to packet=20
> loss, thereby giving everyone a slightly higher packet loss ratio=20
> (including myself), but I probably get better quality (at least as far=20=

> as packet loss is concerned).
>
> So which would you choose? Right! And everyone else would do the same=20=

> (unless you're charged according to bandwidth) and the congestion and=20=

> packet loss goes up and the net gain is negative.
>
> This plan works if there is an intelligent entity in the middle that=20=

> forces all users to switch to a lower rate codec (and accept lower=20
> quality). This is clearly not the answer in the Internet model.
>
> Mike Pierce
> Artel
>
>

--Apple-Mail-31-588494679
Content-Type: text/enriched;
	charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable



On Tuesday, December 3, 2002, at 12:19 PM, Mpierce1@aol.com wrote:


<excerpt><fontfamily><param>Arial</param><smaller>In a message dated
12/3/2002 1:21:13 PM Eastern Standard Time, rja@extremenetworks.com
writes:



</smaller></fontfamily>On Tuesday, Dec 3, 2002, at 10:28
America/Montreal, Mpierce1@aol.com

wrote:

> [MAP] So this says that congestion can occur and there is no way for

> the application (SIP) to deal with it, like by restricting new call

> setups, so the IP-layer is left completely on its own to deal with
it,

> and the only way it can is by discarding packets.


Hmm. =A0Almost, but not quite right. =A0The application end-points, that
is the

application software on either end of the actual VoIP session, *can*
and *do*

know that they are experiencing X% packet loss or that jitter is
increasing

from W to Z, or that latency has increased from A to B, or any
combination

of the above. =A0Having that knowledge, which is in the VoIP endpoints
not in the

SIP server, the endpoints could react in any of several ways. =A0An
example

that seems most obvious to me would be to switch from G.711 to a more
robust

CODEC that consumes less bandwidth (e.g. some form of CELP).





=
<fontfamily><param>Arial</param><color><param>0000,0000,0000</param><small=
er>[MAP]

This is an interesting one. There's been a lot of talk about this idea
in other places too. But let's examine it in reality.


If my application is handling my phone call, first, I don't know what
functionality tells it about packet loss or latency or jitter at the
other end, unless we add specific measuring protocols (which I thing
are being worked on).=20

</smaller></color></fontfamily></excerpt>

That thing is called RTCP and has been an integral part of RTP since
its inception.


=
<excerpt><fontfamily><param>Arial</param><color><param>0000,0000,0000</par=
am><smaller>If
my application had a way to know about excess packet loss at the other
end (presumably caused by there being too many other people trying to
talk over the same link), what could it do about it? It could switch
to a lower speed codec, thereby giving everyone a slighty lower packet
loss ratio (including myself), but I would surely get worse quality
from the lower speed codec. (If it were a better codec, I would have
selected it in the first place.)

</smaller></color></fontfamily></excerpt>

In the case of iLBC, you would consume more CPU on your phone or
computer but your quality would improve (from loss concealment) more
than it would deteriorate (from compression).   so it actually
benefits you to switch, and it becomes even better if other switch as
well.


thanks,

-rohan


=
<excerpt><fontfamily><param>Arial</param><color><param>0000,0000,0000</par=
am><smaller>Or
my application could switch to a higher bandwidth utilitzation, like
adding more forward error correction to make up for the lost packets
or to a higher rate codec that is less suseptable to packet loss,
thereby giving everyone a slightly higher packet loss ratio (including
myself), but I probably get better quality (at least as far as packet
loss is concerned).


So which would you choose? Right! And everyone else would do the same
(unless you're charged according to bandwidth) and the congestion and
packet loss goes up and the net gain is negative.


This plan works if there is an intelligent entity in the middle that
forces all users to switch to a lower rate codec (and accept lower
quality). This is clearly not the answer in the Internet model.


Mike Pierce

Artel



</smaller></color></fontfamily></excerpt>=

--Apple-Mail-31-588494679--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec  4 13:31:26 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27137
	for <ieprep-archive@odin.ietf.org>; Wed, 4 Dec 2002 13:31:26 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB4IXnt09681
	for ieprep-archive@odin.ietf.org; Wed, 4 Dec 2002 13:33:49 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4IXnv09678
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 4 Dec 2002 13:33:49 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27115
	for <ieprep-web-archive@ietf.org>; Wed, 4 Dec 2002 13:30:55 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4IX3v09636;
	Wed, 4 Dec 2002 13:33:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4IWHv09560
	for <ieprep@optimus.ietf.org>; Wed, 4 Dec 2002 13:32:17 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA27014
	for <ieprep@ietf.org>; Wed, 4 Dec 2002 13:29:23 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id A18B167104; Wed,  4 Dec 2002 09:57:59 -0500 (EST)
Date: Wed, 4 Dec 2002 13:32:13 -0500
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Rohan Mahy <rohan@cisco.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <3CC03986-07AF-11D7-9AE1-0003938AF740@cisco.com>
Message-Id: <B4EC7759-07B6-11D7-A471-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Wednesday, Dec 4, 2002, at 12:38 America/Montreal, Rohan Mahy wrote:
> A very good choice here is iLBC (Internet Low Bitrate Codec), a 
> royalty-free low-bitrate codec designed for use on the Internet, and 
> therefore very tolerant of loss, jitter, and delay.

Thanks much.  Is there an I-D or RFC describing iLBC ?

>> Now I'm not an application-layer guy very often, so I don't know the 
>> details
>> of SIP or the details of how the voice is being sent (presumably over 
>> RTP/UDP/IP
>> somehow).  However, it is clear that protocol engineering could be 
>> done to add
>> this kind of capability (e.g. codec negotiation) if folks in that 
>> part of the
>> application community cared to do it.
>
> It's already there.

Thanks again.

Which RFC(s) is this documented in ?
And are there special phrases that one should grep for to find the 
right section ?

>> An IAB comment by Sally Floyd at the IEprep meeting in Atlanta bears 
>> repeating.
>> Because the current VoIP stuff does not include congestion avoidance, 
>> the tendency
>> will be for the VoIP traffic to "crowd out" the congestion avoiding 
>> traffic
>> (e.g. anything running over TCP or SCTP).  This is a major problem.  
>> Somehow
>> the community needs to add congestion avoidance into the VoIP 
>> applications
>> protocols.
>
> Actually there is work now to do this. The goal is to run
> RTP and RTCP over DCCP (Datagram Congestion Control Protocol).

Once again thanks.
And same question, is there an I-D or RFC on this work in progress ?

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec  4 14:02:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28571
	for <ieprep-archive@odin.ietf.org>; Wed, 4 Dec 2002 14:02:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB4J4sU11618
	for ieprep-archive@odin.ietf.org; Wed, 4 Dec 2002 14:04:54 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4J4sv11615
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 4 Dec 2002 14:04:54 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28552
	for <ieprep-web-archive@ietf.org>; Wed, 4 Dec 2002 14:02:00 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4J44v11584;
	Wed, 4 Dec 2002 14:04:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4J3Zv11544
	for <ieprep@optimus.ietf.org>; Wed, 4 Dec 2002 14:03:35 -0500
Received: from chiron.nge.isi.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA28517
	for <ieprep@ietf.org>; Wed, 4 Dec 2002 14:00:40 -0500 (EST)
Received: from chiron (csp@localhost)
	by chiron.nge.isi.edu (8.11.6/8.11.6) with ESMTP id gB4J3Np24931;
	Wed, 4 Dec 2002 14:03:23 -0500
Message-Id: <200212041903.gB4J3Np24931@chiron.nge.isi.edu>
To: RJ Atkinson <rja@extremenetworks.com>
cc: Rohan Mahy <rohan@cisco.com>, ieprep <ieprep@ietf.org>
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others 
In-Reply-To: Your message of "Wed, 04 Dec 2002 13:32:13 EST."
             <B4EC7759-07B6-11D7-A471-00039357A82A@extremenetworks.com> 
Date: Wed, 04 Dec 2002 14:03:23 -0500
From: Colin Perkins <csp@isi.edu>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

--> RJ Atkinson writes:
>
>On Wednesday, Dec 4, 2002, at 12:38 America/Montreal, Rohan Mahy wrote:
>> A very good choice here is iLBC (Internet Low Bitrate Codec), a 
>> royalty-free low-bitrate codec designed for use on the Internet, and 
>> therefore very tolerant of loss, jitter, and delay.
>
>Thanks much.  Is there an I-D or RFC describing iLBC ?

draft-ietf-avt-rtp-ilbc-00.txt
draft-ietf-avt-ilbc-codec-00.txt

>>> Now I'm not an application-layer guy very often, so I don't know the 
>>> details
>>> of SIP or the details of how the voice is being sent (presumably over 
>>> RTP/UDP/IP
>>> somehow).  However, it is clear that protocol engineering could be 
>>> done to add
>>> this kind of capability (e.g. codec negotiation) if folks in that 
>>> part of the
>>> application community cared to do it.
>>
>> It's already there.
>
>Thanks again.
>
>Which RFC(s) is this documented in ?
>And are there special phrases that one should grep for to find the 
>right section ?

RFCs 3261 and 3264, amongst others.

>>> An IAB comment by Sally Floyd at the IEprep meeting in Atlanta bears 
>>> repeating.
>>> Because the current VoIP stuff does not include congestion avoidance, 
>>> the tendency
>>> will be for the VoIP traffic to "crowd out" the congestion avoiding 
>>> traffic
>>> (e.g. anything running over TCP or SCTP).  This is a major problem.  
>>> Somehow
>>> the community needs to add congestion avoidance into the VoIP 
>>> applications
>>> protocols.
>>
>> Actually there is work now to do this. The goal is to run
>> RTP and RTCP over DCCP (Datagram Congestion Control Protocol).
>
>Once again thanks.
>And same question, is there an I-D or RFC on this work in progress ?

Very much work in progress. The DCCP working group has a number of drafts,
but I don't think there is an I-D on RTP over DCCP yet. It's not clear, to
me, if the services DCCP provides are suitable for use in VoIP telephony
due to the limited scope for adaptation in common voice codecs (it is
clearly useful for streaming video).

Colin
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec  4 17:48:36 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA08191
	for <ieprep-archive@odin.ietf.org>; Wed, 4 Dec 2002 17:48:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB4Mp3X27397
	for ieprep-archive@odin.ietf.org; Wed, 4 Dec 2002 17:51:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4Mp3v27394
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 4 Dec 2002 17:51:03 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA08154
	for <ieprep-web-archive@ietf.org>; Wed, 4 Dec 2002 17:48:05 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4Mo9v27370;
	Wed, 4 Dec 2002 17:50:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB4MnBv27332
	for <ieprep@optimus.ietf.org>; Wed, 4 Dec 2002 17:49:11 -0500
Received: from sj-msg-core-2.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA08080
	for <ieprep@ietf.org>; Wed, 4 Dec 2002 17:46:13 -0500 (EST)
Received: from imop.cisco.com (IDENT:mirapoint@imop.cisco.com [171.69.11.44])
	by sj-msg-core-2.cisco.com (8.12.2/8.12.2) with ESMTP id gB4Mn20R016425;
	Wed, 4 Dec 2002 14:49:02 -0800 (PST)
Received: from cisco.com (ssh-sjc-1.cisco.com [171.68.225.134])
	by imop.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABK13074;
	Wed, 4 Dec 2002 14:45:56 -0800 (PST)
Date: Wed, 4 Dec 2002 14:49:15 -0800
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: RJ Atkinson <rja@extremenetworks.com>
From: Rohan Mahy <rohan@cisco.com>
In-Reply-To: <B4EC7759-07B6-11D7-A471-00039357A82A@extremenetworks.com>
Message-Id: <9D5F8A5E-07DA-11D7-9AE1-0003938AF740@cisco.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Wednesday, December 4, 2002, at 10:32 AM, RJ Atkinson wrote:

>
> On Wednesday, Dec 4, 2002, at 12:38 America/Montreal, Rohan Mahy wrote:
>> A very good choice here is iLBC (Internet Low Bitrate Codec), a 
>> royalty-free low-bitrate codec designed for use on the Internet, and 
>> therefore very tolerant of loss, jitter, and delay.
>
> Thanks much.  Is there an I-D or RFC describing iLBC ?

draft-ietf-avt-ilbc-codec-00.txt  describes the codec itself
draft-ietf-avt-rtp-ilbc-00.txt describes its use in RTP (the RTP 
payload format)


>>> Now I'm not an application-layer guy very often, so I don't know the 
>>> details
>>> of SIP or the details of how the voice is being sent (presumably 
>>> over RTP/UDP/IP
>>> somehow).  However, it is clear that protocol engineering could be 
>>> done to add
>>> this kind of capability (e.g. codec negotiation) if folks in that 
>>> part of the
>>> application community cared to do it.
>>
>> It's already there.
>
> Thanks again.
>
> Which RFC(s) is this documented in ?

RFC3264 describes how this negotiation works in SIP.  In general, SDP 
(draft-ietf-mmusic-sdp-new-11.txt) describes how you can use its 
semantics to facilitate negotiation

almost the whole of both documents touch negotiation, but sections 5.1, 
6.1, 9, and 10.2 or RFC 3264 are especially relevant (and also pretty 
short).

> And are there special phrases that one should grep for to find the 
> right section ?
>
>>> An IAB comment by Sally Floyd at the IEprep meeting in Atlanta bears 
>>> repeating.
>>> Because the current VoIP stuff does not include congestion 
>>> avoidance, the tendency
>>> will be for the VoIP traffic to "crowd out" the congestion avoiding 
>>> traffic
>>> (e.g. anything running over TCP or SCTP).  This is a major problem.  
>>> Somehow
>>> the community needs to add congestion avoidance into the VoIP 
>>> applications
>>> protocols.
>>
>> Actually there is work now to do this. The goal is to run
>> RTP and RTCP over DCCP (Datagram Congestion Control Protocol).
>
> Once again thanks.
> And same question, is there an I-D or RFC on this work in progress ?

the charter is a good start.  it explicitly calls out the relationship 
with RTP which is nice.
http://www.ietf.org/html.charters/dccp-charter.html

draft-ietf-dccp-problem-00.txt  is self explanatory ;-)
draft-ietf-dccp-spec-00.txt is the proposed base spec (still way early)
draft-ietf-dccp-ccid3-00.txt  is the DCCP profile most suitable for RTP

hope this helps out.
thanks,
-rohan



> Ran
> rja@extremenetworks.com
>

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec  5 07:57:09 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA09752
	for <ieprep-archive@odin.ietf.org>; Thu, 5 Dec 2002 07:57:09 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB5CxUD20102
	for ieprep-archive@odin.ietf.org; Thu, 5 Dec 2002 07:59:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5CxUv20099
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 5 Dec 2002 07:59:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA09743
	for <ieprep-web-archive@ietf.org>; Thu, 5 Dec 2002 07:56:38 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5Ct7v19878;
	Thu, 5 Dec 2002 07:55:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5CsSv19834
	for <ieprep@optimus.ietf.org>; Thu, 5 Dec 2002 07:54:28 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA09618
	for <ieprep@ietf.org>; Thu, 5 Dec 2002 07:51:36 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB5CmHo21435;
	Thu, 5 Dec 2002 04:48:17 -0800 (PST)
Date: Thu, 5 Dec 2002 04:48:17 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: Mpierce1@aol.com
cc: rja@extremenetworks.com, <ieprep@ietf.org>
Subject: Re: [Ieprep] Architectural considerations
In-Reply-To: <20.3a387e9.2b1e6bde@aol.com>
Message-ID: <Pine.GSO.4.44.0212050446190.21190-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Tue, 3 Dec 2002 Mpierce1@aol.com wrote:
    > I'm sure that a lot of the  reliability of the Internet depends on
    > such "rerouting" at the physical layer.

No, it's _packet switched_.  Rerouting is principally done at layer
_three_.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec  5 08:13:48 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10123
	for <ieprep-archive@odin.ietf.org>; Thu, 5 Dec 2002 08:13:48 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB5DG9E21366
	for ieprep-archive@odin.ietf.org; Thu, 5 Dec 2002 08:16:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DG9v21363
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 5 Dec 2002 08:16:09 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10117
	for <ieprep-web-archive@ietf.org>; Thu, 5 Dec 2002 08:13:16 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DCBv21201;
	Thu, 5 Dec 2002 08:12:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5D7Nv20931
	for <ieprep@optimus.ietf.org>; Thu, 5 Dec 2002 08:07:23 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA09916
	for <ieprep@ietf.org>; Thu, 5 Dec 2002 08:04:31 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB5D1Ao21508;
	Thu, 5 Dec 2002 05:01:10 -0800 (PST)
Date: Thu, 5 Dec 2002 05:01:10 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: Mpierce1@aol.com
cc: sob@harvard.edu, <rja@extremenetworks.com>, <ieprep@ietf.org>
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
In-Reply-To: <1ac.ce9e43a.2b1d2c94@aol.com>
Message-ID: <Pine.GSO.4.44.0212050458310.21457-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Mon, 2 Dec 2002 Mpierce1@aol.com wrote:
    > I meant that whatever entity was
    > setting up calls (SIP proxy) would have some knowledge of how many it could
    > attempt to setup simultaneously, at least based on the provisioned capacity.

I assume someone's already explained that the provisioned capacity cannot
be determined?  And that the proxy, even if it were able to
instantaneously observe the degree of congestion on path A-B of an
existing call would have no way of using that to determine in advance the
degree of congestion that a call on path C-B might experience?  And that
all that's moot anyhow, since the degree of conjestion this millisecond
isn't likely to be similar to the degree of conjestion next millisecond
anyway, since THIS IS PACKET SWITCHED?

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec  5 08:19:45 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10273
	for <ieprep-archive@odin.ietf.org>; Thu, 5 Dec 2002 08:19:45 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB5DM6721708
	for ieprep-archive@odin.ietf.org; Thu, 5 Dec 2002 08:22:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DM6v21705
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 5 Dec 2002 08:22:06 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10270
	for <ieprep-web-archive@ietf.org>; Thu, 5 Dec 2002 08:19:14 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DI2v21430;
	Thu, 5 Dec 2002 08:18:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DHTv21397
	for <ieprep@optimus.ietf.org>; Thu, 5 Dec 2002 08:17:29 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10149
	for <ieprep@ietf.org>; Thu, 5 Dec 2002 08:14:36 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB5DBGo21623;
	Thu, 5 Dec 2002 05:11:17 -0800 (PST)
Date: Thu, 5 Dec 2002 05:11:16 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
cc: "'RJ Atkinson'" <rja@extremenetworks.com>, ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149A7@chntex02.is.dyncorp.com>
Message-ID: <Pine.GSO.4.44.0212050502010.21457-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Mon, 2 Dec 2002, Gunn, Janet wrote:
    > Is there a consistent way to identify emergency traffic across several
    > different "entities that are addressing congestion on the access links"?

At what layer?  The lower the layer, the more expensive doing so would be,
and the more difficult it would be to avoid spoofing or misuse.

    > And, is there a way to pass that "identification as emergency traffic" from
    > the access links at the originating node to the access links at the
    > terminating node (if the terminating node is in another domain)? (From what
    > I have read here, and ToS bits based marking would be lost in the
    > intervening ISP.)

Excuse the BUAG.  Topology looks like this:

 Cust A                  |     ISP A    |     ISP B     |          Cust B

 LAN--CPE rtr--tail ckt--acc rtr--core--ix--core--acc rtr--tail--CPE--LAN

The phones and SIP devices are all on those LANs.  The CPE routers aren't
capable of manipulating anything or doing any QoS or anything like that.
The tail circuits are the portion where there's economic incentive to
congest them occasionally.  The portion between the access routers is the
part under the ISPs control, collectively, but it doesn't experience
congestion, and can't afford the high failure rate associated with QoS,
even if QoS had some beneficial effect.

So effectively, what you've got to play with is the stuff going onto the
LANs.  But fortunately, that's where you get your big win, since that's
what does all the call setup, origination, answering, redirection, et
cetera.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec  5 08:31:43 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10592
	for <ieprep-archive@odin.ietf.org>; Thu, 5 Dec 2002 08:31:43 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB5DY5W22311
	for ieprep-archive@odin.ietf.org; Thu, 5 Dec 2002 08:34:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DY5v22308
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 5 Dec 2002 08:34:05 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10583
	for <ieprep-web-archive@ietf.org>; Thu, 5 Dec 2002 08:31:12 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DU2v22062;
	Thu, 5 Dec 2002 08:30:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DTEv22031
	for <ieprep@optimus.ietf.org>; Thu, 5 Dec 2002 08:29:14 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10458
	for <ieprep@ietf.org>; Thu, 5 Dec 2002 08:26:21 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.139])
	by gnat.inet.org (Postfix) with ESMTP id CEE3367104
	for <ieprep@ietf.org>; Thu,  5 Dec 2002 04:55:05 -0500 (EST)
Date: Thu, 5 Dec 2002 08:29:09 -0500
Subject: Re: [Ieprep] identification of packets
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
From: RJ Atkinson <rja@extremenetworks.com>
To: ieprep <ieprep@ietf.org>
Content-Transfer-Encoding: 7bit
In-Reply-To: <Pine.GSO.4.44.0212050502010.21457-100000@paixhost.pch.net>
Message-Id: <88CF72A7-0855-11D7-B233-00039357A82A@extremenetworks.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Thursday, Dec 5, 2002, at 08:11 America/Montreal, Bill Woodcock 
wrote:
> Excuse the BUAG.  Topology looks like this:
>
>  Cust A                  |     ISP A    |     ISP B     |          
> Cust B
>
>  LAN--CPE rtr--tail ckt--acc rtr--core--ix--core--acc 
> rtr--tail--CPE--LAN
>
> The phones and SIP devices are all on those LANs.  The CPE routers 
> aren't
> capable of manipulating anything or doing any QoS or anything like 
> that.

To be fair, Sprint and some other ISPs let some kinds of single-router
class-based queuing happen in CPE routers (and sometimes in a more 
limited way
also in Access Routers).  That queuing could be based on ToS bits of 
packets
from LAN to CPE or based on some combination of (address/protocol/port).
Good news, is that this sort of stuff will work on the link ("tail ckt")
that is most likely to be congested.

> The tail circuits are the portion where there's economic incentive to
> congest them occasionally.  The portion between the access routers
  (i.e. acc-rtr -- core -- ix -- core -- acc-rtr)

> is the part under the ISPs control, collectively, but it doesn't 
> experience
> congestion, and can't afford the high failure rate associated with QoS,
> even if QoS had some beneficial effect.

Right.  That is the "inter-domain QoS" set of issues.

> So effectively, what you've got to play with is the stuff going onto 
> the
> LANs.  But fortunately, that's where you get your big win, since that's
> what does all the call setup, origination, answering, redirection, et
> cetera.

Right.  Downside from a contractual perspective is that one generally 
will
need contracts with each end site if one wants that site's routers to 
have
a particular configuration deployed ==> lots of contracts to manage.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec  5 08:34:35 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10709
	for <ieprep-archive@odin.ietf.org>; Thu, 5 Dec 2002 08:34:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB5DavW22501
	for ieprep-archive@odin.ietf.org; Thu, 5 Dec 2002 08:36:57 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5Dauv22498
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 5 Dec 2002 08:36:56 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10694
	for <ieprep-web-archive@ietf.org>; Thu, 5 Dec 2002 08:34:03 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DX1v22169;
	Thu, 5 Dec 2002 08:33:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DWYv22140
	for <ieprep@optimus.ietf.org>; Thu, 5 Dec 2002 08:32:34 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10475
	for <ieprep@ietf.org>; Thu, 5 Dec 2002 08:29:42 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB5DQLo21733;
	Thu, 5 Dec 2002 05:26:21 -0800 (PST)
Date: Thu, 5 Dec 2002 05:26:21 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
cc: "'Mpierce1@aol.com '" <Mpierce1@aol.com>,
        "'rja@extremenetworks.com '" <rja@extremenetworks.com>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
Subject: RE: [Ieprep] Internet faults, VoIP demo, etc and others
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E779D8@mcl-its-exs01.mail.saic.com>
Message-ID: <Pine.GSO.4.44.0212050512340.21457-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Monday, Dec 2, 2002, Mpierce1@aol.com wrote:
    > So take this simple case and apply an emergency. (I'd like to
    > see the performance last week right after the bombing in Kenya!)

There was no audible or measurable difference.  I don't think that you're
getting that "emergency" means something very different to you than to the
engineering community.  An "emergency" to us would tend to mean something
that has a deleterious impact on the network.  Not something that causes
TV reporters to interrupt advertisements, necessarily.  Those tend to not
generally be associated with any serious impact on our operations.

    > > If 10 or 20 or whatever number of those "several thousands" of end
    > > users tried to make a VoIP call via that single 384K link, what would
    > > have happened?
    > > 1. all calls to be of unacceptable quality

Correct, that's a reasonable assumption.

    > I'll repeat the question which I don't believe I've gotten any
    > concrete answer to: In the case above of the 10 callers from Nairobi, if
    > one of them was authorized to make emergency calls and the other 9
    > weren't...

Authorized by whom?

    > what specific standardized mechanism (not products) could be
    > used to provide the required service across the congested 384 kbit/s
    > satellite link? (The required service is that their call gets set up and
    > they can talk.)

If they were authorized by _me_, that is, if they were paying _me_ to
deliver their call in preference to the other nine callers, I'd
policy-route their SIP traffic into a separate PVC at the CPE router.
That's what I did back in the distant past, when congestion still occurred
occasionally, and it seemed to work quite satisfactorily at the time.

Of course, that assumes xsomeone competent controlling a reasonable CPE
router, neither of which are useful assumptions.  It also assumes that the
person performing the authorization is someone with influence over network
operations, which is also not a useful assumption.

      On Mon, 2 Dec 2002, King, Kimberly  S. wrote:
    > But, under a different set of assumptions, the answer is different.

Dr. King's examples of two different contexts under which there might or
might not be useful ways of controlling access are very good ones.  Again,
it's important to note, as she points out, that this function is quite
feasible at the application layer, in the SIP end-nodes.  The proxies
really don't have anything to do with it.  It's the phones themselves, the
end-nodes that control the amount of data in and out of each one.  And
they're the only devices that have any visibility whatsoever into the
degree of congestion that they may be experiencing, end-to-end.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec  5 08:40:46 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10841
	for <ieprep-archive@odin.ietf.org>; Thu, 5 Dec 2002 08:40:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB5Dh7L23360
	for ieprep-archive@odin.ietf.org; Thu, 5 Dec 2002 08:43:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5Dh7v23357
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 5 Dec 2002 08:43:07 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10835
	for <ieprep-web-archive@ietf.org>; Thu, 5 Dec 2002 08:40:15 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5Dd4v23175;
	Thu, 5 Dec 2002 08:39:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5Dcfv23152
	for <ieprep@optimus.ietf.org>; Thu, 5 Dec 2002 08:38:41 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10739
	for <ieprep@ietf.org>; Thu, 5 Dec 2002 08:35:49 -0500 (EST)
Received: from mcl-its-ieg02.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Thu, 5 Dec 2002 08:38:04 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg02.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120508375227625
 ; Thu, 05 Dec 2002 08:37:52 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YG2Y4DPL>; Thu, 5 Dec 2002 08:38:08 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E779FD@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Bill Woodcock '" <woody@pch.net>,
        "'Gunn, Janet '" <Janet.Gunn@dyncorp.com>
Cc: "''RJ Atkinson' '" <rja@extremenetworks.com>,
        "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
Date: Thu, 5 Dec 2002 08:38:19 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Bill said:
>The CPE routers aren't capable of manipulating anything or doing any QoS 
>or anything like that.

I'm not following you here.  Are you saying you don't believe a CPE router
can make a decision about how to queue packets moving from the LAN to the
tail circuit?  I know a lot depends upon the definition of "manipulating
anything" or "doing any QoS" but can't some such routers set DSCP based upon
a variety of information such as protocol, etc.  Can't such routers support
a low latency queue and perhaps other queues served in round-robin fashion
with a scheduler that tells what goes when or something of that form?  Of
course, this doesn't come standard -- one has to set it up this way -- but
I'm not following what you mean by CPE router not being able to manipulate
or do any QoS, etc.  Am I missing something?

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec  5 08:42:47 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10945
	for <ieprep-archive@odin.ietf.org>; Thu, 5 Dec 2002 08:42:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB5Dj9R23527
	for ieprep-archive@odin.ietf.org; Thu, 5 Dec 2002 08:45:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5Dj8v23524
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 5 Dec 2002 08:45:08 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10919
	for <ieprep-web-archive@ietf.org>; Thu, 5 Dec 2002 08:42:16 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5Df1v23277;
	Thu, 5 Dec 2002 08:41:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DeFv23239
	for <ieprep@optimus.ietf.org>; Thu, 5 Dec 2002 08:40:15 -0500
Received: from cs.columbia.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10753
	for <ieprep@ietf.org>; Thu, 5 Dec 2002 08:37:22 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB5DeAbb012570
	(version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NOT);
	Thu, 5 Dec 2002 08:40:11 -0500 (EST)
Received: from cs.columbia.edu (tallgrass.netlab.uky.edu [204.198.76.66])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gB5De9dG023193
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Thu, 5 Dec 2002 08:40:10 -0500 (EST)
Message-ID: <3DEF56AD.7070204@cs.columbia.edu>
Date: Thu, 05 Dec 2002 08:37:49 -0500
From: Henning Schulzrinne <hgs@cs.columbia.edu>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.2.1) Gecko/20021130
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: RJ Atkinson <rja@extremenetworks.com>
CC: ieprep <ieprep@ietf.org>
Subject: Re: [Ieprep] identification of packets
References: <88CF72A7-0855-11D7-B233-00039357A82A@extremenetworks.com>
In-Reply-To: <88CF72A7-0855-11D7-B233-00039357A82A@extremenetworks.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

> To be fair, Sprint and some other ISPs let some kinds of single-router
> class-based queuing happen in CPE routers (and sometimes in a more 
> limited way
> also in Access Routers).  That queuing could be based on ToS bits of 
> packets
> from LAN to CPE or based on some combination of (address/protocol/port).

Unfortunately, this COS arrangement is highly susceptible to 
denial-of-service attacks. For voice communications, there is no fixed 
source or destination address (even assuming you could do reliable 
address filtering), and in all likelihood, no fixed port. Thus, you'd 
have to assign this by TOS byte or protocol (say, UDP) only. Highly 
spoofable.

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec  5 08:51:59 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA11259
	for <ieprep-archive@odin.ietf.org>; Thu, 5 Dec 2002 08:51:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB5DsLa24136
	for ieprep-archive@odin.ietf.org; Thu, 5 Dec 2002 08:54:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DsLv24133
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 5 Dec 2002 08:54:21 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA11238
	for <ieprep-web-archive@ietf.org>; Thu, 5 Dec 2002 08:51:28 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5Do2v23762;
	Thu, 5 Dec 2002 08:50:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5DmJv23660
	for <ieprep@optimus.ietf.org>; Thu, 5 Dec 2002 08:48:19 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA11047
	for <ieprep@ietf.org>; Thu, 5 Dec 2002 08:45:26 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB5Dfto21864;
	Thu, 5 Dec 2002 05:41:55 -0800 (PST)
Date: Thu, 5 Dec 2002 05:41:55 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
cc: "'Gunn, Janet '" <Janet.Gunn@dyncorp.com>,
        "''RJ Atkinson' '" <rja@extremenetworks.com>,
        "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E779FD@mcl-its-exs01.mail.saic.com>
Message-ID: <Pine.GSO.4.44.0212050534190.21457-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Thu, 5 Dec 2002, King, Kimberly  S. wrote:
    > >The CPE routers aren't capable of manipulating anything or doing any QoS
    > >or anything like that.
    > I'm not following you here.  Are you saying you don't believe a CPE router
    > can make a decision about how to queue packets moving from the LAN to the
    > tail circuit?

I'm making a generalization...  While it's quite possible to put a capable
router in the CPE role, and to have a capable person responsible for
configuring it and maintaining it, that combination is so rare as to be
negligible, because it would be three orders of magnitude more expensive,
with too small a return on that investment to be worthwhile for the vast
majority of customers.

CPE routers used to be a lot more similar to access and core routers than
they are today.  At that time, they ran similar operating systems, and had
similar, if slower, processors, and were configured similarly, if more
simply.  And the amortization period on CPE is typically a lot longer than
on infrastructure, so some of those still exist out there.  However,
they're typically on older lines, ADNs and frac T1s and so forth, and
generally going out of service over time.

Newer CPE routers don't have these capabilities, and don't have fast
enough processors, or operating systems with legacy QoS support, and
they're not designed to be configured by operators, anyway.  Everything
gets more efficient over time, so excess feature support goes away, and
new vendors don't bother to support old stuff like IPX or X.25 or QoS.
They need to keep prices down, so they won't go out of business so
quickly.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec  5 09:01:30 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA11652
	for <ieprep-archive@odin.ietf.org>; Thu, 5 Dec 2002 09:01:30 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB5E3qV24502
	for ieprep-archive@odin.ietf.org; Thu, 5 Dec 2002 09:03:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5E3qv24499
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 5 Dec 2002 09:03:52 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA11638
	for <ieprep-web-archive@ietf.org>; Thu, 5 Dec 2002 09:00:59 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5Dv3v24248;
	Thu, 5 Dec 2002 08:57:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5Du2v24199
	for <ieprep@optimus.ietf.org>; Thu, 5 Dec 2002 08:56:02 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA11291
	for <ieprep@ietf.org>; Thu, 5 Dec 2002 08:53:09 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB5Dngo21911;
	Thu, 5 Dec 2002 05:49:42 -0800 (PST)
Date: Thu, 5 Dec 2002 05:49:42 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
cc: "'RJ Atkinson'" <rja@extremenetworks.com>, <Mpierce1@aol.com>,
        ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: Looking for governmental feedback as well
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149A0@chntex02.is.dyncorp.com>
Message-ID: <Pine.GSO.4.44.0212050546540.21457-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Mon, 2 Dec 2002, Gunn, Janet wrote:
    > I don't think anyone is asking the ISPs to solve the problem of congestion
    > on access links (though it seems that some ISPs (e.g., Sprint) are
    > addressing it).
    > What is being asked is for IETF to help identify emergency traffic WHEN SOME
    > ENTITY is addressing congestion on the access links.

Okay, so, bearing in mind that engineering and deployment time are
precious, and nobody has capex, when might that be?

I'm not being facetious...  I'm just going back to the needs-assesment
issue.  If there's no one for whom this is a need, that is, if there's no
one who's willing to buy this product if it's made available, then all the
development work that goes into it is wasted, and could better have been
spent addressing the problem in some other, more likely to be adopted,
way.

Who would ever act against their own self-interest to reduce congestion on
their access link?


                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec  5 10:28:48 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA16191
	for <ieprep-archive@odin.ietf.org>; Thu, 5 Dec 2002 10:28:48 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB5FVBB31645
	for ieprep-archive@odin.ietf.org; Thu, 5 Dec 2002 10:31:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5FVBv31642
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 5 Dec 2002 10:31:11 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA16147
	for <ieprep-web-archive@ietf.org>; Thu, 5 Dec 2002 10:28:16 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5FR5v31484;
	Thu, 5 Dec 2002 10:27:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5FQsv31451
	for <ieprep@optimus.ietf.org>; Thu, 5 Dec 2002 10:26:54 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA15988
	for <ieprep@ietf.org>; Thu, 5 Dec 2002 10:23:59 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAGVP5>; Thu, 5 Dec 2002 10:24:24 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149BC@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'Bill Woodcock'" <woody@pch.net>
Cc: "'RJ Atkinson'" <rja@extremenetworks.com>, ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
Date: Thu, 5 Dec 2002 10:24:15 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

The higher the layer the better, I think.

Hypothetical example (sticking to VoIP as the example, but I don't want to
restrict the discussion to JUST VoIP):

Caller is at a site A, served by SPRINT, with their congestion control
mechanism implemented on the access link.

Celled party is at site B served by Genuity, with their congestion control
mechanism implemented on the access link.

Assume that both site A and Site B have configured their access lines so
that there is  category of traffic which is called (by people, not
necessarily machines) "emergency"?

Assume, for the time being, that the sites are owned by "FEMA" and the "Red
Cross", so that you know in advance that there is likely to be emergency
traffic between them during an "incident", and any business arrangements are
worked out in advance. (I am also interested in the "opportunistic" case
where there is not a business arrangement in advance (e.g., Internet Cafe),
but that seems much harder, and much more open to DoS abuse.)

Then the question is: "Is there a way for site A to mark "the traffic" (at
whatever layer is most likely to work) so that it recognized as "emergency
traffic" on both access links?  And for the traffic to be given the Sprint
"emergency treatment" at site A's access link and Genuity's "emergency
treatment" at site B, which might actually be different treatments?"

If it is possible, is it practical, and what are the drawbacks, and to whom?

I am asking nothing from the ISP core, except to pass the "marking" (so ToS
bits won't work).  


-----Original Message-----
From: Bill Woodcock [mailto:woody@pch.net]
Sent: Thursday, December 05, 2002 8:11 AM
To: Gunn, Janet
Cc: 'RJ Atkinson'; ieprep
Subject: RE: [Ieprep] identification of packets


      On Mon, 2 Dec 2002, Gunn, Janet wrote:
    > Is there a consistent way to identify emergency traffic across several
    > different "entities that are addressing congestion on the access
links"?

At what layer?  The lower the layer, the more expensive doing so would be,
and the more difficult it would be to avoid spoofing or misuse.

    > And, is there a way to pass that "identification as emergency traffic"
from
    > the access links at the originating node to the access links at the
    > terminating node (if the terminating node is in another domain)? (From
what
    > I have read here, and ToS bits based marking would be lost in the
    > intervening ISP.)

Excuse the BUAG.  Topology looks like this:

 Cust A                  |     ISP A    |     ISP B     |          Cust B

 LAN--CPE rtr--tail ckt--acc rtr--core--ix--core--acc rtr--tail--CPE--LAN

The phones and SIP devices are all on those LANs.  The CPE routers aren't
capable of manipulating anything or doing any QoS or anything like that.
The tail circuits are the portion where there's economic incentive to
congest them occasionally.  The portion between the access routers is the
part under the ISPs control, collectively, but it doesn't experience
congestion, and can't afford the high failure rate associated with QoS,
even if QoS had some beneficial effect.

So effectively, what you've got to play with is the stuff going onto the
LANs.  But fortunately, that's where you get your big win, since that's
what does all the call setup, origination, answering, redirection, et
cetera.

                                -Bill

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec  5 13:20:47 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA22304
	for <ieprep-archive@odin.ietf.org>; Thu, 5 Dec 2002 13:20:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB5INAH12165
	for ieprep-archive@odin.ietf.org; Thu, 5 Dec 2002 13:23:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5INAv12162
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 5 Dec 2002 13:23:10 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA22298
	for <ieprep-web-archive@ietf.org>; Thu, 5 Dec 2002 13:20:16 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5IHlv11706;
	Thu, 5 Dec 2002 13:17:47 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB5IGav11678
	for <ieprep@optimus.ietf.org>; Thu, 5 Dec 2002 13:16:36 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA21959
	for <ieprep@ietf.org>; Thu, 5 Dec 2002 13:13:40 -0500 (EST)
Received: from mcl-its-ieg02.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Thu, 5 Dec 2002 13:15:58 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg02.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120513154727866
 ; Thu, 05 Dec 2002 13:15:47 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YG2Y49CQ>; Thu, 5 Dec 2002 13:16:03 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A01@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Gunn, Janet '" <Janet.Gunn@dyncorp.com>,
        "''Bill Woodcock' '" <woody@pch.net>
Cc: "''RJ Atkinson' '" <rja@extremenetworks.com>,
        "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
Date: Thu, 5 Dec 2002 13:16:14 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Janet asked:
>Is there a way for site A to mark "the traffic" (at
>whatever layer is most likely to work) so that it recognized as
>"emergency traffic" on both access links?  And for the traffic to 
>be given the Sprint "emergency treatment" at site A's access link and 
>Genuity's "emergency treatment" at site B, which might actually be 
>different treatments?"

>If it is possible, is it practical, and what are the drawbacks, and to
>whom?

>I am asking nothing from the ISP core, except to pass the "marking" 
>(so ToS bits won't work).  

Janet,

Thank you for asking such good questions.  I take it by your last statement
(so ToS bits won't work) that you recognize that you can't expect ToS bits
to remain unmodified as they pass over the Internet.  For example, many ISPs
will zero these bits out.

Note that you are asking for a form of multiple ISP QoS that has not been
deployed to date for some pretty serious business and technical reasons (see
http://www1.ietf.org/mail-archive/working-groups/ieprep/current/msg00832.htm
l).  

The only thing I can think of at this time that might be possible if you
could get the providers to agree to it would be to provide some kind of
secure tunnel between them.  The ISPs would continue to zero out ToS on all
incoming packets but it wouldn't matter because the emergency ones would be
marked accordingly within the tunnel.  

Here I'm assuming a secure tunnel between ISP X and ISP Y.  There would have
to be some way to figure out how to put emergency traffic and only emergency
traffic into that tunnel.  Traffic can be remarked according to local use
perhaps requiring ISP X and Y to share information.  They would have to
trust each other (gulp).  I suppose they could police it to somewhat limit
DoS.  

Drawbacks include complexity adds security risks (plus all the ones Ran
listed associated with complexity in an earlier email), the design isn't
scalable, nicely identifying and aggregating "emergency traffic" will
perhaps make it an easier target for attackers, the cost could well exceed
the price of simply buying bigger access pipes on these locations that are
already recognized to carry emergency traffic, and I'm sure there are lots
more.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 04:08:04 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA29599
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 04:08:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB69AZ910830
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 04:10:35 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB69AZv10827
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 04:10:35 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA29580
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 04:07:33 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB699Fv10797;
	Fri, 6 Dec 2002 04:09:15 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB698cv10768
	for <ieprep@optimus.ietf.org>; Fri, 6 Dec 2002 04:08:38 -0500
Received: from devil.jaist.ac.jp (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id EAA29490
	for <ieprep@ietf.org>; Fri, 6 Dec 2002 04:05:35 -0500 (EST)
Received: (qmail 14793 invoked from network); 6 Dec 2002 09:07:30 -0000
Received: from localhost (127.0.0.1)
  by localhost with SMTP; 6 Dec 2002 09:07:30 -0000
Date: Fri, 06 Dec 2002 18:07:29 +0900 (JST)
Message-Id: <20021206.180729.861023506.izawa@izawa.org>
To: ieprep@ietf.org
From: Yukimitsu Izawa <izawa@izawa.org>
X-fingerprint20: 2BB0 B30E E740 EFC6 DE51  8817 6EC4 EAD0 E818 98F2
X-URL: http://izawa.org/
X-Mailer: Mew version 3.0.66 on XEmacs 21.1.14 (Cuyahoga Valley)
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] Short description of IAA.
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Hello, My name is Yukimitsu Izawa and I am a researcher in IAA which
is research project in WIDE.
I heard from other people, IAA becomes a topic of conversation in
IEPREP WG Mailing list.

Then, I explain about IAA with sort description.
If you have more interest in IAA, please read following papers.

--- Purpose of IAA:

The IAA system is designed to one of disaster-support information
database for large-scale disaster like Hanshin-Awaji Earthquake[1].
System requirements and key technologies are as follows:

* The IAA system must be scalable and flexible. Because we cannot
  predict the size of a disaster in advance, capacity of the system must
  be easily extendible. To consider scalability and flexibility of
  the system, the IAA system consists of IAA cluster nodes and
  IAA transport mechanism.

* The IAA system must be robust and fault tolerant. It means that
  the IAA system must consist multiple components which provide the same
  functionality and they are distributed geographically.
  Even if any part of the system crashes or network congested,
  the total system should continue service without interruption.
  And when crashed parts recover, the system should start to work with
  all of them. To keep robustness and fault tolerant and avoid network
  congestion, all IAA clusters exchange each registered victim information
  by IAA transport mechanism and keeps identical victim information among them.
  Queries are handled by any IAA cluster nodes and they are navigated
  by load distribution mechanism such as weighted round robin.

* The IAA system must protect personal privacy. Personal safety
  information is strongly related to privacy information, and the system
  must protect queries which violate the privacy information.
  Security of the privacy-related information transmitted through the Internet
  must also be considered.
  To consider privacy and security issues even when disaster occurs,
  IAA cluster handles victim information as an encrypted message in the
  IAA transport mechanism. And to avoid unexpected exploitation of privacy
  information, queries are required to fix the individual victim information. 

* The IAA system must provide accurate information. Personal safety
  information from oneself and indirect information from others should
  be discriminated.
  To provide data accuracy of registered victim information, two types of
  registration format are prepared. One registration format is that for
  oneself and the other is that for other persons situation. IAA cluster
  treats the former information as high accurate data and people who get
  victims' information can easily judge the accuracy of the results.

* The IAA system must provide flexible user interfaces for supporting
  many kinds of people. People may not be familiar with computers or
  there are not enough facilities to register their information in the
  disaster area.


[1] N. Tada, Y. Izawa, M. Kimoto, T. Maruyama, H. Ohno and M. Nakayama,
   "IAA System (``I Am Alive''): The Experiences of the Internet Disaster
    Drills", Proc. of INET2000, ISOC, Jul. 2000.



-- Technology used in IAA system:

Current prototype implementation of the IAA cluster is an user area
application on FreeBSD platform. And IAA cluster uses many standard
software packages to eliminate system dependency.
For example, database component uses PostgreSQL package, web server
portion uses Apache program and IAA transport mechanism uses INN package.
In addition to these packages, Perl scripts are composed to handle
operations of IAA system requirements. User interface programs are
developed separately.
We developed Web client, FAX interface, telephone interface,
windows client and i-mode mobile phone interface for IAA cluster.
Other detailed implementation of transport mechanism and FAX interface
program are  shown in [2,3].


[2] Y. Izawa, S. Ishii, N. Tada and M. Nakayama,
    "Implementation and Evaluation of Widely Distributed Database System
    Using Satellite based Multicast and NetNews System for the Transport
    Mechanism", Proc. of IWS98, IEICE, Mar. 1998.

[3] H. Honma, A. Noda and H. Ohno,
    "An Alternative User Interface for the IAA System: Using OCR/OMR as
    on-ramp Gateway for the Internet", Proc. of IWS98, IEICE, Mar. 1998.


--- L3 requirement for IAA system:

To synchronize victims' information between IAA cluster nodes, TCP/IP
connectivity is necessary. In our experimental system, we have redundant
information exchange paths between IAA cluster nodes.
We don't use any priority based L3 services between IAA cluster nodes.

---- IAA-Alliance
 
There are several implementations of IAA system.  One is developed by
WIDE Project, and one is developed under direction of CRL
(Communications Research Laboratory), other implementations also
exist. And there are many systems which has similar facilities to IAA.
At this point, technical problems are protocol interconnectivity and
meta-data portability. To establish interconnectivity of these systems,
IAA Alliance has organized in 2002.  IAA Alliance may deal with
important roles for establishing a standard with other international
standard bodies.


Thanks for reading this mail through.

-- izawa
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 08:16:11 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA04911
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 08:16:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB6DIXE23215
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 08:18:33 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6DIXv23212
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 08:18:33 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA04870
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 08:15:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6DH9v23147;
	Fri, 6 Dec 2002 08:17:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3MnSv23107
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 17:49:28 -0500
Received: from chiron.nge.isi.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA23236
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 17:46:32 -0500 (EST)
Received: from chiron (csp@localhost)
	by chiron.nge.isi.edu (8.11.6/8.11.6) with ESMTP id gB3MnLe22085;
	Tue, 3 Dec 2002 17:49:21 -0500
Message-Id: <200212032249.gB3MnLe22085@chiron.nge.isi.edu>
To: Mpierce1@aol.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others 
In-Reply-To: Your message of "Tue, 03 Dec 2002 17:37:38 EST."
             <14b.1849b180.2b1e8c32@aol.com> 
Date: Tue, 03 Dec 2002 17:49:21 -0500
From: Colin Perkins <csp@isi.edu>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

--> Mpierce1@aol.com writes:
>In a message dated 12/3/2002 3:52:30 PM Eastern Standard Time, csp@isi.edu 
>writes:
>
>> If you implement RTP correctly, the RTP control protocol provides feedback
>> on the packet loss, latency and jitter at the other end [RFC 1889].
>> 
>
>But:
>
>In a message dated 12/2/2002 1:42:11 PM Eastern Standard Time, 
>sob@harvard.edu writes:
>
>
>> > 2. new call attempts would have been rejected
>> 
>> no- the congested link would not cause the SIP proxies to reject calls
>> n
>
>
>[MAP] Guess I'm missing something, since the RTP functionality mentioned by 
>Colin is the type of thing that I thought would allow a call control 
>application to know that it should stop trying to set up more calls. Scott 
>(and others) say there is no way to do it.

The media doesn't necessarily follow the same path as the signalling, and
so the SIP proxy may not see the RTP reception quality reports.

Colin
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 08:16:16 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA04926
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 08:16:16 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB6DIce23250
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 08:18:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6DIcv23247
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 08:18:38 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA04878
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 08:15:44 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6DH4v23114;
	Fri, 6 Dec 2002 08:17:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3Knsv14043
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 15:49:54 -0500
Received: from chiron.nge.isi.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA17299
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 15:46:58 -0500 (EST)
Received: from chiron (csp@localhost)
	by chiron.nge.isi.edu (8.11.6/8.11.6) with ESMTP id gB3KnlN21049;
	Tue, 3 Dec 2002 15:49:47 -0500
Message-Id: <200212032049.gB3KnlN21049@chiron.nge.isi.edu>
To: Mpierce1@aol.com
cc: rja@extremenetworks.com, ieprep@ietf.org
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others 
In-Reply-To: Your message of "Tue, 03 Dec 2002 15:19:40 EST."
             <1a8.cf7fd2e.2b1e6bdc@aol.com> 
Date: Tue, 03 Dec 2002 15:49:47 -0500
From: Colin Perkins <csp@isi.edu>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

--> Mpierce1@aol.com writes:
>
>--part1_1a8.cf7fd2e.2b1e6bdc_boundary
>Content-Type: text/plain; charset="US-ASCII"
>Content-Transfer-Encoding: 7bit
>
>In a message dated 12/3/2002 1:21:13 PM Eastern Standard Time, 
>rja@extremenetworks.com writes:
>
>
>> On Tuesday, Dec 3, 2002, at 10:28 America/Montreal, Mpierce1@aol.com 
>> wrote:
>> > [MAP] So this says that congestion can occur and there is no way for 
>> > the application (SIP) to deal with it, like by restricting new call 
>> > setups, so the IP-layer is left completely on its own to deal with it, 
>> > and the only way it can is by discarding packets.
>> 
>> Hmm.  Almost, but not quite right.  The application end-points, that is the
>> application software on either end of the actual VoIP session, *can* and 
>> *do*
>> know that they are experiencing X% packet loss or that jitter is increasing
>> from W to Z, or that latency has increased from A to B, or any combination
>> of the above.  Having that knowledge, which is in the VoIP endpoints not in 
>> the
>> SIP server, the endpoints could react in any of several ways.  An example
>> that seems most obvious to me would be to switch from G.711 to a more robust
>> CODEC that consumes less bandwidth (e.g. some form of CELP).
>
>[MAP]
>This is an interesting one. There's been a lot of talk about this idea in 
>other places too. But let's examine it in reality.
>
>If my application is handling my phone call, first, I don't know what 
>functionality tells it about packet loss or latency or jitter at the other 
>end, unless we add specific measuring protocols (which I thing are being 
>worked on). If my application had a way to know about excess packet loss at 
>the other end (presumably caused by there being too many other people trying 
>to talk over the same link), what could it do about it? It could switch to a 
>lower speed codec, thereby giving everyone a slighty lower packet loss ratio 
>(including myself), but I would surely get worse quality from the lower speed 
>codec. (If it were a better codec, I would have selected it in the first 
>place.)

If you implement RTP correctly, the RTP control protocol provides feedback
on the packet loss, latency and jitter at the other end [RFC 1889].

>Or my application could switch to a higher bandwidth utilitzation, like 
>adding more forward error correction to make up for the lost packets or to a 
>higher rate codec that is less suseptable to packet loss, thereby giving 
>everyone a slightly higher packet loss ratio (including myself), but I 
>probably get better quality (at least as far as packet loss is concerned).
>
>So which would you choose? Right! And everyone else would do the same (unless 
>you're charged according to bandwidth) and the congestion and packet loss 
>goes up and the net gain is negative.

The hope is that we can apply clue to VoIP implementations, although that
may not be realistic without policing. Certainly the RFCs that describe FEC
use with RTP talk about these issues.

Colin
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 08:52:01 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA05999
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 08:52:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB6DsNx24987
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 08:54:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6DsNv24984
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 08:54:23 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA05958
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 08:51:29 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6Dr3v24959;
	Fri, 6 Dec 2002 08:53:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6Dqtv24941
	for <ieprep@optimus.ietf.org>; Fri, 6 Dec 2002 08:52:55 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA05928
	for <ieprep@ietf.org>; Fri, 6 Dec 2002 08:50:01 -0500 (EST)
Received: from extremenetworks.com (unknown [10.112.40.61])
	by gnat.inet.org (Postfix) with ESMTP
	id A495167104; Fri,  6 Dec 2002 05:18:57 -0500 (EST)
Date: Fri, 6 Dec 2002 08:52:50 -0500
Subject: Re: [Ieprep] identification of packets
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep '" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E779FD@mcl-its-exs01.mail.saic.com>
Message-Id: <025AFE2D-0922-11D7-B135-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Thursday, Dec 5, 2002, at 08:38 America/Montreal, King, Kimberly S. 
wrote:
>  Can't such routers support
> a low latency queue and perhaps other queues served in round-robin 
> fashion
> with a scheduler that tells what goes when or something of that form?

Many COTS routers can do that today.  And since it is not intrinsically
an "over the wire" feature, that sort of class-based-queuing isn't 
something
that IETF would standardise.  The most common scheduling algorithms are
WFQ, RED, and hybrids/variations on those two.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 08:57:28 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA06261
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 08:57:28 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB6Dxpi25249
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 08:59:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6Dxpv25246
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 08:59:51 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA06218
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 08:56:57 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6DwNv25177;
	Fri, 6 Dec 2002 08:58:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6Dvgv25136
	for <ieprep@optimus.ietf.org>; Fri, 6 Dec 2002 08:57:42 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA06143
	for <ieprep@ietf.org>; Fri, 6 Dec 2002 08:54:48 -0500 (EST)
Received: from extremenetworks.com (unknown [10.112.40.61])
	by gnat.inet.org (Postfix) with ESMTP
	id 81D2867103; Fri,  6 Dec 2002 05:23:44 -0500 (EST)
Date: Fri, 6 Dec 2002 08:57:37 -0500
Subject: Re: [Ieprep] identification of packets
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Henning Schulzrinne <hgs@cs.columbia.edu>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <3DEF56AD.7070204@cs.columbia.edu>
Message-Id: <AD49B807-0922-11D7-B135-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Thursday, Dec 5, 2002, at 08:37 America/Montreal, Henning 
Schulzrinne wrote:

>> To be fair, Sprint and some other ISPs let some kinds of single-router
>> class-based queuing happen in CPE routers (and sometimes in a more 
>> limited way
>> also in Access Routers).  That queuing could be based on ToS bits of 
>> packets
>> from LAN to CPE or based on some combination of 
>> (address/protocol/port).
>
> Unfortunately, this COS arrangement is highly susceptible to 
> denial-of-service attacks.

All QoS mechanisms within the IP layer are highly susceptible to 
denial-of-service
attacks.  So I have no argument on your claim above.   Within a single 
administrative
domain, there is much more potential to manage/reduce risk from such 
attacks than
exists in the inter-domain case.

> For voice communications, there is no fixed source or destination 
> address (even assuming you could do reliable address filtering), and 
> in all likelihood, no fixed port. Thus, you'd have to assign this by 
> TOS byte or protocol (say, UDP) only. Highly spoofable.

One might know that a particular IP destination or source address 
within a given
administrative domain is the gateway to the "emergency call centre", the
"crisis management facility", "the 911 centre", "the NOC subnet", the
"VoIP subnet", or other such like.

So how much filtering and such one can do will directly vary with how 
thoughtful
a given domain has organised its networks, servers, and VoIP phones.  
Within
a single administrative domain, one can set policies throughout that 
domain,
so to the extent the networks/servers/IP-phones have been thoughtfully 
done,
one can deploy a corresponding amount of protective filtering.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 09:03:02 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA06480
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 09:03:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB6E5PI25510
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 09:05:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6E5Pv25507
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 09:05:25 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA06451
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 09:02:31 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6E41v25419;
	Fri, 6 Dec 2002 09:04:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6E3Ev25377
	for <ieprep@optimus.ietf.org>; Fri, 6 Dec 2002 09:03:14 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA06349
	for <ieprep@ietf.org>; Fri, 6 Dec 2002 09:00:20 -0500 (EST)
Received: from extremenetworks.com (unknown [10.112.40.61])
	by gnat.inet.org (Postfix) with ESMTP
	id B8E7267103; Fri,  6 Dec 2002 05:29:16 -0500 (EST)
Date: Fri, 6 Dec 2002 09:03:09 -0500
Subject: Re: [Ieprep] identification of packets
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep '" <ieprep@ietf.org>
To: Bill Woodcock <woody@pch.net>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <Pine.GSO.4.44.0212050534190.21457-100000@paixhost.pch.net>
Message-Id: <7357265A-0923-11D7-B135-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Thursday, Dec 5, 2002, at 08:41 America/Montreal, Bill Woodcock 
wrote:
> Newer CPE routers don't have these capabilities, and don't have fast
> enough processors, or operating systems with legacy QoS support, and
> they're not designed to be configured by operators, anyway.

I'm not sure I really believe this is as common today as the above 
suggests.
At least several vendors still make/sell CPE routers that have those
capabilities.  (I don't want to get into vendor-specific conversations
on list, but will happily chat in more detail in private email.)

> Everything gets more efficient over time, so excess feature support 
> goes away,
> and new vendors don't bother to support old stuff like IPX or X.25 or 
> QoS.
> They need to keep prices down, so they won't go out of business so
> quickly.

Some vendors find that the best way to reduce costs is to build custom
ASICs or use commodity ASICs.  In either case, implementing QoS queuing
and filtering in a custom ASIC is straight-forward if the vendor thinks
about it.  At least some commodity ASICs have the necessary support 
already.
And at least one vendor chose not to rip out IPX support in later 
chipsets
simply because cost of removal exceeded cost of leaving it in place.

Now this is not to say that "all vendors" have any particular feature 
set
that one might need, just that some/many likely do have it and that the 
ability
to buy economical products with those features does not appear to be 
disappearing
for now (as near as I can tell).

Ran
rja@extremenetworks.com


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 09:10:03 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA04924
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 08:16:15 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB6DIc523235
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 08:18:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6DIcv23228
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 08:18:38 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA04875
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 08:15:43 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6DH8v23131;
	Fri, 6 Dec 2002 08:17:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB3MUAv21570
	for <ieprep@optimus.ietf.org>; Tue, 3 Dec 2002 17:30:10 -0500
Received: from chiron.nge.isi.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA22571
	for <ieprep@ietf.org>; Tue, 3 Dec 2002 17:27:15 -0500 (EST)
Received: from chiron (csp@localhost)
	by chiron.nge.isi.edu (8.11.6/8.11.6) with ESMTP id gB3MU2a21780;
	Tue, 3 Dec 2002 17:30:03 -0500
Message-Id: <200212032230.gB3MU2a21780@chiron.nge.isi.edu>
To: RJ Atkinson <rja@extremenetworks.com>
cc: ieprep <ieprep@ietf.org>
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others 
In-Reply-To: Your message of "Tue, 03 Dec 2002 16:57:37 EST."
             <3C9B612C-070A-11D7-A694-00039357A82A@extremenetworks.com> 
Date: Tue, 03 Dec 2002 17:30:02 -0500
From: Colin Perkins <csp@isi.edu>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

--> RJ Atkinson writes:
>On Tuesday, Dec 3, 2002, at 15:49 America/Montreal, Colin Perkins wrote:
>> The hope is that we can apply clue to VoIP implementations,
>
>Colin,
>
>I share that hope. :-)
>
>> although that may not be realistic without policing. Certainly the RFCs
>> that describe FEC use with RTP talk about these issues.
>
>For folks like me who aren't up on RTP, could you please list some
>assigned-reading by listing the most relevant RFCs on this thread
>(including the current incarnation of the base RTP spec) ??

There are three parts to the RTP framework:
 - The base RTP spec 
 - A set of profiles
 - A set of payload formats

The base RTP spec is RFC 1889, updated by draft-ietf-avt-rtp-new-11.txt
(the update has been approved for publication, and is currently winding
its way from IESG to RFC Editor). The base spec defines a data transfer
protocol (framing, timing, sequencing, payload identification, etc) and 
a control protocol (reception quality feedback, user identification and 
lip synchronisation). 

The profiles adapt RTP to particular environments. For voice-over-IP or
video teleconferencing, the appropriate profile is RFC 1890 (updated by
draft-ietf-avt-profile-new-12.txt, also on its way to RFC publication 
as a draft standard).

Payload formats define the mapping from particular codecs onto RTP. The
profile defines some payload formats (e.g. for G.711), others are defined
in separate RFCs (e.g. H.263 in RFC 2429, G.722.1 in RFC 3047). FEC is
either implemented as part of a payload format for a particular codec (e.g.
see the discussion in RFC 3267 for AMR audio) or as a standalone payload
format (e.g. RFC 2733 or RFC 2198). 

There are also guidelines for authors of payload formats in RFC 2736 that
may be of interest. 

This is all handled by the AVT group: http://www.ietf.org/html.charters/avt-charter.html

Cheers,
Colin
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 09:24:05 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07299
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 09:24:05 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB6EQS427030
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 09:26:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6EQSv27027
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 09:26:28 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07287
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 09:23:34 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6EP4v26986;
	Fri, 6 Dec 2002 09:25:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6EOTv26955
	for <ieprep@optimus.ietf.org>; Fri, 6 Dec 2002 09:24:29 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07229
	for <ieprep@ietf.org>; Fri, 6 Dec 2002 09:21:34 -0500 (EST)
Received: from extremenetworks.com (unknown [10.112.40.61])
	by gnat.inet.org (Postfix) with ESMTP
	id F20D067103; Fri,  6 Dec 2002 05:50:29 -0500 (EST)
Date: Fri, 6 Dec 2002 09:24:22 -0500
Subject: Re: [Ieprep] Janet's example
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149BC@chntex02.is.dyncorp.com>
Message-Id: <6A2C67A3-0926-11D7-B135-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Thursday, Dec 5, 2002, at 10:24 America/Montreal, Gunn, Janet wrote:
> Hypothetical example (sticking to VoIP as the example, but I don't 
> want to
> restrict the discussion to JUST VoIP):

... (and thanks for being so detailed with the example, I think having
a specific example to focus on will in turn help focus discussion)...

> Then the question is: "Is there a way for site A to mark "the traffic" 
> (at
> whatever layer is most likely to work) so that it recognized as 
> "emergency
> traffic" on both access links?

	Inside SIP, subject to stuff in draft-carlberg-ets-telephony-*.txt
and Henning's draft getting added (if not already there in SIP),
then clearly it ought to be possible to mark things from a SIP
application-layer perspective.  So this will help both for VoIP
stuff and in pass-through (PSTN--VoIP--PSTN) cases.  Marking inside
SIP does NOT imply anything about what happens at a layer below SIP,
however.

> And for the traffic to be given the Sprint "emergency treatment" at 
> site A's
> access link and Genuity's "emergency treatment" at site B, which might 
> actually
> be different treatments?"

	Yes.  And each router would need to be specifically configured.

This configuration might look like a filter (source = FEMA VoIP subnets,
destination = RedCross VoIP subnets, protocol = UDP, and port = {member 
of
the particular port range used by VoIP} combined with some specialised 
queuing
treatment (e.g. Weighted Fair Queuing, with preferred weight for the
special VoIP traffic).

If one makes the further assumption that each site is using modern 
IP/Ethernet
equipment internal to the site, one might want to create DiffServ ToS 
and/or
802.1p ToS markings on the last acccess router in the path -- so that 
the deployed
Ethernet and IP infrastructure within the site might be able to use 
those labels
(intra-domain) for preferential queuing in routers and switches.

Also, if that further assumption is made, then one choice would be to 
config
the originating IP Telephone device to mark its outbound packets with 
some
DiffServ/IP ToS and 802.1p ToS markings so that the originating site 
might
be able to use those labels (intra-domain) for preferential queuing in 
routers
and switches.

> If it is possible, is it practical, and what are the drawbacks, and to 
> whom?

I think it is practical.
Drawbacks are:
	- need to add the special configurations to the applicable 
routers/switches:
		- minimum in the site's access routers
		- preferably also in the ISP's access router on the other side of the
			access link
		- preferably also in the switches and routers within each site
	- increased complexity of the router configurations
	- need to deploy source address filtering throughout the FEMA site
	  and the RedCross site to prevent source address spoofing
	  (IMHO, not a big deal since that is commercial best practice anyway
	  as a more general security measure for any end site) [1]
	- need to have equipment with the right feature set (or spend money
		to purchase/deploy equipment with the right feature set).
	- probably need to pay the ISP to perform the needed config on its
		access router on its side of the access link (depending on ISP,
		this might or might not be a service offered)
	- might need to change ISPs from an ISP that won't enable special
		queuing (on its access router on its side of the access link)
		to an ISP that will do it (presumably for an extra fee).
	- depending on the knowledge level of existing IT/LAN staff at each
		site, might need to either provide more training or hire a
		consultant to help figure out the desired configs for all the
		equipment.
	- more complicated configs are harder to debug if a problem arises
	- more complicated configs are harder to replicate if equipment
		fails and/or is otherwise replaced by other equipment
	- the intra-site network design/deployment is more complicated.

This increased complexity translates into several high-level issues:
	- probably a longer MTTR for the system overall
	- probably a shorter MTBF for the system overall
	- possibly increased capital costs at each site
		(depends on whether existing equipment has the right features already)
	- probably increased ISP uplink costs
	- certainly increased operational costs for IT staff
		(either more training for existing staff, more expensive staff,
			or hiring outside consultants)

I hope this helps.

Ran
rja@extremenetworks.com

[1] For reasons I described in detail in an earlier note, it is not 
practical
for an ISP to deploy source-address-spoofing filters on links to most 
other
ISPs (i.e. inter-domain situation).

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 09:31:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07622
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 09:31:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB6EXX027666
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 09:33:33 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6EXXv27662
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 09:33:33 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07582
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 09:30:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6ET1v27243;
	Fri, 6 Dec 2002 09:29:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6ES3v27153
	for <ieprep@optimus.ietf.org>; Fri, 6 Dec 2002 09:28:03 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07353
	for <ieprep@ietf.org>; Fri, 6 Dec 2002 09:25:07 -0500 (EST)
Received: from extremenetworks.com (unknown [10.112.40.61])
	by gnat.inet.org (Postfix) with ESMTP
	id 0B86167103; Fri,  6 Dec 2002 05:54:04 -0500 (EST)
Date: Fri, 6 Dec 2002 09:27:56 -0500
Subject: Re: [Ieprep] Kimberly's concept for Janet's example
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep '" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A01@mcl-its-exs01.mail.saic.com>
Message-Id: <E9B542FE-0926-11D7-B135-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Thursday, Dec 5, 2002, at 13:16 America/Montreal, King, Kimberly S. 
wrote:
> The only thing I can think of at this time that might be possible if 
> you
> could get the providers to agree to it would be to provide some kind of
> secure tunnel between them.  The ISPs would continue to zero out ToS 
> on all
> incoming packets but it wouldn't matter because the emergency ones 
> would be
> marked accordingly within the tunnel.
>
> Here I'm assuming a secure tunnel between ISP X and ISP Y.  There 
> would have
> to be some way to figure out how to put emergency traffic and only 
> emergency
> traffic into that tunnel.  Traffic can be remarked according to local 
> use
> perhaps requiring ISP X and Y to share information.  They would have to
> trust each other (gulp).  I suppose they could police it to somewhat 
> limit
> DoS.

Another approach is to encrypt all traffic between sites A and B (good
security practice anyway), but use filters inside routers inside each
site to set appropriate IP ToS bits on the relevant packets.

> Drawbacks include complexity adds security risks (plus all the ones Ran
> listed associated with complexity in an earlier email), the design 
> isn't
> scalable, nicely identifying and aggregating "emergency traffic" will
> perhaps make it an easier target for attackers, the cost could well 
> exceed
> the price of simply buying bigger access pipes on these locations that 
> are
> already recognized to carry emergency traffic, and I'm sure there are 
> lots
> more.

The above should also work.  It is more complex than the scenario I 
outlined
in email a moment ago (because it needs most of the stuff I outlined 
plus
an IPsec VPN box) -- but it provides site-to-site security (which is
probably strongly desirable anyway).

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 13:25:55 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA17965
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 13:25:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB6ISMs11629
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 13:28:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6ISLv11626
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 13:28:21 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA17919
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 13:25:24 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6IQNv11504;
	Fri, 6 Dec 2002 13:26:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6INZv11392
	for <ieprep@optimus.ietf.org>; Fri, 6 Dec 2002 13:23:35 -0500
Received: from imo-r03.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA17699
	for <ieprep@ietf.org>; Fri, 6 Dec 2002 13:20:38 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r03.mx.aol.com (mail_out_v34.13.) id b.148.445fa65 (4394);
	Fri, 6 Dec 2002 13:22:10 -0500 (EST)
Message-ID: <148.445fa65.2b2244d2@aol.com>
Date: Fri, 6 Dec 2002 13:22:10 EST
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
To: woody@pch.net, KIMBERLY.S.KING@saic.com
CC: rja@extremenetworks.com, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_148.445fa65.2b2244d2_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_148.445fa65.2b2244d2_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/5/2002 8:32:51 AM Eastern Standard Time, woody@pch.net 
writes:


>    > > [MAP] If 10 or 20 or whatever number of those "several thousands" of 
> end
>     > > users tried to make a VoIP call via that single 384K link, what 
> would
>     > > have happened?
>     > > 1. all calls to be of unacceptable quality
> 
> [BW] Correct, that's a reasonable assumption.
> 
>     > [MAP] I'll repeat the question which I don't believe I've gotten any
>     > concrete answer to: In the case above of the 10 callers from Nairobi, 
> if
>     > one of them was authorized to make emergency calls and the other 9
>     > weren't...
> 
> [BW] Authorized by whom?
> 
>     > [MAP] what specific standardized mechanism (not products) could be
>     > used to provide the required service across the congested 384 kbit/s
>     > satellite link? (The required service is that their call gets set up 
> and
>     > they can talk.)
> 
> [BW] If they were authorized by _me_, that is, if they were paying _me_ to
> deliver their call in preference to the other nine callers, I'd
> policy-route their SIP traffic into a separate PVC at the CPE router.
> That's what I did back in the distant past, when congestion still occurred
> occasionally, and it seemed to work quite satisfactorily at the time.
> 

[MAP] I'll try one more time. The question presumes the following situation:

1. There is a government agency that "authorizes" indiviuduals to get 
priority treatment. (How or why is not the issue here)
2. There is a service provider who has agreed by contract to support this 
capabilitity. The service provider does not "authorize" anyone, but presume 
that they must "authenticate" its use. (Whether they are "paid" to do this is 
not the issue.)
3. There is a single 384 kbit/s path that must carry all the traffic from the 
area (the redundant paths may have been taken out by the terrorist act). 
There is no other "PVC" to route traffic onto. (Why the service provider 
didn't have sufficient backup is not the issue.)
4. There is excessive congestion on the 384 kbit/s path, caused by too many 
voice calls, all of which are getting unacceptable quality. (Any data or 
elastic traffic has already been squeezed out.)

If you don't accept the above as the case being discussed, please do not 
respond.

How is the 1 authorized individual able to get their call through when the 
link is congested by users without the special authorization from the 
government?

Mike Pierce
Artel


--part1_148.445fa65.2b2244d2_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/5/2002 8:32:51 AM Eastern Standard Time, woody@pch.net writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px"> &nbsp;&nbsp;&gt; &gt; [MAP] If 10 or 20 or whatever number of those "several thousands" of end
<BR> &nbsp;&nbsp;&nbsp;&gt; &gt; users tried to make a VoIP call via that single 384K link, what would
<BR> &nbsp;&nbsp;&nbsp;&gt; &gt; have happened?
<BR> &nbsp;&nbsp;&nbsp;&gt; &gt; 1. all calls to be of unacceptable quality
<BR>
<BR>[BW] Correct, that's a reasonable assumption.
<BR>
<BR> &nbsp;&nbsp;&nbsp;&gt; [MAP] I'll repeat the question which I don't believe I've gotten any
<BR> &nbsp;&nbsp;&nbsp;&gt; concrete answer to: In the case above of the 10 callers from Nairobi, if
<BR> &nbsp;&nbsp;&nbsp;&gt; one of them was authorized to make emergency calls and the other 9
<BR> &nbsp;&nbsp;&nbsp;&gt; weren't...
<BR>
<BR>[BW] Authorized by whom?
<BR>
<BR> &nbsp;&nbsp;&nbsp;&gt; [MAP] what specific standardized mechanism (not products) could be
<BR> &nbsp;&nbsp;&nbsp;&gt; used to provide the required service across the congested 384 kbit/s
<BR> &nbsp;&nbsp;&nbsp;&gt; satellite link? (The required service is that their call gets set up and
<BR> &nbsp;&nbsp;&nbsp;&gt; they can talk.)
<BR>
<BR>[BW] If they were authorized by _me_, that is, if they were paying _me_ to
<BR>deliver their call in preference to the other nine callers, I'd
<BR>policy-route their SIP traffic into a separate PVC at the CPE router.
<BR>That's what I did back in the distant past, when congestion still occurred
<BR>occasionally, and it seemed to work quite satisfactorily at the time.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>[MAP] I'll try one more time. The question presumes the following situation:
<BR>
<BR>1. There is a government agency that "authorizes" indiviuduals to get priority treatment. (How or why is not the issue here)
<BR>2. There is a service provider who has agreed by contract to support this capabilitity. The service provider does not "authorize" anyone, but presume that they must "authenticate" its use. (Whether they are "paid" to do this is not the issue.)
<BR>3. There is a single 384 kbit/s path that must carry all the traffic from the area (the redundant paths may have been taken out by the terrorist act). There is no other "PVC" to route traffic onto. (Why the service provider didn't have sufficient backup is not the issue.)
<BR>4. There is excessive congestion on the 384 kbit/s path, caused by too many voice calls, all of which are getting unacceptable quality. (Any data or elastic traffic has already been squeezed out.)
<BR>
<BR>If you don't accept the above as the case being discussed, please do not respond.
<BR>
<BR>How is the 1 authorized individual able to get their call through when the link is congested by users without the special authorization from the government?
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_148.445fa65.2b2244d2_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 15:09:50 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA22564
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 15:09:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB6KCGu17818
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 15:12:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6KCGv17806
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 15:12:16 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA22554
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 15:09:18 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6K9dv17658;
	Fri, 6 Dec 2002 15:09:39 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6K6Sv16984
	for <ieprep@optimus.ietf.org>; Fri, 6 Dec 2002 15:06:28 -0500
Received: from sj-msg-core-4.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA22326
	for <ieprep@ietf.org>; Fri, 6 Dec 2002 15:03:31 -0500 (EST)
Received: from imop.cisco.com (IDENT:mirapoint@imop.cisco.com [171.69.11.44])
	by sj-msg-core-4.cisco.com (8.12.2/8.12.2) with ESMTP id gB6K6KLW014160;
	Fri, 6 Dec 2002 12:06:20 -0800 (PST)
Received: from cisco.com (ssh-sjc-1.cisco.com [171.68.225.134])
	by imop.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABK37017;
	Fri, 6 Dec 2002 12:03:14 -0800 (PST)
Date: Fri, 6 Dec 2002 12:06:36 -0800
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others 
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: Mpierce1@aol.com, ieprep@ietf.org
To: Colin Perkins <csp@isi.edu>
From: Rohan Mahy <rohan@cisco.com>
In-Reply-To: <200212032249.gB3MnLe22085@chiron.nge.isi.edu>
Message-Id: <394597BC-0956-11D7-9AE1-0003938AF740@cisco.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, December 3, 2002, at 02:49 PM, Colin Perkins wrote:

> --> Mpierce1@aol.com writes:
>> In a message dated 12/3/2002 3:52:30 PM Eastern Standard Time, 
>> csp@isi.edu
>> writes:
>>
>>> If you implement RTP correctly, the RTP control protocol provides 
>>> feedback
>>> on the packet loss, latency and jitter at the other end [RFC 1889].
>>>
>>
>> But:
>>
>> In a message dated 12/2/2002 1:42:11 PM Eastern Standard Time,
>> sob@harvard.edu writes:
>>
>>
>>>> 2. new call attempts would have been rejected
>>>
>>> no- the congested link would not cause the SIP proxies to reject 
>>> calls
>>> n
>>
>>
>> [MAP] Guess I'm missing something, since the RTP functionality 
>> mentioned by
>> Colin is the type of thing that I thought would allow a call control
>> application to know that it should stop trying to set up more calls. 
>> Scott
>> (and others) say there is no way to do it.
>
> The media doesn't necessarily follow the same path as the signalling, 
> and
> so the SIP proxy may not see the RTP reception quality reports.

To be more specific.  It is extremely unlikely that SIP intermediaries 
would be privy to or care about RTCP quality reports.  The endpoints 
would decide that one or more media flows had insufficient quality, and 
would modify the session to modify or exclude those flows (possibly in 
favor of a lower-bitrate audio flow or a flow for text chat) or one end 
would simply terminate the session.

thanks,
-rohan

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 16:12:20 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA24575
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 16:12:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB6LEls21212
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 16:14:47 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6LElv21209
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 16:14:47 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA24569
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 16:11:49 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6LD4v21143;
	Fri, 6 Dec 2002 16:13:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB6LCfv21129
	for <ieprep@optimus.ietf.org>; Fri, 6 Dec 2002 16:12:41 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA24462
	for <ieprep@ietf.org>; Fri, 6 Dec 2002 16:09:43 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Fri, 6 Dec 2002 16:11:55 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120616121420727
 ; Fri, 06 Dec 2002 16:12:14 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPJ1BS0>; Fri, 6 Dec 2002 16:12:01 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A0F@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
Cc: "'Henry.Sinnreich@wcom.com'" <Henry.Sinnreich@wcom.com>
Date: Fri, 6 Dec 2002 16:12:12 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] commercial SIP service
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Worldcom offers IP Communications, a converged voice/data service using SIP
mainly targeted at small to medium businesses.   I contacted them on behalf
of ieprep as a result of last week's discussions and Henry Sinnreich was
kind enough to provide some answers.

They will:

Provision CPE to support a fixed maximum number of voice calls over a
customer access link.  This is based upon policy of UDP to TCP ratios and
also maximum data burst sizes.  What this means is that the customers may
choose more or fewer maximum voice connections (as a function of the link
speed) with the tradeoff of how fast they can send data during times when
all those virtual voice channels are in use.

They use a QoS appliance at the customer premise to track the number of
calls.  (Kimberly note:  These QoS appliances are alternate way that some
consider much easier to use than QoS provisioning of a CPE router.  The
appliances usually sport a GUI interface.) 

They achieve the same call quality when there is an IP-PSTN call. 
And Henry claims IP-IP is better quality than the PSTN, since it is the
ultimate 4-wire link: 
Close to 4 kHz vs. 3.1 kHz (on PSTN), no noise, no distortion, no 
copper-frequency attenuation, no echo, if correctly implemented.  

The service is similar to an IP-Centrix with SIP servers in the Worldcom
backbone providing the call signaling function.

They are not using QoS in their backbone.

They provide a list of firewalls that interoperate with the SIP service.

Most of the 60,000 WorldCom employees have access to the IP phones at their
desks. 

Some employees may also use VPN technology and "take their office phone" to
their new location accessing over the Internet. They can use the SIP 
based Windows Messenger 4.7 or the Pingtel PC soft phones. The narrowband
G.723.1 codec in the Windows Messenger, allows business quality voice calls
over RASNet (UUNet) dial-up to the Internet (such as from hotel rooms). The
Windows Messenger will be part of the Connection early in 1Q03.

Should a "SIP Resource Priority Mechanism" become available, they do not
anticipate difficulties in supporting such a mechanism at gateways to the
PSTN if the market calls for such a need.  This could be a potential
enhancement of the GETS service.  

What would happen if someone dialed a GETS call today using IP Comm?
Either the "shared local gateway" in the WorldCom network, or a physical
local gateway box will connect such calls (911, 800, 411, etc) to the local
phone network.

What types of PSTN signaling are used?  
"Network Gateways" connect to the PSTN using SS7 or PRI, while "Enterprise
Gateways" connect the PBX to the local IP/LAN using Q.931 or CAS signaling. 



- Kimberly




_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec  6 20:59:24 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA05320
	for <ieprep-archive@odin.ietf.org>; Fri, 6 Dec 2002 20:59:23 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB721qL05513
	for ieprep-archive@odin.ietf.org; Fri, 6 Dec 2002 21:01:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB721qv05510
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 6 Dec 2002 21:01:52 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA05311
	for <ieprep-web-archive@ietf.org>; Fri, 6 Dec 2002 20:58:52 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB720Ev05443;
	Fri, 6 Dec 2002 21:00:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB71x1v05392
	for <ieprep@optimus.ietf.org>; Fri, 6 Dec 2002 20:59:01 -0500
Received: from sj-msg-core-4.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA05252
	for <ieprep@ietf.org>; Fri, 6 Dec 2002 20:56:01 -0500 (EST)
Received: from imop.cisco.com (IDENT:mirapoint@imop.cisco.com [171.69.11.44])
	by sj-msg-core-4.cisco.com (8.12.2/8.12.2) with ESMTP id gB71whKv025033;
	Fri, 6 Dec 2002 17:58:43 -0800 (PST)
Received: from cisco.com (ssh-sjc-1.cisco.com [171.68.225.134])
	by imop.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABK40975;
	Fri, 6 Dec 2002 17:55:37 -0800 (PST)
Date: Fri, 6 Dec 2002 17:59:00 -0800
Subject: Re: [Ieprep] identification of packets
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'Bill Woodcock'" <woody@pch.net>,
        "'RJ Atkinson'" <rja@extremenetworks.com>, ieprep <ieprep@ietf.org>
To: "Gunn, Janet" <Janet.Gunn@dyncorp.com>
From: Rohan Mahy <rohan@cisco.com>
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149BC@chntex02.is.dyncorp.com>
Message-Id: <74761E44-0987-11D7-9AE1-0003938AF740@cisco.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Hi Janet,

Your scenario: FEMA wants to communicate with Red Cross is a useful 
one.  While a few folks have responded with QoS related solutions and 
VPNs, it is quite possible that the cost for FEMA and Red Cross to each 
get several (diverse) big pipes well connected to the core may cost 
much less than turning on QoS on smaller pipes.  I think these sites 
could certainly justify a couple of OC3s which is where the queuing 
delay starts to become negligible.

thanks,
-rohan


On Thursday, December 5, 2002, at 07:24 AM, Gunn, Janet wrote:

> The higher the layer the better, I think.
>
> Hypothetical example (sticking to VoIP as the example, but I don't 
> want to
> restrict the discussion to JUST VoIP):
>
> Caller is at a site A, served by SPRINT, with their congestion control
> mechanism implemented on the access link.
>
> Celled party is at site B served by Genuity, with their congestion 
> control
> mechanism implemented on the access link.
>
> Assume that both site A and Site B have configured their access lines 
> so
> that there is  category of traffic which is called (by people, not
> necessarily machines) "emergency"?
>
> Assume, for the time being, that the sites are owned by "FEMA" and the 
> "Red
> Cross", so that you know in advance that there is likely to be 
> emergency
> traffic between them during an "incident", and any business 
> arrangements are
> worked out in advance. (I am also interested in the "opportunistic" 
> case
> where there is not a business arrangement in advance (e.g., Internet 
> Cafe),
> but that seems much harder, and much more open to DoS abuse.)
>
> Then the question is: "Is there a way for site A to mark "the traffic" 
> (at
> whatever layer is most likely to work) so that it recognized as 
> "emergency
> traffic" on both access links?  And for the traffic to be given the 
> Sprint
> "emergency treatment" at site A's access link and Genuity's "emergency
> treatment" at site B, which might actually be different treatments?"
>
> If it is possible, is it practical, and what are the drawbacks, and to 
> whom?
>
> I am asking nothing from the ISP core, except to pass the "marking" 
> (so ToS
> bits won't work).
>
>
> -----Original Message-----
> From: Bill Woodcock [mailto:woody@pch.net]
> Sent: Thursday, December 05, 2002 8:11 AM
> To: Gunn, Janet
> Cc: 'RJ Atkinson'; ieprep
> Subject: RE: [Ieprep] identification of packets
>
>
>       On Mon, 2 Dec 2002, Gunn, Janet wrote:
>> Is there a consistent way to identify emergency traffic across several
>> different "entities that are addressing congestion on the access
> links"?
>
> At what layer?  The lower the layer, the more expensive doing so would 
> be,
> and the more difficult it would be to avoid spoofing or misuse.
>
>> And, is there a way to pass that "identification as emergency traffic"
> from
>> the access links at the originating node to the access links at the
>> terminating node (if the terminating node is in another domain)? (From
> what
>> I have read here, and ToS bits based marking would be lost in the
>> intervening ISP.)
>
> Excuse the BUAG.  Topology looks like this:
>
>  Cust A                  |     ISP A    |     ISP B     |          
> Cust B
>
>  LAN--CPE rtr--tail ckt--acc rtr--core--ix--core--acc 
> rtr--tail--CPE--LAN
>
> The phones and SIP devices are all on those LANs.  The CPE routers 
> aren't
> capable of manipulating anything or doing any QoS or anything like 
> that.
> The tail circuits are the portion where there's economic incentive to
> congest them occasionally.  The portion between the access routers is 
> the
> part under the ISPs control, collectively, but it doesn't experience
> congestion, and can't afford the high failure rate associated with QoS,
> even if QoS had some beneficial effect.
>
> So effectively, what you've got to play with is the stuff going onto 
> the
> LANs.  But fortunately, that's where you get your big win, since that's
> what does all the call setup, origination, answering, redirection, et
> cetera.
>
>                                 -Bill
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec  7 00:43:04 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10052
	for <ieprep-archive@odin.ietf.org>; Sat, 7 Dec 2002 00:43:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB75jai15020
	for ieprep-archive@odin.ietf.org; Sat, 7 Dec 2002 00:45:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75jav15017
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 7 Dec 2002 00:45:36 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10043
	for <ieprep-web-archive@ietf.org>; Sat, 7 Dec 2002 00:42:33 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75huv14986;
	Sat, 7 Dec 2002 00:43:56 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75gKv14965
	for <ieprep@optimus.ietf.org>; Sat, 7 Dec 2002 00:42:20 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA09960
	for <ieprep@ietf.org>; Sat, 7 Dec 2002 00:39:17 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB75Zvo06568;
	Fri, 6 Dec 2002 21:35:57 -0800 (PST)
Date: Fri, 6 Dec 2002 21:35:57 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
cc: "'RJ Atkinson'" <rja@extremenetworks.com>, ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149BC@chntex02.is.dyncorp.com>
Message-ID: <Pine.GSO.4.44.0212062129470.6478-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Thu, 5 Dec 2002, Gunn, Janet wrote:
    > Then the question is: "Is there a way for site A to mark "the traffic" (at
    > whatever layer is most likely to work) so that it recognized as "emergency
    > traffic" on both access links?  And for the traffic to be given the Sprint
    > "emergency treatment" at site A's access link and Genuity's "emergency
    > treatment" at site B, which might actually be different treatments?"
    > If it is possible, is it practical, and what are the drawbacks, and to whom?
    > I am asking nothing from the ISP core, except to pass the "marking" (so ToS
    > bits won't work).

ISP B's access router would need to recognize and operate upon the
marking, as would ISP A's access router for the reply traffic, so they do
both have to understand it.  It's quite possible _using_ the ToS bits, but
for all the aforestated reasons, that's not practical, largely because it
would be abused.

If we know _in advance_ that the emergency traffic is going to be between
A and B, _or_ if we postulate somewhat smarter CPE routers than we have
today, then we can do this with static configuration in the access
routers and no new markings...  A's CPE router (or a separate
tunnel-endpoint termination device) policy-routes ToS-marked traffic into
tunnels, and both A and B just pay their ISPs a bit more for a service
which prioritizes traffic in established tunnels.  (As opposed to set-up
traffic, which could be spoofed to DoS an endpoint.)

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec  7 00:46:24 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10158
	for <ieprep-archive@odin.ietf.org>; Sat, 7 Dec 2002 00:46:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB75mu415319
	for ieprep-archive@odin.ietf.org; Sat, 7 Dec 2002 00:48:56 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75muv15316
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 7 Dec 2002 00:48:56 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10151
	for <ieprep-web-archive@ietf.org>; Sat, 7 Dec 2002 00:45:53 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75l2v15189;
	Sat, 7 Dec 2002 00:47:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75kOv15091
	for <ieprep@optimus.ietf.org>; Sat, 7 Dec 2002 00:46:24 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10081
	for <ieprep@ietf.org>; Sat, 7 Dec 2002 00:43:21 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB75dwo06592;
	Fri, 6 Dec 2002 21:39:58 -0800 (PST)
Date: Fri, 6 Dec 2002 21:39:58 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
cc: "'Gunn, Janet '" <Janet.Gunn@dyncorp.com>,
        "''RJ Atkinson' '" <rja@extremenetworks.com>,
        "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A01@mcl-its-exs01.mail.saic.com>
Message-ID: <Pine.GSO.4.44.0212062136360.6478-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Thu, 5 Dec 2002, King, Kimberly  S. wrote:
    > Note that you are asking for a form of multiple ISP QoS...

Not necessarily...  Janet managed to break the problem down into
bite-sized market-addressable pieces...  If A and B can each
independently negotiate market-priced services with their own ISPs, and
there doesn't need to be any unfunded many-to-many relationship between
the ISPs themselves to make it work, then we've got a potential solution.

    > The only thing I can think of at this time that might be possible if you
    > could get the providers to agree to it would be to provide some kind of
    > secure tunnel between them.

Between the endpoints...  Ideally between the phones themselves, in fact.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec  7 00:49:15 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10292
	for <ieprep-archive@odin.ietf.org>; Sat, 7 Dec 2002 00:49:15 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB75plm15461
	for ieprep-archive@odin.ietf.org; Sat, 7 Dec 2002 00:51:47 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75plv15458
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 7 Dec 2002 00:51:47 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10239
	for <ieprep-web-archive@ietf.org>; Sat, 7 Dec 2002 00:48:44 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75o1v15385;
	Sat, 7 Dec 2002 00:50:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75nFv15337
	for <ieprep@optimus.ietf.org>; Sat, 7 Dec 2002 00:49:15 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10155
	for <ieprep@ietf.org>; Sat, 7 Dec 2002 00:46:12 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB75gvo06630;
	Fri, 6 Dec 2002 21:42:57 -0800 (PST)
Date: Fri, 6 Dec 2002 21:42:57 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: RJ Atkinson <rja@extremenetworks.com>
cc: "'ieprep '" <ieprep@ietf.org>
Subject: Re: [Ieprep] identification of packets
In-Reply-To: <7357265A-0923-11D7-B135-00039357A82A@extremenetworks.com>
Message-ID: <Pine.GSO.4.44.0212062141230.6478-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Fri, 6 Dec 2002, RJ Atkinson wrote:
    > At least several vendors still make/sell CPE routers that have those
    > capabilities.

Oh, no argument there...  Lots of vendors are still designing and building
high-end equipment which they're trying to sell as CPE...  The issue is
that fewer and fewer customers are buying it as such.  More and more are
getting Ethernet-to-Ethernet or DSL-to-Ethernet devices instead.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec  7 00:54:48 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10455
	for <ieprep-archive@odin.ietf.org>; Sat, 7 Dec 2002 00:54:48 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB75vKI15786
	for ieprep-archive@odin.ietf.org; Sat, 7 Dec 2002 00:57:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75vKv15783
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 7 Dec 2002 00:57:20 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10432
	for <ieprep-web-archive@ietf.org>; Sat, 7 Dec 2002 00:54:16 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75u2v15749;
	Sat, 7 Dec 2002 00:56:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB75tlv15732
	for <ieprep@optimus.ietf.org>; Sat, 7 Dec 2002 00:55:47 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id AAA10368
	for <ieprep@ietf.org>; Sat, 7 Dec 2002 00:52:44 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB75nQo06677;
	Fri, 6 Dec 2002 21:49:26 -0800 (PST)
Date: Fri, 6 Dec 2002 21:49:26 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: Mpierce1@aol.com
cc: KIMBERLY.S.KING@saic.com, <rja@extremenetworks.com>, <ieprep@ietf.org>
Subject: Re: [Ieprep] Internet faults, VoIP demo, etc and others
In-Reply-To: <148.445fa65.2b2244d2@aol.com>
Message-ID: <Pine.GSO.4.44.0212062144390.6478-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Fri, 6 Dec 2002 Mpierce1@aol.com wrote:
    > 1. There is a government agency that "authorizes" indiviuduals to get
    > priority treatment.

And that would be what, the New World Order government?  Sign me up,
Dogbert!  The example you gave spans two countries plus outer space.  Last
I knew neither outer space nor Kenya were suburbs of Washington D.C.

    > 2. There is a service provider who has agreed by contract to support this
    > capabilitity. The service provider does not "authorize" anyone, but presume
    > that they must "authenticate" its use. (Whether they are "paid" to do this is
    > not the issue.)

I thought you were worried about traffic getting dropped?  Any traffic
that isn't paid for is going to get dropped.  So you want to make sure
your solution includes some method of at least not guaranteeing that all
of this traffic is going to get dropped.

    > 3. There is a single 384 kbit/s path that must carry all the traffic from the
    > area (the redundant paths may have been taken out by the terrorist act).
    > There is no other "PVC" to route traffic onto. (Why the service provider
    > didn't have sufficient backup is not the issue.)

PVC means "Permanent Virtual Circuit."  A PVC is not an alternate or
backup path.  It's a preexisting and widely-used layer-2 tagging
mechanism.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec  7 08:07:40 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA27389
	for <ieprep-archive@odin.ietf.org>; Sat, 7 Dec 2002 08:07:40 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB7DA4R12241
	for ieprep-archive@odin.ietf.org; Sat, 7 Dec 2002 08:10:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB7DA4v12238
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 7 Dec 2002 08:10:04 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA27386
	for <ieprep-web-archive@ietf.org>; Sat, 7 Dec 2002 08:07:09 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB7D8Kv12176;
	Sat, 7 Dec 2002 08:08:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB7D7Dv11998
	for <ieprep@optimus.ietf.org>; Sat, 7 Dec 2002 08:07:13 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA27326
	for <ieprep@ietf.org>; Sat, 7 Dec 2002 08:04:18 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Sat, 7 Dec 2002 08:06:37 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120708065632618
 ; Sat, 07 Dec 2002 08:06:56 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPJ18GD>; Sat, 7 Dec 2002 08:06:43 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A10@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Rohan Mahy '" <rohan@cisco.com>,
        "'Gunn, Janet '" <Janet.Gunn@dyncorp.com>
Cc: "''Bill Woodcock' '" <woody@pch.net>,
        "''RJ Atkinson' '" <rja@extremenetworks.com>,
        "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
Date: Sat, 7 Dec 2002 08:06:55 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Rohan said:
>Hi Janet,

>Your scenario: FEMA wants to communicate with Red Cross is a useful 
>one.  While a few folks have responded with QoS related solutions and 
>VPNs, it is quite possible that the cost for FEMA and Red Cross to each 
>get several (diverse) big pipes well connected to the core may cost 
>much less than turning on QoS on smaller pipes.  I think these sites 
>could certainly justify a couple of OC3s which is where the queuing 
>delay starts to become negligible.

I completely agree with Rohan and it is also useful to keep in mind that
while the kind of QoS related solutions would be difficult enough between
two sites, it gets wildly out of control if you wanted to ensure that kind
of situation between the top 100 sites.  For each site to maintain special
information between itself and 99 other sites is just asking for major
disaster (no pun intended).

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  8 17:20:05 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA11173
	for <ieprep-archive@odin.ietf.org>; Sun, 8 Dec 2002 17:20:05 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB8MMXM12874
	for ieprep-archive@odin.ietf.org; Sun, 8 Dec 2002 17:22:33 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB8MMXv12871
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 8 Dec 2002 17:22:33 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA11167
	for <ieprep-web-archive@ietf.org>; Sun, 8 Dec 2002 17:19:34 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB8MKVv12806;
	Sun, 8 Dec 2002 17:20:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB8M3nv11705
	for <ieprep@optimus.ietf.org>; Sun, 8 Dec 2002 17:03:49 -0500
Received: from joy.songbird.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA10785
	for <ieprep@ietf.org>; Sun, 8 Dec 2002 17:00:49 -0500 (EST)
Received: from bbprime.brandenburg.com (208.184.79.252.songbird.com [208.184.79.252] (may be forged))
	by joy.songbird.com (8.9.3/8.9.3) with ESMTP id OAA26151
	for <ieprep@ietf.org>; Sun, 8 Dec 2002 14:03:26 -0800
Date: Sun, 8 Dec 2002 12:10:39 -0800
From: Dave Crocker <dhc@dcrocker.net>
X-Mailer: The Bat! (v1.61) Personal
Organization: TribalWise
X-Priority: 3 (Normal)
Message-ID: <6464409896.20021208121039@tribalwise.com>
To: ieprep@ietf.org
Subject: Re: [Ieprep] identification of packets
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A10@mcl-its-exs01.mail.saic.com>
References: 
 <B8030EB94AF1D51196D70002A589D64207E77A10@mcl-its-exs01.mail.saic.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Folks


>>Your scenario: FEMA wants to communicate with Red Cross is a useful
>>one.  While a few folks have responded with QoS related solutions and 
>>VPNs, it is quite possible that the cost for FEMA and Red Cross to each 
>>get several (diverse) big pipes well connected to the core may cost 
>>much less than turning on QoS on smaller pipes.

Kimberly> I completely agree with Rohan

I also believe that messing with infrastructure requirements, for this stage
of IEPrep effort, will not be productive.

However, I thought that IEPrep was intended for use in special circumstances
that challenge normal operation. For example, a hurricane has hit.

How do FEMA and Red Cross go about getting those big fat pipes when there
are almost not data pipes around at all?

And for that matter, let's consider that it won't be just FEMA and Red
Cross.  It will be a mass of local, provincial, and national organizations
(and even inter-national organizations when entire regions are affected.)
Under these circumstances, a pair-wise solution does not scale to handle
everyone.

d/
-- 
 Dave Crocker  <mailto:dhc@dcrocker.net>
 TribalWise <http://www.tribalwise.com>
 t +1.408.246.8253; f +1.408.850.1850

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sun Dec  8 21:00:07 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA14469
	for <ieprep-archive@odin.ietf.org>; Sun, 8 Dec 2002 21:00:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB922cH21437
	for ieprep-archive@odin.ietf.org; Sun, 8 Dec 2002 21:02:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB922cv21434
	for <ieprep-web-archive@optimus.ietf.org>; Sun, 8 Dec 2002 21:02:38 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA14448
	for <ieprep-web-archive@ietf.org>; Sun, 8 Dec 2002 20:59:36 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB91w9v21313;
	Sun, 8 Dec 2002 20:58:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB91vwv21292
	for <ieprep@optimus.ietf.org>; Sun, 8 Dec 2002 20:57:58 -0500
Received: from kc-msxproto2.kc.umkc.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA14373
	for <ieprep@ietf.org>; Sun, 8 Dec 2002 20:54:56 -0500 (EST)
Received: from KC-MAIL2.kc.umkc.edu ([134.193.143.162] RDNS failed) by kc-msxproto2.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.5329);
	 Sun, 8 Dec 2002 19:57:49 -0600
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Subject: Re: [Ieprep] identification of packets
Date: Sun, 8 Dec 2002 19:57:49 -0600
Message-ID: <A29143A40AEAE3459FF829D1DD43316101D9B078@KC-MAIL2.kc.umkc.edu>
Thread-Topic: Re: [Ieprep] identification of packets
Thread-Index: AcKfJmBTPaRLNnTcSGauIQOEnU4zDA==
From: "Ayyasamy, Senthilkumar  (UMKC-Student)" <saq66@umkc.edu>
To: "Dave Crocker" <dhc@dcrocker.net>, <ieprep@ietf.org>
X-OriginalArrivalTime: 09 Dec 2002 01:57:49.0906 (UTC) FILETIME=[60A90B20:01C29F26]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gB91vwv21293
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


> I also believe that messing with infrastructure
> requirements, for this stage
> of IEPrep effort, will not be productive.

I agree with you here.

  While a number of discussions in this list are 
interesting but most of them (including many of my
previous mails) are not relevant.

 First of all, this working group is not just about 
"operational" requirements. But, requirements which 
other WG can use as a reference when developing a solution 
for ieprep. I accept that it is difficult to
separate ieprep requirements which protocols (like SIP etc)has 
to solve from considerations for deploying such a requirement. 
But, for getting some work done, we have to treat them 
separately.

 Second, we often discuss solutions leaving the important 
requirements. For example, most of the ieprep requirements 
can be satisfied by over-provisioning. I can provide you with 
more solutions for ieprep(multicast, sensor networks, separate 
spectrum as proposed in EMTEL, overlay architecture,
VPN, ISLAY, QoS models etc...) I remember Fred slamming many for 
discussing the solution space. So, first lets do some 
requirements work before thinking about solution. It is relevant 
only when we talk about BCPs. 

 third, it is important for IETF to hear from other communities 
too (here, emergency response community.) At last, they are the 
once who are going to pay you(ISPs)for routing their packets. 

 So, why cant we start a separate interest list (like diffserv
interest list.... If your in Diffserv list, How many mail of yours 
have been bypassed to diffserv-interest list by Brian? ) to talk about 
all this other issues and leave this list for requirements 
discussion. 

 Also, I suggest we keep some separate threads while 
discussing each issue (like the same way
SIP requirements was discussed.) Else, it will boil 
down to another over provisioning discussion. 


  
 
   

 


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 05:44:25 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04412
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 05:44:25 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9Akne24942
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 05:46:49 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Aknv24939
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 05:46:49 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04406
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 05:43:54 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9AjCv24912;
	Mon, 9 Dec 2002 05:45:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9AiUv24877
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 05:44:30 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA04383
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 05:41:36 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 9 Dec 2002 05:44:04 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120905442402574
 ; Mon, 09 Dec 2002 05:44:24 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPH530M>; Mon, 9 Dec 2002 05:45:56 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A11@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Dave Crocker '" <dhc@dcrocker.net>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
Date: Mon, 9 Dec 2002 05:44:17 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Dave said:
>I also believe that messing with infrastructure requirements, for 
>this stage of IEPrep effort, will not be productive.

Agree.

>How do FEMA and Red Cross go about getting those big fat pipes when
>there are almost not data pipes around at all?

See http://tsp.ncs.gov/ for a solution to the above.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 06:18:20 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA04845
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 06:18:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9BKiR26288
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 06:20:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9BKiv26285
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 06:20:44 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA04838
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 06:17:49 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9BJFv26212;
	Mon, 9 Dec 2002 06:19:15 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9BI5v26191
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 06:18:05 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA04810
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 06:15:10 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 9 Dec 2002 06:17:34 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120906175405885
 ; Mon, 09 Dec 2002 06:17:54 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPJG17R>; Mon, 9 Dec 2002 06:17:41 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A12@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Ayyasamy, Senthilkumar  (UMKC-Student) '" <saq66@umkc.edu>,
        "'Dave Crocker '" <dhc@dcrocker.net>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
Date: Mon, 9 Dec 2002 06:17:45 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] identification of problem
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

>Second, we often discuss solutions leaving the important 
>requirements. For example, most of the ieprep requirements 
>can be satisfied by over-provisioning. I can provide you with 
>more solutions for ieprep(multicast, sensor networks, separate 
>spectrum as proposed in EMTEL, overlay architecture,
>VPN, ISLAY, QoS models etc...) I remember Fred slamming many for 
>discussing the solution space. So, first lets do some 
>requirements work before thinking about solution. It is relevant 
>only when we talk about BCPs. 

Fred was absolutely right in discouraging discussions of solutions space
before requirements.  However, something even more fundamental is missing--a
concise statement of the problem to be solved.  

Over two *years* ago, Sean Donelan asked some probing questions.
( http://www.iepscheme.net/archive/0045.html )
"What problems have emergency service users actually had using the
Internet?"

"Have they really had identical problems as with the switched telephone
network, or different problems which don't happen with the switched phone
network."

While the first question is just as open today, we do have some evidence
related to the second one.  Please read the outstanding report "The Internet
Under Crisis Conditions: Learning from September 11 (2002)"
http://www.nap.edu/books/0309087023/html/.  

At the crux of the matter, the Internet is not centrally controlled, is
extremely survivable, and was specifically engineered to avoid congestion
collapse so *if* the problem is to avoid congestion collapse in an emergency
(as defined by the way the PSTN collapses in an emergency) then the answer
probably is that the Internet already does that.  The Internet adapts to
congestion and additionally the evidence shows that people adapt to how they
use the Internet (the communications become more succinct and less rich)
when a traditional emergency hits.  Neither adaptation is possible in the
rigid PSTN world of applications that reserve constant bit rate bandwidth.

More realistically, Internet communications might be more likely to
temporarily collapse from various widespread security attacks than from
increases in user usage or some physical damage as the result of an
emergency.  

In order to discuss requirements, there has to be a crisp definition of the
problem to be solved.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 06:54:57 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05338
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 06:54:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9BvLp27948
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 06:57:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9BvLv27945
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 06:57:21 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05323
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 06:54:26 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Bt5v27845;
	Mon, 9 Dec 2002 06:55:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Brev27753
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 06:53:40 -0500
Received: from kc-msxproto4.kc.umkc.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA05220
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 06:50:45 -0500 (EST)
Received: from KC-MAIL2.kc.umkc.edu ([134.193.143.162] RDNS failed) by kc-msxproto4.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.5329);
	 Mon, 9 Dec 2002 05:53:37 -0600
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="utf-8"
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Date: Mon, 9 Dec 2002 05:53:36 -0600
Message-ID: <A29143A40AEAE3459FF829D1DD43316101D9B07A@KC-MAIL2.kc.umkc.edu>
Thread-Topic: identification of problem
Thread-Index: AcKfdKSusxis7hMnTP+i2e4wiuQwCwAAGlal
From: "Ayyasamy, Senthilkumar  (UMKC-Student)" <saq66@umkc.edu>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>,
        "Dave Crocker " <dhc@dcrocker.net>, <ieprep@ietf.org>
X-OriginalArrivalTime: 09 Dec 2002 11:53:37.0955 (UTC) FILETIME=[9C286F30:01C29F79]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from base64 to 8bit by www1.ietf.org id gB9Brev27754
Subject: [Ieprep] RE: identification of problem
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

> In order to discuss requirements, there has to be a crisp definition of the
> problem to be solved.
    make the charter narrow with specific problems.
 Lets work on problem statement and emergency user requirement
(from various emergency response agencies)  before concentrating
on elastic/inelastic requirements. 

> http://www.nap.edu/books/0309087023/html/. 
      I like the report for its measurement perspective which helps
 us understand how redundant/robust Internet 
 when compared to PSTN.  Note that the report says we still
lack some  measurement infrastruture to accurately measure
the performance.
    But, their are lot of difference between their goal and ours.
They studied robustness of Internet whereas we are focusing
on facilitating effective communication between emergency users 
during times like 9/11. 

Yes studying Robustness of Internet is interesting..e.g.
     - How can we make BGP robust to all code-red/misconfigurations
     - How can we make DNS more robust to DoS attacks
     - How can we make TCP robust to packet re-ordering
     - Any enhancements to RED for robustness (RED-nounce )
     -  How can i use triggers to make robust transport communication
    - What are the features to track DoS 
    - any architectural considerations for making robust systems
    - Robust coding issues (making all protocol stacks to be resistent to
      DoS )
   etc etc.
But all the above issues are totally out of scope. But, sure making
various components of Internet structure  more robust is _one_ of 
many solutionns to support emergency users.  Hence, I accept with 
kimberly that we have to write problem statement/emergency user 
requirements before diving into that part.
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 08:04:47 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA07539
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 08:04:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9D7C332007
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 08:07:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9D7Cv32004
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 08:07:12 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA07502
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 08:04:15 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9D55v31564;
	Mon, 9 Dec 2002 08:05:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9D50v31535
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 08:05:00 -0500
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA07040;
	Mon, 9 Dec 2002 08:02:03 -0500 (EST)
Message-Id: <200212091302.IAA07040@ietf.org>
Mime-Version: 1.0
Content-Type: Multipart/Mixed; Boundary="NextPart"
To: IETF-Announce: ;
Cc: ieprep@ietf.org
From: Internet-Drafts@ietf.org
Reply-to: Internet-Drafts@ietf.org
Date: Mon, 09 Dec 2002 08:02:03 -0500
Subject: [Ieprep] I-D ACTION:draft-ietf-ieprep-sip-reqs-03.txt
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

--NextPart

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

	Title		: Requirements for Resource Priority Mechanisms for the 
                          Session Initiation Protocol
	Author(s)	: H. Schulzrinne
	Filename	: draft-ietf-ieprep-sip-reqs-03.txt
	Pages		: 16
	Date		: 2002-12-6
	
This document summarizes requirements for prioritizing access to
circuit-switched network, end system and proxy resources for
emergency preparedness communications using the Session Initiation
Protocol (SIP).

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-ieprep-sip-reqs-03.txt

To remove yourself from the IETF Announcement list, send a message to 
ietf-announce-request with the word unsubscribe in the body of the message.

Internet-Drafts are also available by anonymous FTP. Login with the username
"anonymous" and a password of your e-mail address. After logging in,
type "cd internet-drafts" and then
	"get draft-ietf-ieprep-sip-reqs-03.txt".

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


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

Send a message to:
	mailserv@ietf.org.
In the body type:
	"FILE /internet-drafts/draft-ietf-ieprep-sip-reqs-03.txt".
	
NOTE:	The mail server at ietf.org can return the document in
	MIME-encoded form by using the "mpack" utility.  To use this
	feature, insert the command "ENCODING mime" before the "FILE"
	command.  To decode the response(s), you will need "munpack" or
	a MIME-compliant mail reader.  Different MIME-compliant mail readers
	exhibit different behavior, especially when dealing with
	"multipart" MIME messages (i.e. documents which have been split
	up into multiple messages), so check your local documentation on
	how to manipulate these messages.
		
		
Below is the data which will enable a MIME compliant mail reader
implementation to automatically retrieve the ASCII version of the
Internet-Draft.

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

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

Content-Type: text/plain
Content-ID:	<2002-12-6140159.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-ieprep-sip-reqs-03.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-ieprep-sip-reqs-03.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<2002-12-6140159.I-D@ietf.org>

--OtherAccess--

--NextPart--


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 10:37:00 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14186
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 10:37:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9FdQh09569
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 10:39:26 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9FdPv09566
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 10:39:25 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14159
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 10:36:29 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Fa7v08784;
	Mon, 9 Dec 2002 10:36:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9FZXv08724
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 10:35:33 -0500
Received: from imo-m02.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA13927
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 10:32:37 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m02.mx.aol.com (mail_out_v34.13.) id f.140.49b58af (25508);
	Mon, 9 Dec 2002 10:34:46 -0500 (EST)
Message-ID: <140.49b58af.2b261216@aol.com>
Date: Mon, 9 Dec 2002 10:34:46 EST
Subject: Re: [Ieprep] identification of packets
To: saq66@umkc.edu, dhc@dcrocker.net, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_140.49b58af.2b261216_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_140.49b58af.2b261216_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/8/2002 8:58:30 PM Eastern Standard Time, saq66@umkc.edu 
writes:

> 
> third, it is important for IETF to hear from other communities 
> too (here, emergency response community.) At last, they are the 
> 


[MAP] There have been many comments made by two major potential users of this 
capability - NCS and DoD in the US - but their comments have been routinely 
slammed.

> 
> So, why cant we start a separate interest list (like diffserv
> interest list.... If your in Diffserv list, How many mail of yours 
> have been bypassed to diffserv-interest list by Brian? ) to talk about 
> all this other issues and leave this list for requirements 
> discussion. 
> 


[MAP] I can't afford to be on too many more lists. This one needs to come to 
decisions/agreements on requirements, then we can participate in "solutions" 
discussions (maybe in DiffServ).

> Also, I suggest we keep some separate threads while 
> discussing each issue (like the same way
> SIP requirements was discussed.) Else, it will boil 
> down to another over provisioning discussion. 
> 
> 

[MAP] Regardless of any other discussion, I would like to see an end to any 
discussion on this list about "over-provisioning". We all agree a carrier can 
do it if they want, but it's not the solution to the problem.

Mike Pierc
Artel


--part1_140.49b58af.2b261216_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/8/2002 8:58:30 PM Eastern Standard Time, saq66@umkc.edu writes:
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>third, it is important for IETF to hear from other communities 
<BR>too (here, emergency response community.) At last, they are the 
<BR>once who are going to pay you(ISPs)for routing their packets. </BLOCKQUOTE>
<BR>
<BR>
<BR>[MAP] There have been many comments made by two major potential users of this capability - NCS and DoD in the US - but their comments have been routinely slammed.
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
<BR>So, why cant we start a separate interest list (like diffserv
<BR>interest list.... If your in Diffserv list, How many mail of yours 
<BR>have been bypassed to diffserv-interest list by Brian? ) to talk about 
<BR>all this other issues and leave this list for requirements 
<BR>discussion. 
<BR></BLOCKQUOTE>
<BR>
<BR>
<BR>[MAP] I can't afford to be on too many more lists. This one needs to come to decisions/agreements on requirements, then we can participate in "solutions" discussions (maybe in DiffServ).
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">Also, I suggest we keep some separate threads while 
<BR>discussing each issue (like the same way
<BR>SIP requirements was discussed.) Else, it will boil 
<BR>down to another over provisioning discussion. 
<BR>
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>[MAP] Regardless of any other discussion, I would like to see an end to any discussion on this list about "over-provisioning". We all agree a carrier can do it if they want, but it's not the solution to the problem.
<BR>
<BR>Mike Pierc
<BR>Artel
<BR></FONT></HTML>

--part1_140.49b58af.2b261216_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 10:50:23 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14596
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 10:50:23 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9FqnU10201
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 10:52:49 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Fqmv10198
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 10:52:48 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14581
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 10:49:52 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Fm1v10000;
	Mon, 9 Dec 2002 10:48:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Fljv09976
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 10:47:45 -0500
Received: from imo-r05.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA14452
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 10:44:49 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r05.mx.aol.com (mail_out_v34.13.) id a.e6.32b3398d (25508);
	Mon, 9 Dec 2002 10:46:40 -0500 (EST)
Message-ID: <e6.32b3398d.2b2614e0@aol.com>
Date: Mon, 9 Dec 2002 10:46:40 EST
Subject: Re: [Ieprep] identification of problem
To: KIMBERLY.S.KING@saic.com, saq66@umkc.edu, dhc@dcrocker.net,
        ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_e6.32b3398d.2b2614e0_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_e6.32b3398d.2b2614e0_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/9/2002 6:20:04 AM Eastern Standard Time, 
KIMBERLY.S.KING@saic.com writes:


> At the crux of the matter, the Internet is not centrally controlled, is
> extremely survivable, and was specifically engineered to avoid congestion
> collapse so *if* the problem is to avoid congestion collapse in an emergency
> (as defined by the way the PSTN collapses in an emergency) then the answer
> probably is that the Internet already does that. 


[MAP] I may have missed something. Can someone give an example of when and 
where "the PSTN collapsed" during an emergency? I don't consider local 
congestion or inability for some customers to get dial tone as "collapsed".

I also note that the report cited states two things in its summary:

"perfomance [of the Internet on Sept 11] does not necesarily indicate how it 
might respond to being directly targeted"
"their [ISPs] facilities were not concentrated at 140 West Street"

Mike Pierce
Artel


--part1_e6.32b3398d.2b2614e0_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/9/2002 6:20:04 AM Eastern Standard Time, KIMBERLY.S.KING@saic.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">At the crux of the matter, the Internet is not centrally controlled, is
<BR>extremely survivable, and was specifically engineered to avoid congestion
<BR>collapse so *if* the problem is to avoid congestion collapse in an emergency
<BR>(as defined by the way the PSTN collapses in an emergency) then the answer
<BR>probably is that the Internet already does that. </FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>[MAP] I may have missed something. Can someone give an example of when and where "the PSTN collapsed" during an emergency? I don't consider local congestion or inability for some customers to get dial tone as "collapsed".
<BR>
<BR>I also note that the report cited states two things in its summary:
<BR>
<BR>"perfomance [of the Internet on Sept 11] does not necesarily indicate how it might respond to being directly targeted"
<BR>"their [ISPs] facilities were not concentrated at 140 West Street"
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_e6.32b3398d.2b2614e0_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 13:56:57 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24926
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 13:56:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9IxPs23449
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 13:59:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9IxPv23446
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 13:59:25 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24914
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 13:56:26 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Iw2v23375;
	Mon, 9 Dec 2002 13:58:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Iv6v23340
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 13:57:06 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA24716
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 13:54:07 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 09C3D67114; Mon,  9 Dec 2002 10:23:42 -0500 (EST)
Date: Mon, 9 Dec 2002 13:57:00 -0500
Subject: Re: [Ieprep] commercial SIP service
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>,
        "'Henry.Sinnreich@wcom.com'" <Henry.Sinnreich@wcom.com>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A0F@mcl-its-exs01.mail.saic.com>
Message-Id: <FF57504E-0BA7-11D7-8878-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Dec 6, 2002, at 16:12 America/Montreal, King, Kimberly S. 
wrote:

> Worldcom offers IP Communications, ...

Which part of Worldcom is offering this particular service/product ?
There are many parts of Worldcom, many of which are operated separately
from each other.

For example, is this an MCI service or a UUnet service or an MFS service
or some other construct ?

Does it use the UUnet backbone or a specially-constructed backbone of
some other operating unit of Worldcom ?

Thanks,

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 14:05:55 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25466
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 14:05:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9J8Np24542
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 14:08:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9J8Nv24539
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 14:08:23 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25422
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 14:05:23 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9J70v23837;
	Mon, 9 Dec 2002 14:07:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9J6Uv23754
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 14:06:30 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25246
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 14:03:30 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 9 Dec 2002 14:05:47 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120914060730846
 ; Mon, 09 Dec 2002 14:06:07 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPH5501>; Mon, 9 Dec 2002 14:07:39 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A16@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>,
        "'Henry.Sinnreich@wcom.com'" <Henry.Sinnreich@wcom.com>
Subject: RE: [Ieprep] commercial SIP service
Date: Mon, 9 Dec 2002 14:05:57 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Ran,
 
> On Friday, Dec 6, 2002, at 16:12 America/Montreal, King, Kimberly S. 
> wrote:
> 
> > Worldcom offers IP Communications, ...
> 
> Which part of Worldcom is offering this particular service/product ?
> There are many parts of Worldcom, many of which are operated 
> separately
> from each other.

Henry is out of the country this week but perhaps this link offers more
information.  

http://www.worldcom.com/us/products/connection/
 
> For example, is this an MCI service or a UUnet service or an 
> MFS service
> or some other construct ?
> 
> Does it use the UUnet backbone or a specially-constructed backbone of
> some other operating unit of Worldcom ?

I read somewhere this isn't the UUnet backbone but rather another IP
network.  I'll send you some contacts offline too.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 14:05:56 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25479
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 14:05:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9J8OP24557
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 14:08:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9J8Ov24554
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 14:08:24 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25426
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 14:05:24 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9J73v23879;
	Mon, 9 Dec 2002 14:07:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9J6Wv23764
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 14:06:32 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25253
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 14:03:32 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 8D9EA67103; Mon,  9 Dec 2002 10:33:07 -0500 (EST)
Date: Mon, 9 Dec 2002 14:06:25 -0500
Subject: Re: [Ieprep] identification of packets
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep '" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A10@mcl-its-exs01.mail.saic.com>
Message-Id: <507AD5F6-0BA9-11D7-8878-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Saturday, Dec 7, 2002, at 08:06 America/Montreal, King, Kimberly S. 
wrote:
> I completely agree with Rohan and it is also useful to keep in mind 
> that
> while the kind of QoS related solutions would be difficult enough 
> between
> two sites, it gets wildly out of control if you wanted to ensure that 
> kind
> of situation between the top 100 sites.  For each site to maintain 
> special
> information between itself and 99 other sites is just asking for major
> disaster (no pun intended).

It might not be so crazy -- or at least USG doesn't think it is.
Part of the whole idea behind GOVnet was to create a large private
network that would be dedicated to intra-government connectivity.

Designing such a large private network to provide adequate capacity
such that QoS mechanisms need not be deployed would simply be a matter
of using current commercial best practices, nothing magical.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 14:14:59 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA26045
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 14:14:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9JHRC25049
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 14:17:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9JHRv25046
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 14:17:27 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA26033
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 14:14:28 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9JG3v24963;
	Mon, 9 Dec 2002 14:16:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9JFRv24923
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 14:15:27 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25852
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 14:12:28 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 0CDF267103; Mon,  9 Dec 2002 10:42:04 -0500 (EST)
Date: Mon, 9 Dec 2002 14:15:22 -0500
Subject: Re: [Ieprep] identification of packets
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <140.49b58af.2b261216@aol.com>
Message-Id: <9026E43C-0BAA-11D7-8878-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 9, 2002, at 10:34 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] There have been many comments made by two major potential users 
> of this capability - NCS and DoD in the US - but their comments have 
> been routinely slammed.

When anyone makes a silly comment, the comment is likely to get slammed.
(It happens to me all the time and is just part of life.)  When folks 
make
non-silly comments or ask reasonable questions, those generally get 
taken
seriously (e.g. Janet Gunn's recent scenario discussion).

> So, why cant we start a separate interest list (like diffserv
> interest list.... If your in Diffserv list, How many mail of yours
> have been bypassed to diffserv-interest list by Brian? ) to talk about
> all this other issues and leave this list for requirements
> discussion.

The point of the comment was that it is silly to try to talk about
"all these other issues" when the problem definition is not generally
agreed to yet (and clearly it is not generally agreed to as of now).

> Also, I suggest we keep some separate threads while
> discussing each issue (like the same way
> SIP requirements was discussed.)

I keep trying to edit Subject: lines to reflect the current
topics.  It would not hurt for others to do the same when
the topic changes.

> [MAP] Regardless of any other discussion, I would like to see an end
> to any discussion on this list about "over-provisioning". We all agree
> a carrier can do it if they want, but it's not the solution to the 
> problem.

It is the best solution in town, for a wide range of situations, so it 
is
pure fantasy to think it won't keep being brought up here.  In fact, 
over
provisioning seems to be the current DISN strategy for at least the 
CONUS IP
networks [1] (and I wish them every success in getting it deployed for 
their
networks).


Ran

[1] There appears to be an RFI out from DISA regarding availability of 
dark fibre
in CONUS.  (I think it is a safe bet DISA aren't planning to use the 
fibre
for long-haul cable television links. :-)

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 14:15:58 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA26096
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 14:15:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9JIQw25120
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 14:18:26 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9JIQv25117
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 14:18:26 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA26082
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 14:15:27 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9JH0v25000;
	Mon, 9 Dec 2002 14:17:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9JGOv24977
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 14:16:24 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA25952
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 14:13:24 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 9 Dec 2002 14:15:46 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120914160625943
 ; Mon, 09 Dec 2002 14:16:06 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPJHXRD>; Mon, 9 Dec 2002 14:15:52 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A18@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] identification of packets
Date: Mon, 9 Dec 2002 14:16:04 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


> On Saturday, Dec 7, 2002, at 08:06 America/Montreal, King, 
> Kimberly S. 
> For each site to maintain 
> > special
> > information between itself and 99 other sites is just 
> asking for major
> > disaster (no pun intended).

Ran said:
> It might not be so crazy -- or at least USG doesn't think it is.
> Part of the whole idea behind GOVnet was to create a large private
> network that would be dedicated to intra-government connectivity.

Creating a large priviate network dedicated to intra-government connectivity
is one thing.  Using special filters at each router configured and
maintained for each of the other sites to support QoS over the Internet is
another.

> Designing such a large private network to provide adequate capacity
> such that QoS mechanisms need not be deployed would simply be a matter
> of using current commercial best practices, nothing magical.

That sounds reasonable.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 14:18:01 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA26306
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 14:18:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9JKUg25264
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 14:20:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9JKUv25261
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 14:20:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA26242
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 14:17:30 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9JJEv25176;
	Mon, 9 Dec 2002 14:19:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9JIev25142
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 14:18:40 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA26089
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 14:15:41 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 77E9D67103; Mon,  9 Dec 2002 10:45:16 -0500 (EST)
Date: Mon, 9 Dec 2002 14:18:34 -0500
Subject: Re: [Ieprep] identification of problem
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <e6.32b3398d.2b2614e0@aol.com>
Message-Id: <02E14361-0BAB-11D7-8878-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 9, 2002, at 10:46 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] I may have missed something. Can someone give an example of when 
> and where "the PSTN collapsed" during an emergency? I don't consider 
> local congestion or inability for some customers to get dial tone as 
> "collapsed".

Local collapse would certainly qualify as "collapsed" in my own book.
Clearly your milage varies.

> I also note that the report cited states two things in its summary:
>
> "perfomance [of the Internet on Sept 11] does not necesarily indicate 
> how it might respond to being directly targeted"
> "their [ISPs] facilities were not concentrated at 140 West Street"

A number of ISPs were at locations in lower Manhatten that suffered
fibre cuts, lack of power, lack of cooling, and other serious problems
on 9/11.  However, the Internet as a whole stayed up.  140 West Street
is not the only address that suffered serious problems on 9/11/02
-- as you ought to know if you didn't.

And you seem to think that "local congestion" doesn't qualify as 
collapsed
for the PSTN (which I disagree with), but why do you have such a 
blatent double
standard that it does qualify as collapsed in the IP network case ?

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 15:12:06 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA28970
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 15:12:05 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9KEZE29190
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 15:14:35 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9KEYv29187
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 15:14:34 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA28954
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 15:11:34 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9KD7v29153;
	Mon, 9 Dec 2002 15:13:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9KCsv29123
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 15:12:54 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA28873
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 15:09:54 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB9K5Ao28283;
	Mon, 9 Dec 2002 12:05:11 -0800 (PST)
Date: Mon, 9 Dec 2002 12:05:10 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: Mpierce1@aol.com
cc: KIMBERLY.S.KING@saic.com, <saq66@umkc.edu>, <dhc@dcrocker.net>,
        <ieprep@ietf.org>
Subject: Re: [Ieprep] identification of problem
In-Reply-To: <e6.32b3398d.2b2614e0@aol.com>
Message-ID: <Pine.GSO.4.44.0212091158490.28222-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Mon, 9 Dec 2002 Mpierce1@aol.com wrote:
    > I don't consider inability to get dial tone as "collapsed".

Uh, what _would_ you consider a failure of the PSTN, then?  And if that
doesn't constitute failure of the PSTN, why would it constitute failure of
the Internet?  Sure looks like failure to me.

    > Can someone give an example of when and where "the PSTN collapsed"
    > during an emergency?

Just since I don't remember anyone having cited it as an example yet, Loma
Prieta.  My parents heard news reports that the bay bridge had collapsed
and that San Francisco had burned to the ground, and neither they
(inbound) nor I (outbound) were able to get any calls through for 36+
hours to contradict such sillyness.  The Internet, as usual, worked
fine; unfortunately they didn't have email yet at that time.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 15:15:00 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA29140
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 15:14:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9KHTI29364
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 15:17:29 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9KHTv29361
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 15:17:29 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA29114
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 15:14:28 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9KG2v29272;
	Mon, 9 Dec 2002 15:16:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9KFqv29253
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 15:15:52 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA29009
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 15:12:52 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gB9K8Bo28369;
	Mon, 9 Dec 2002 12:08:11 -0800 (PST)
Date: Mon, 9 Dec 2002 12:08:11 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: RJ Atkinson <rja@extremenetworks.com>
cc: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>,
        "'ieprep '" <ieprep@ietf.org>
Subject: Re: [Ieprep] identification of packets
In-Reply-To: <507AD5F6-0BA9-11D7-8878-00039357A82A@extremenetworks.com>
Message-ID: <Pine.GSO.4.44.0212091206450.28222-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Mon, 9 Dec 2002, RJ Atkinson wrote:
    > It might not be so crazy -- or at least USG doesn't think it is.
    > Part of the whole idea behind GOVnet was to create a large private
    > network that would be dedicated to intra-government connectivity.

Yeah, that's true...  I recently saw a grant offer from the USG to bring
any NGOs which didn't already have >=DS3 up to speed, and that's got to be
targeting thousands of organizations.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 15:38:06 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00241
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 15:38:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9KeZK30843
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 15:40:35 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9KeZv30840
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 15:40:35 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA00237
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 15:37:35 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Kd5v30718;
	Mon, 9 Dec 2002 15:39:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9KcAv30667
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 15:38:10 -0500
Received: from [140.32.132.66] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA29996
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 15:35:09 -0500 (EST)
Received: from mail.nps.navy.mil by [140.32.132.66]
          via smtpd (for [132.151.6.1]) with ESMTP; Mon, 9 Dec 2002 12:33:34 -0800
Received: from ziggy ([131.120.179.220])
	by capella.nps.navy.mil (8.12.2/8.12.2) with ESMTP id gB9KY2bR006171;
	Mon, 9 Dec 2002 12:34:02 -0800 (PST)
Subject: Re: [Ieprep] identification of problem
From: Rex Buddenberg <budden@nps.navy.mil>
To: Bill Woodcock <woody@pch.net>
Cc: Mpierce1@aol.com, KIMBERLY.S.KING@saic.com, saq66@umkc.edu,
        dhc@dcrocker.net, ieprep@ietf.org
In-Reply-To: <Pine.GSO.4.44.0212091158490.28222-100000@paixhost.pch.net>
References: <Pine.GSO.4.44.0212091158490.28222-100000@paixhost.pch.net>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) 
Date: 09 Dec 2002 12:29:49 -0800
Message-Id: <1039465789.2748.52.camel@ziggy>
Mime-Version: 1.0
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

On Mon, 2002-12-09 at 12:05, Bill Woodcock wrote:

> MAP:
>     > Can someone give an example of when and where "the PSTN collapsed"
>     > during an emergency?
> 

Hinsdale, Ill fire that burned the central office ... Didn't require the
emergency to collapse there.

Kobe, Japan earthquate (see previous post)

Richard Nixon's resignation speech.



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 15:58:57 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01537
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 15:58:57 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9L1RI31553
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 16:01:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9L1Rv31550
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 16:01:27 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01525
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 15:58:26 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9L02v31506;
	Mon, 9 Dec 2002 16:00:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9KxBv31438
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 15:59:11 -0500
Received: from kc-msxproto2.kc.umkc.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01382
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 15:56:11 -0500 (EST)
Received: from KC-MAIL2.kc.umkc.edu ([134.193.143.162] RDNS failed) by kc-msxproto2.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.5329);
	 Mon, 9 Dec 2002 14:59:05 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Date: Mon, 9 Dec 2002 14:59:04 -0600
Message-ID: <A29143A40AEAE3459FF829D1DD43316101D9B07B@KC-MAIL2.kc.umkc.edu>
Thread-Topic: [Ieprep] identification of problem
Thread-Index: AcKfv1TVEuQJtOzdRLS3mRtZYUrZ5gAAGivw
From: "Ayyasamy, Senthilkumar  (UMKC-Student)" <saq66@umkc.edu>
To: "Bill Woodcock" <woody@pch.net>, <Mpierce1@aol.com>
Cc: <KIMBERLY.S.KING@saic.com>, <dhc@dcrocker.net>, <ieprep@ietf.org>
X-OriginalArrivalTime: 09 Dec 2002 20:59:05.0472 (UTC) FILETIME=[CF473800:01C29FC5]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gB9KxBv31439
Subject: [Ieprep] PSTN collapse
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


>       On Mon, 9 Dec 2002 Mpierce1@aol.com wrote:
>     > I don't consider inability to get dial tone as "collapsed".

No Dial tone = No connection (PSTN) 
Welcome to the world of SIP, where you get dial tone but still
remain disconnected. This is due to the interesting concept of 
separation of session establishment and resource reservation in
SIP based systems. If for some reason the resource reservation 
fails, the call will be set up even though network doesnt support.
So, your phone may be ringing even though the network cannot
support the call. It was one of the problems discussed during
initial days of SIP. But, there are many proposed fixes to avoid 
the above situation. 


>     > Can someone give an example of when and where "the PSTN 
> collapsed"
>     > during an emergency?

Please go through all the trade press articles about 9/11.
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 16:50:18 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA04167
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 16:50:18 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9Lqkh02318
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 16:52:46 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Lqkv02315
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 16:52:46 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA04136
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 16:49:47 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9LoUv02206;
	Mon, 9 Dec 2002 16:50:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Ln3v02128
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 16:49:03 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA03922
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 16:46:04 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 9 Dec 2002 16:48:24 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002120916484413747
 ; Mon, 09 Dec 2002 16:48:44 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPJ2J1R>; Mon, 9 Dec 2002 16:48:30 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A19@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Ayyasamy, Senthilkumar  (UMKC-Student) '" <saq66@umkc.edu>,
        "'Bill Woodcock '" <woody@pch.net>,
        "'Mpierce1@aol.com '" <Mpierce1@aol.com>
Cc: "'KIMBERLY.S.KING@saic.com '" <KIMBERLY.S.KING@saic.com>,
        "'dhc@dcrocker.net '" <dhc@dcrocker.net>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
Date: Mon, 9 Dec 2002 16:48:40 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] SIP and Resource Reservation
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


Ayyasamy, Senthilkumar said:

>No Dial tone = No connection (PSTN) 
>Welcome to the world of SIP, where you get dial tone but still
>remain disconnected. This is due to the interesting concept of 
>separation of session establishment and resource reservation in
>SIP based systems. If for some reason the resource reservation 
>fails, the call will be set up even though network doesnt support.

Resource reservation is not typically supported for SIP systems running over
the Internet.  Instead, there is a graceful degradation for all callers over
a congested link rather than a complete inability to achieve a connection.
As we have talked about (and I believe agreed to!), congested links
typically occur in access links and congestion is not much of an issue in
core U.S. ISPs.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 17:08:07 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA04961
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 17:08:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9MAas04031
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 17:10:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9MAav04028
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 17:10:36 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA04934
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 17:07:36 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9M94v03993;
	Mon, 9 Dec 2002 17:09:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9M8Bv03967
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 17:08:11 -0500
Received: from kc-msxproto4.kc.umkc.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA04852
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 17:05:11 -0500 (EST)
Received: from KC-MAIL2.kc.umkc.edu ([134.193.143.162] RDNS failed) by kc-msxproto4.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.5329);
	 Mon, 9 Dec 2002 16:08:04 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Date: Mon, 9 Dec 2002 16:08:03 -0600
Message-ID: <A29143A40AEAE3459FF829D1DD43316101D9B07C@KC-MAIL2.kc.umkc.edu>
Thread-Topic: SIP and Resource Reservation
Thread-Index: AcKfzMZt4/7LWaZcS+Wt3o2IXT3mkAAAKqMA
From: "Ayyasamy, Senthilkumar  (UMKC-Student)" <saq66@umkc.edu>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>,
        "Bill Woodcock " <woody@pch.net>, <Mpierce1@aol.com>
Cc: <dhc@dcrocker.net>, <ieprep@ietf.org>
X-OriginalArrivalTime: 09 Dec 2002 22:08:04.0194 (UTC) FILETIME=[72261420:01C29FCF]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gB9M8Bv03968
Subject: [Ieprep] RE: SIP and Resource Reservation
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


> >No Dial tone = No connection (PSTN) 
> >Welcome to the world of SIP, where you get dial tone but still
> >remain disconnected. This is due to the interesting concept of 
> >separation of session establishment and resource reservation in
> >SIP based systems. If for some reason the resource reservation 
> >fails, the call will be set up even though network doesnt support.
> 
> Resource reservation is not typically supported for SIP 
> systems running over  the Internet. Instead, there is a 
> graceful degradation for  all callers over  a congested link rather 
> than a complete inability to achieve  a connection.
> As we have talked about (and I believe agreed to!), congested links
> typically occur in access links and congestion is not much of 
> an issue in  core U.S. ISPs.

I completely agree. I used that as an example to explain that their 
may be situations where dial tone doesn't always mean connection is
ok ( as in PSTN.) Also, the assumption being using RSVP for resource 
reservation. Try reading RFC 3312. 

( Note: I did not consider any deployment feasibility for this 
example.) 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 17:17:08 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05315
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 17:17:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9MJbo04341
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 17:19:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9MJbv04338
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 17:19:37 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05248
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 17:16:37 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9MI4v04278;
	Mon, 9 Dec 2002 17:18:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9MHxv04253
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 17:17:59 -0500
Received: from imo-m03.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05211
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 17:14:59 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m03.mx.aol.com (mail_out_v34.13.) id l.6.41ae3d5 (657)
	 for <ieprep@ietf.org>; Mon, 9 Dec 2002 17:17:47 -0500 (EST)
Message-ID: <6.41ae3d5.2b26708b@aol.com>
Date: Mon, 9 Dec 2002 17:17:47 EST
To: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_6.41ae3d5.2b26708b_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Subject: [Ieprep] Problem?
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_6.41ae3d5.2b26708b_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

To all,

The following quote, taken from a recent e-mail, certainly shows why we are 
having such difficulty on this list in getting anywhere.

"If we know _in advance_ that the emergency traffic is going to be between
A and B, ..."

There seem to be some (maybe only one?) who still do not grasp the problem - 
emergencies are not preplanned.

Mike Pierce
Artel


--part1_6.41ae3d5.2b26708b_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>To all,
<BR>
<BR>The following quote, taken from a recent e-mail, certainly shows why we are having such difficulty on this list in getting anywhere.
<BR>
<BR>"If we know _in advance_ that the emergency traffic is going to be between
<BR>A and B, ..."
<BR>
<BR>There seem to be some (maybe only one?) who still do not grasp the problem - emergencies are not preplanned.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_6.41ae3d5.2b26708b_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 17:19:12 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05453
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 17:19:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9MLfo04494
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 17:21:41 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9MLdv04491
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 17:21:39 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05426
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 17:18:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9MK2v04411;
	Mon, 9 Dec 2002 17:20:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9MJjv04353
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 17:19:45 -0500
Received: from imo-r03.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05268
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 17:16:45 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r03.mx.aol.com (mail_out_v34.13.) id f.18f.128c5bd4 (657);
	Mon, 9 Dec 2002 17:17:39 -0500 (EST)
Message-ID: <18f.128c5bd4.2b267082@aol.com>
Date: Mon, 9 Dec 2002 17:17:38 EST
To: saq66@umkc.edu, woody@pch.net
CC: KIMBERLY.S.KING@saic.com, dhc@dcrocker.net, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_18f.128c5bd4.2b267082_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Subject: [Ieprep] Re: PSTN collapse
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_18f.128c5bd4.2b267082_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/9/2002 3:59:12 PM Eastern Standard Time, saq66@umkc.edu 
writes:


> >       On Mon, 9 Dec 2002 Mpierce1@aol.com wrote:
> >     > I don't consider inability to get dial tone as "collapsed".
> 
> No Dial tone = No connection (PSTN) 
> Welcome to the world of SIP, where you get dial tone but still
> remain disconnected. This is due to the interesting concept of 
> separation of session establishment and resource reservation in
> SIP based systems. If for some reason the resource reservation 
> fails, the call will be set up even though network doesnt support.
> So, your phone may be ringing even though the network cannot
> support the call. It was one of the problems discussed during
> initial days of SIP. But, there are many proposed fixes to avoid 
> the above situation. 
> 
> 
> >     > Can someone give an example of when and where "the PSTN 
> > collapsed"
> >     > during an emergency?
> 
> Please go through all the trade press articles about 9/11.
> 



As usual, I see that you can not answer a direct question regarding a wild 
claim about how bad the PSTN is. I don't have time to read all the trade 
press articles on 9/11, certainly not all the ones that tell about how 
wonderful the Internet performed. (I don't dispute that.) I'd like a pointer 
to one that describes a "collapse" of the PSTN, on 9/11 or any other time.

I was:
- in San Fransisco when the earthquake hit in 1989
- in South Florida when the big hurricane hit
- in Baltimore when a hurricane went over and we lost power for a week
- at a meeting in the area right after Hinsdale burned (we waited on the 
ground in Miami an extra 2 hours while ATC circuits were rerouted.)

(Sounds like disasters follow me.)

and in none of these cases did the PSTN "collapse". There were local outages 
depending on the condition. Sure, if a tree fell across the wires to your 
house, you didn't get dial tone. That's not a "collapse" of the PSTN!

Mike Pierce



--part1_18f.128c5bd4.2b267082_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/9/2002 3:59:12 PM Eastern Standard Time, saq66@umkc.edu writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">&gt; &nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;On Mon, 9 Dec 2002 Mpierce1@aol.com wrote:
<BR>&gt; &nbsp;&nbsp;&nbsp;&nbsp;&gt; I don't consider inability to get dial tone as "collapsed".
<BR>
<BR>No Dial tone = No connection (PSTN) 
<BR>Welcome to the world of SIP, where you get dial tone but still
<BR>remain disconnected. This is due to the interesting concept of 
<BR>separation of session establishment and resource reservation in
<BR>SIP based systems. If for some reason the resource reservation 
<BR>fails, the call will be set up even though network doesnt support.
<BR>So, your phone may be ringing even though the network cannot
<BR>support the call. It was one of the problems discussed during
<BR>initial days of SIP. But, there are many proposed fixes to avoid 
<BR>the above situation. 
<BR>
<BR>
<BR>&gt; &nbsp;&nbsp;&nbsp;&nbsp;&gt; Can someone give an example of when and where "the PSTN 
<BR>&gt; collapsed"
<BR>&gt; &nbsp;&nbsp;&nbsp;&nbsp;&gt; during an emergency?
<BR>
<BR>Please go through all the trade press articles about 9/11.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>
<BR>As usual, I see that you can not answer a direct question regarding a wild claim about how bad the PSTN is. I don't have time to read all the trade press articles on 9/11, certainly not all the ones that tell about how wonderful the Internet performed. (I don't dispute that.) I'd like a pointer to one that describes a "collapse" of the PSTN, on 9/11 or any other time.
<BR>
<BR>I was:
<BR>- in San Fransisco when the earthquake hit in 1989
<BR>- in South Florida when the big hurricane hit
<BR>- in Baltimore when a hurricane went over and we lost power for a week
<BR>- at a meeting in the area right after Hinsdale burned (we waited on the ground in Miami an extra 2 hours while ATC circuits were rerouted.)
<BR>
<BR>(Sounds like disasters follow me.)
<BR>
<BR>and in none of these cases did the PSTN "collapse". There were local outages depending on the condition. Sure, if a tree fell across the wires to your house, you didn't get dial tone. That's not a "collapse" of the PSTN!
<BR>
<BR>Mike Pierce
<BR>
<BR></FONT></HTML>

--part1_18f.128c5bd4.2b267082_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 17:25:01 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05644
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 17:25:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9MRUS04740
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 17:27:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9MRUv04737
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 17:27:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05602
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 17:24:30 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9MPXv04624;
	Mon, 9 Dec 2002 17:25:33 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9MN0v04543
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 17:23:00 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA05487
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 17:20:01 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAG7XT>; Mon, 9 Dec 2002 17:20:25 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149CC@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'King, Kimberly  S.'" <KIMBERLY.S.KING@saic.com>,
        "'ieprep@ietf.org'"
	 <ieprep@ietf.org>
Cc: "'Henry.Sinnreich@wcom.com'" <Henry.Sinnreich@wcom.com>
Subject: RE: [Ieprep] commercial SIP service
Date: Mon, 9 Dec 2002 17:20:15 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Just a clarification.

Does this mean that only the 800 based "alternate GETS numbers" would work,
or would the 710 based numbers work too?

-----Original Message-----
From: King, Kimberly S. [mailto:KIMBERLY.S.KING@saic.com]
Sent: Friday, December 06, 2002 4:12 PM
To: 'ieprep@ietf.org'
Cc: 'Henry.Sinnreich@wcom.com'
Subject: [Ieprep] commercial SIP service


Worldcom offers IP Communications, a converged voice/data service using SIP
mainly targeted at small to medium businesses.   I contacted them on behalf
of ieprep as a result of last week's discussions and Henry Sinnreich was
kind enough to provide some answers.

...

What would happen if someone dialed a GETS call today using IP Comm?
Either the "shared local gateway" in the WorldCom network, or a physical
local gateway box will connect such calls (911, 800, 411, etc) to the local
phone network.

...


- Kimberly




_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 17:55:13 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06720
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 17:55:13 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9Mvg006468
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 17:57:42 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Mvgv06465
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 17:57:42 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06709
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 17:54:41 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Mu5v06420;
	Mon, 9 Dec 2002 17:56:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Mtcv06391
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 17:55:38 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06679
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 17:52:37 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAG7ZK>; Mon, 9 Dec 2002 17:53:02 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149CD@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'Mpierce1@aol.com'" <Mpierce1@aol.com>, ieprep@ietf.org
Subject: RE: [Ieprep] Problem?
Date: Mon, 9 Dec 2002 17:52:53 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01C29FD5.B5362470"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

This message is in MIME format. Since your mail reader does not understand
this format, some or all of this message may not be legible.

------_=_NextPart_001_01C29FD5.B5362470
Content-Type: text/plain;
	charset="iso-8859-1"

Mike,
 
My point was to "back off" and restrict the problem to a really simple
example that COULD be addressed- then see how far we can expand it.
 
I agree that enhanced service between two pre-defined locations isn't
particularly valuable in addressing the overall problem- but I am trying to
explore WHERE the line between "doable/practical", and "not possible/not
practical" is.  
 
While "any to any" opportunistic priority service (like GETS) is desirable,
it is not a hard and fast "requirement". When we expanded GETS to WPS, we
gave up part of the opportunistic element- it will only work on
pre-subscribed cell phones, not any cell phone.  But that is still a big
improvement over the "no wireless priority" that existed before.
 
Opportunistic "any to any" priority service on the Internet would be highly
desirable.  But if it can't be done from a practical perspective (I am not
giving up on it, just conceding it for the time being to see what CAN be
done), even priority service between 100 pre-defined sites (or IP domains)
could definitely be an improvement over the status quo.
 
Bill referred to a government program to fund "bigger pipes" for NGOs.  That
would presumably cover the Red Cross in my example, but wouldn't cover the
"critical infrastructure" side of things (oil. gas, electric utilities,
transportation companies, drug companies, etc.)
 
We DON'T know in advance where the emergency will be.  But we probably CAN
predict in advance where AT LEAST SOME of the emergency traffic will be.

-----Original Message-----
From: Mpierce1@aol.com [mailto:Mpierce1@aol.com]
Sent: Monday, December 09, 2002 5:18 PM
To: ieprep@ietf.org
Subject: [Ieprep] Problem?


To all, 

The following quote, taken from a recent e-mail, certainly shows why we are
having such difficulty on this list in getting anywhere. 

"If we know _in advance_ that the emergency traffic is going to be between 
A and B, ..." 

There seem to be some (maybe only one?) who still do not grasp the problem -
emergencies are not preplanned. 

Mike Pierce 
Artel 



------_=_NextPart_001_01C29FD5.B5362470
Content-Type: text/html;
	charset="iso-8859-1"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=iso-8859-1">


<META content="MSHTML 5.50.4134.600" name=GENERATOR></HEAD>
<BODY>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff 
size=2>Mike,</FONT></SPAN></DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff size=2>My 
point was to "back off" and restrict the problem to a really simple example that 
COULD be addressed- then see how far we can expand it.</FONT></SPAN></DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff size=2>I 
agree that enhanced service between two pre-defined locations isn't particularly 
valuable in addressing the overall problem- but I am trying to explore WHERE the 
line between "doable/practical", and "not possible/not practical" is.&nbsp; 
</FONT></SPAN></DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff size=2>While 
"any to any" opportunistic priority service (like GETS) is desirable, it is not 
a hard and fast "requirement". When we expanded GETS to WPS, we gave up part of 
the opportunistic element- it will only work on pre-subscribed cell phones, not 
any cell phone.&nbsp; But that is still a big improvement over the "no wireless 
priority" that existed before.</FONT></SPAN></DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff 
size=2>Opportunistic "any to any" priority service on the Internet would be 
highly desirable.&nbsp; But if it can't be done from a practical perspective (I 
am not giving up on it, just conceding it for the time being to see what CAN be 
done), even priority service between 100 pre-defined sites (or IP domains) could 
definitely be an improvement over the status quo.</FONT></SPAN></DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff size=2>Bill 
referred to a government program to fund "bigger pipes" for NGOs.&nbsp; That 
would presumably cover the Red Cross in my example, but wouldn't cover the 
"critical infrastructure" side of things (oil. gas, electric utilities, 
transportation companies, drug companies, etc.)</FONT></SPAN></DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff 
size=2></FONT></SPAN>&nbsp;</DIV>
<DIV><SPAN class=130142822-09122002><FONT face=Arial color=#0000ff size=2>We 
DON'T know in advance where the emergency will be.&nbsp; But we probably CAN 
predict in advance where AT LEAST SOME of the emergency traffic will 
be.</FONT></SPAN></DIV>
<BLOCKQUOTE>
  <DIV class=OutlookMessageHeader dir=ltr align=left><FONT face=Tahoma 
  size=2>-----Original Message-----<BR><B>From:</B> Mpierce1@aol.com 
  [mailto:Mpierce1@aol.com]<BR><B>Sent:</B> Monday, December 09, 2002 5:18 
  PM<BR><B>To:</B> ieprep@ietf.org<BR><B>Subject:</B> [Ieprep] 
  Problem?<BR><BR></FONT></DIV><FONT face=arial,helvetica><FONT size=2>To all, 
  <BR><BR>The following quote, taken from a recent e-mail, certainly shows why 
  we are having such difficulty on this list in getting anywhere. <BR><BR>"If we 
  know _in advance_ that the emergency traffic is going to be between <BR>A and 
  B, ..." <BR><BR>There seem to be some (maybe only one?) who still do not grasp 
  the problem - emergencies are not preplanned. <BR><BR>Mike Pierce <BR>Artel 
  <BR></BLOCKQUOTE></FONT></FONT></BODY></HTML>

------_=_NextPart_001_01C29FD5.B5362470--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 17:57:09 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06850
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 17:57:09 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9MxcJ06561
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 17:59:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Mxcv06558
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 17:59:38 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06800
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 17:56:38 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Mw2v06500;
	Mon, 9 Dec 2002 17:58:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9MvVv06453
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 17:57:31 -0500
Received: from kc-msxproto4.kc.umkc.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06706
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 17:54:31 -0500 (EST)
Received: from KC-MAIL2.kc.umkc.edu ([134.193.143.162] RDNS failed) by kc-msxproto4.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.5329);
	 Mon, 9 Dec 2002 16:57:24 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Date: Mon, 9 Dec 2002 16:57:23 -0600
Message-ID: <A29143A40AEAE3459FF829D1DD43316101D9B07D@KC-MAIL2.kc.umkc.edu>
Thread-Topic: PSTN collapse
Thread-Index: AcKf0NQFBYWUz0o6TSarNpmCEyzOzgAABn/Q
From: "Ayyasamy, Senthilkumar  (UMKC-Student)" <saq66@umkc.edu>
To: <Mpierce1@aol.com>, <woody@pch.net>
Cc: <KIMBERLY.S.KING@saic.com>, <dhc@dcrocker.net>, <ieprep@ietf.org>
X-OriginalArrivalTime: 09 Dec 2002 22:57:24.0358 (UTC) FILETIME=[568B4260:01C29FD6]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gB9MvVv06454
Subject: [Ieprep] RE: PSTN collapse
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit




>>> Can someone give an example of when and where "the PSTN 
>>> collapsed" during an emergency? 

>> Please go through all the trade press articles about 9/11. 

>  As usual, I see that you can not answer a direct question 
>  regarding a wild claim about how bad the PSTN is.

I am aware of only RFC 896, which talks about congestive collapse
and how it leads to "Internet meltdown". 

Please provide reference to PSTN collapse (since i dont have any
access to ITU-T documents.) If local meltdown is not collapse,
then what is your definition of PSTN collapse? 


The _only_ good part of PSTN is to determine malicious calls.
call trace and tapping issue, which i think(?), is not dealt in
IP telephony.



PS: Mike, can you try not to sent htmlized mails.
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 18:00:11 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07075
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 18:00:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9N2ee06730
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 18:02:40 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9N2ev06727
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 18:02:40 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA07057
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 17:59:40 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9N12v06658;
	Mon, 9 Dec 2002 18:01:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9N0Mv06618
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 18:00:22 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA06899
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 17:57:22 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAG7ZS>; Mon, 9 Dec 2002 17:57:47 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149CE@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'Dave Crocker'" <dhc@dcrocker.net>, ieprep@ietf.org
Subject: RE: [Ieprep] Janet's example
Date: Mon, 9 Dec 2002 17:57:38 -0500 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

I quite agree.  I wanted to back off to something simple that COULD be done,
and then ask about scaling/expanding.

So then the next question is: "What WOULD scale to address, say, 100
sites/organizations/IP domains?"  Would  defining some sort of VPN
(recognizing that many different kinds of "beast" use the name "VPN") be a
viable approach?  If not, is there any approach that WOULD be scalable (in
addition to, not necessarily instead of "buying bigger access pipes").

-----Original Message-----
From: Dave Crocker [mailto:dhc@dcrocker.net]

And for that matter, let's consider that it won't be just FEMA and Red
Cross.  It will be a mass of local, provincial, and national organizations
(and even inter-national organizations when entire regions are affected.)
Under these circumstances, a pair-wise solution does not scale to handle
everyone.
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 18:25:41 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07869
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 18:25:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9NSBx08157
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 18:28:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NSBv08154
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 18:28:11 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07853
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 18:25:10 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NP2v08032;
	Mon, 9 Dec 2002 18:25:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NOtv08010
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 18:24:55 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07755
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 18:21:54 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.105])
	by gnat.inet.org (Postfix) with ESMTP
	id 4BB4067116; Mon,  9 Dec 2002 14:51:29 -0500 (EST)
Date: Mon, 9 Dec 2002 18:24:42 -0500
Subject: Re: [Ieprep] Problem?
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep@ietf.org
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <6.41ae3d5.2b26708b@aol.com>
Message-Id: <6547648A-0BCD-11D7-92FE-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


Mike,

	Actually that quote doesn't illustrate what you claim.  Instead,
it was part of a constructive response to a scenario outlined by Janet
Gunn, where part of the scenario specifically involved two organisations
that do know in advance they would want to be connected to handle
emergency traffic.

	You consistently talk about IEprep only for unplanned scenarios,
but other folks quite legitimately talk about scenarios where it is
quite predictable in advance that a pair (or set) of organisations will
want/need to be able to communicate for emergency purposes.

	You don't have any statutory monopoly on the definition of what
constitutes legitimate scenarios.  Anyone participating here can propose
any scenario.  The WG consensus determines how we proceed.

	So your note mainly illustrates how out of step you are as compared
with most of the WG...

Ran

On Monday, Dec 9, 2002, at 17:17 America/Montreal, Mpierce1@aol.com 
wrote:

> To all,
>
> The following quote, taken from a recent e-mail, certainly shows why 
> we are having such difficulty on this list in getting anywhere.
>
> "If we know _in advance_ that the emergency traffic is going to be 
> between
> A and B, ..."
>
> There seem to be some (maybe only one?) who still do not grasp the 
> problem -
> emergencies are not preplanned.
>
> Mike Pierce
> Artel
>

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 18:30:00 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08049
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 18:30:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9NWUJ08324
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 18:32:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NWUv08321
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 18:32:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08003
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 18:29:29 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NQ1v08077;
	Mon, 9 Dec 2002 18:26:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NPCv08041
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 18:25:12 -0500
Received: from mace-in.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07775
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 18:22:12 -0500 (EST)
Received: from [10.32.253.236]
	by 171.70.144.186 (mace-in.cisco.com) with SMTP; 09 Dec 2002 07:26:48 +0000
Message-Id: <5.2.0.9.2.20021209145936.045ec898@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.2.0.9
Date: Mon, 09 Dec 2002 15:24:25 -0800
To: ieprep@ietf.org
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Re: PSTN collapse
In-Reply-To: <18f.128c5bd4.2b267082@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

At 05:17 PM 12/9/2002 -0500, Mpierce1@aol.com wrote:
>I'd like a pointer to one that describes a "collapse" of the PSTN, on 9/11 
>or any other time.

The PSTN did not "collapse". There were a number of switches taken out, one 
way or another, and telephone service was adversely affected as far away as 
(according to anecdotal reports) Chicago. To say that the PSTN worked 
properly on, say, Manhatten Island after 9:03 on 9/11/2001 would not be 
accurate either. I think that's what he is referring to.

By the way, we're having the same behavior in North Carolina at the moment. 
They had an ice storm last week, and while Internet connectivity and 
therefore Cisco's VoIP networks are still working just fine, the local 
telephone company seems to be having some challenges. No slight intended 
against said telephone company; there any number of last mile problems, and 
when the wire is busted, it really doesn't matter what kind of traffic 
you're trying to run over it. But their fundamental problem is with the 
architecture of the "intelligent network": it is subject to brittleness 
under failure conditions where Internet technology will automatically 
re-route.

I have an amusing email from an engineer who reports that he is working 
from home today and has call-forwarded his home phone to his desktop VoIP 
phone. He had a call this morning from the phone company, who wanted to 
know when his service had been restored. They didn't like being told that 
their dial tone was still unavailable and he was being serviced by VoIP 
from 3000 miles away. 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 18:30:08 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08062
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 18:30:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9NWcB08341
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 18:32:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NWcv08338
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 18:32:38 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08007
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 18:29:37 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NV2v08278;
	Mon, 9 Dec 2002 18:31:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NUev08233
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 18:30:40 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA07906
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 18:27:39 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.105])
	by gnat.inet.org (Postfix) with ESMTP
	id DA57667116; Mon,  9 Dec 2002 14:57:13 -0500 (EST)
Date: Mon, 9 Dec 2002 18:30:05 -0500
Subject: Re: [Ieprep] Re: PSTN collapse
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <18f.128c5bd4.2b267082@aol.com>
Message-Id: <25B6DDE8-0BCE-11D7-92FE-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 9, 2002, at 17:17 America/Montreal, Mpierce1@aol.com 
wrote:
> As usual, I see that you can not answer a direct question regarding a 
> wild claim about how bad the PSTN is. I don't have time to read all 
> the trade press articles on 9/11, certainly not all the ones that tell 
> about how wonderful the Internet performed. (I don't dispute that.) 
> I'd like a pointer to one that describes a "collapse" of the PSTN, on 
> 9/11 or any other time.
>
> I was:
> - in San Fransisco when the earthquake hit in 1989
> - in South Florida when the big hurricane hit
> - in Baltimore when a hurricane went over and we lost power for a week
> - at a meeting in the area right after Hinsdale burned (we waited on 
> the ground in Miami an extra 2 hours while ATC circuits were > rerouted.)
>
> (Sounds like disasters follow me.)
>
> and in none of these cases did the PSTN "collapse". There were local 
> outages depending on the condition. Sure, if a tree fell across the 
> wires to your house, you didn't get dial tone. That's not a "collapse" 
> of the PSTN!

Mike,

	You are seriously inconsistent with yourself.  Access links are 
entirely
comparable to "wires to your house" -- because those are common kinds 
of access
links.  Yet, you are totally worked up about IP congestion on access 
links even
though most access links only have one site (e.g. house in the case of 
DSL)
behind the link.  If local outages don't count for the PSTN, then it is 
only
fair, reasonable, and logical that they don't count against the 
Internet.

	You also selectively reply to email.  Several folks gave specific 
examples
where access wires were in place, yet dial tone did not exist for long 
periods
of time (e.g. Bill Woodcock's Loma Prieta example, among several posted 
here
recently).  You seem to have a kind of selective amnesia to facts that 
you
find inconvenient.

	Your hypocritical behaviour is impressively bad -- accusing others of
not providing a direct response while you fail to acknowledge that 
numerous
specific PSTN collapse events HAVE been presented here recently -- and
presented by several different folks.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 18:34:00 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08191
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 18:34:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9NaUs08458
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 18:36:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NaUv08455
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 18:36:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08186
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 18:33:29 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NZ1v08417;
	Mon, 9 Dec 2002 18:35:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NYev08392
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 18:34:40 -0500
Received: from mace-in.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08133
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 18:31:39 -0500 (EST)
Received: from [10.32.253.236]
	by 171.70.144.186 (mace-in.cisco.com) with SMTP; 09 Dec 2002 07:36:16 +0000
Message-Id: <5.2.0.9.2.20021209152513.03467b20@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.2.0.9
Date: Mon, 09 Dec 2002 15:34:30 -0800
To: <ieprep@ietf.org>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] RE: PSTN collapse
In-Reply-To: <A29143A40AEAE3459FF829D1DD43316101D9B07D@KC-MAIL2.kc.umkc.
 edu>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

At 04:57 PM 12/9/2002 -0600, Ayyasamy, Senthilkumar  (UMKC-Student) wrote:
>I am aware of only RFC 896, which talks about congestive collapse
>and how it leads to "Internet meltdown".

You should also take a look at RFC 970, which describes the mechanism for 
congestive collapse. This did in fact occur in late 1987 and most of 1988. 
The issue was that most links in the 56 KBPS NSFNET were running at 
capacity. Your queuing theory courses should have told you that 100% 
utilization implies infinite delay, something experimentally shown to apply 
in that network. The Fuzzball Routers used in much of it had a grand total 
of eight buffers in their grand total of 64 KBPS of data memory and 64 KBPS 
of program memory, which meant that anything resembling a burst was an 
instant overload. You might notice that a lot of Van Jacobsen's work on 
making TCP robust in the presence of loss is dated 1988; there is a reason. 
He was already working on making it robust in the presence of reordering; 
that related to CSNET's X.25 network, which opened up an arbitrary number 
of X.25 circuits between any two routers and distributed load across them, 
resulting in random traffic reordering on a truly large scale.

At the time, Father Time and his minions deployed a mechanism similar to 
that described in RFC 970 to help overcome the issues. That helped, but in 
the end could not create bandwidth. The problem was solved with the 
deployment of the NSFNET DS3 backbone - which had it's own problems, like 
"gray links" that passed perhaps half the messages sent over them without 
link layer errors...

Woody, your turn to chime in :^) 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 18:55:06 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08796
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 18:55:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gB9NvaS09806
	for ieprep-archive@odin.ietf.org; Mon, 9 Dec 2002 18:57:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Nvav09803
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 9 Dec 2002 18:57:36 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08768
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 18:54:35 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9Nu4v09773;
	Mon, 9 Dec 2002 18:56:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gB9NtFv09703
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 18:55:15 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA08680
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 18:52:12 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.115])
	by gnat.inet.org (Postfix) with ESMTP
	id 63A6D67116; Mon,  9 Dec 2002 15:21:25 -0500 (EST)
Date: Mon, 9 Dec 2002 18:54:38 -0500
Subject: Re: [Ieprep] Re: PSTN collapse
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep@ietf.org
To: Fred Baker <fred@cisco.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <5.2.0.9.2.20021209145936.045ec898@mira-sjcm-4.cisco.com>
Message-Id: <93C0C9EC-0BD1-11D7-92FE-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


My employer has a VoIP internal phone system also (built out of 
different
make/model equipment than Fred's :-).  Our RTP office is similarly up 
and
running today for IP and our internal-only VoIP, though many homes and 
businesses
lack telephone service in RTP.

By the way, the CATV infrastructure in RTP seems to be more robust than
the PSTN infrastructure during this particular weather-related scenario.
A nice artifact of IP is that it can use not only the last-mile PSTN
infrastructure, but also wireless, CATV, and other kinds of 
infrastructures.
And with dynamic routing, IP automatically finds the best remaining 
path,
without any special configuration needed to enable that capability.

Ran
rja@extremenetworks.com


On Monday, Dec 9, 2002, at 18:24 America/Montreal, Fred Baker wrote:
> By the way, we're having the same behavior in North Carolina at the 
> moment. They had an ice storm last week, and while Internet 
> connectivity and therefore Cisco's VoIP networks are still working 
> just fine, the local telephone company seems to be having some 
> challenges. No slight intended against said telephone company; there 
> any number of last mile problems, and when the wire is busted, it 
> really doesn't matter what kind of traffic you're trying to run over 
> it. But their fundamental problem is with the architecture of the 
> "intelligent network": it is subject to brittleness under failure 
> conditions where Internet technology will automatically re-route.

Aside:
	It also matters that much of the RTP PSTN infrastructure is on poles.
Poles have much better MTTR but much much worse MTBF for weather damage
(e.g. hurricanes, ice storms).

> I have an amusing email from an engineer who reports that he is 
> working from home today and has call-forwarded his home phone to his 
> desktop VoIP phone. He had a call this morning from the phone company, 
> who wanted to know when his service had been restored. They didn't 
> like being told that their dial tone was still unavailable and he was 
> being serviced by VoIP from 3000 miles away. 

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec  9 23:59:20 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA17340
	for <ieprep-archive@odin.ietf.org>; Mon, 9 Dec 2002 23:59:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBA51qR25211
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 00:01:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBA51qv25208
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 00:01:52 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA17336
	for <ieprep-web-archive@ietf.org>; Mon, 9 Dec 2002 23:58:49 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBA508v25141;
	Tue, 10 Dec 2002 00:00:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBA4xTv25094
	for <ieprep@optimus.ietf.org>; Mon, 9 Dec 2002 23:59:29 -0500
Received: from kcmso2.proxy.att.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA17287
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 23:56:26 -0500 (EST)
Received: from attrh3i.attrh.att.com ([135.71.62.12])
	by kcmso2.proxy.att.com (AT&T IPNS/MSO-4.0) with ESMTP id gBA4x0A9018046
	for <ieprep@ietf.org>; Mon, 9 Dec 2002 22:59:20 -0600 (CST)
Received: from OCCLUST03EVS1.ugd.att.com (135.71.164.11) by attrh3i.attrh.att.com (6.5.032)
        id 3DF17FA4001243DB; Mon, 9 Dec 2002 23:59:08 -0500
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Subject: RE: [Ieprep] Re: PSTN collapse
Date: Mon, 9 Dec 2002 23:59:08 -0500
Message-ID: <456FD0E7B215B24ABBAD711614E4A7A20184252B@OCCLUST03EVS1.ugd.att.com>
Thread-Topic: [Ieprep] Re: PSTN collapse
Thread-Index: AcKf2myQJPtiBW3vSVyErQRGNk/S9QAJO//w
From: "Dvorak, Charles A (Chuck), ALASO" <cdvorak@att.com>
To: "Fred Baker" <fred@cisco.com>, <ieprep@ietf.org>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBA4xTv25095
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

Fred

You said:

<snip>
> there any number of last mile problems, and 
> when the wire is busted, it really doesn't matter what kind 
> of traffic you're trying to run over it. 
<snip>

Thanks for correctly pointing out that even the Internet needs the
physical layer :-)

Many of the statements made about the PSTN here have been so biased that
they do not deserve a reply. True, there are some isolated problems that
continue to plaque some carriers, but for the most part in the switched
network, things are dynamic (like the Internet) and there is no single
point of failure, and in fact pretty sophisicated techniques are used
(like the Internet). 

For example, as long as traffic gets into AT&T's network, where we have
had real-time network routing for many years and dynamic routing _since
1984_, traffic almost always gets through. This was evident on 9/11 when
we carried 50% more traffic than the usual load. 

But of course if the unprotected access pipe of _any_ user is down, any
core network (PSTN or the Internet) never sees the traffic. Take out a
single, dedicated data pipe that carries traffic to an access
router...and the Internet "collapses" (at least for users of that pipe).

The real issue is when people still have access (e.g., wireless) but
there is severe network congestion, say due to a disaster, what do you
do in the network? You've got a bottleneck, and while the PSTN and the
Internet deal with bottlenecks differently, the issue for this group
should be: When some deserving (emergency-related) traffic merits
priority treatment, and authorized users so authenticate, how does the
network increase the probabilty of completion over non-prioritized
traffic? (Note I did not say guarantee...this is only done in private
networks)

The PSTN does this today; a bit gets set in SS7 and the switches give
priority to such calls. Nothing is guaranteed, as under emergency
situations no one knows what facilities remain and what the traffic will
be. But at least the emergency traffic gets prioritized treatment. 

It seems to me we could successfully do something similar in IEPREP,
which would be a proactive step as opposed to PSTN-bashing, or some of
the recent personal attacks by others.

Chuck Dvorak

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 05:40:30 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA03348
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 05:40:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAAgsh20174
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 05:42:54 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAAgrv20171
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 05:42:53 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA03341
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 05:39:58 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAAfBv20146;
	Tue, 10 Dec 2002 05:41:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAAeQv20103
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 05:40:26 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA03318
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 05:37:31 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Tue, 10 Dec 2002 05:39:47 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121005400614898
 ; Tue, 10 Dec 2002 05:40:06 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPH6C6B>; Tue, 10 Dec 2002 05:41:38 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A1A@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Mpierce1@aol.com '" <Mpierce1@aol.com>,
        "'saq66@umkc.edu '" <saq66@umkc.edu>,
        "'woody@pch.net '" <woody@pch.net>
Cc: "'KIMBERLY.S.KING@saic.com '" <KIMBERLY.S.KING@saic.com>,
        "'dhc@dcrocker.net '" <dhc@dcrocker.net>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
Date: Tue, 10 Dec 2002 05:39:56 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] RE: PSTN collapse
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Mike said:
>As usual, I see that you can not answer a direct question regarding a
>wild claim about how bad the PSTN is. 

Mike,

I started this "PSTN collapse" talk when I made the following statement 

"At the crux of the matter, the Internet is not centrally controlled, is
extremely survivable, and was specifically engineered to avoid congestion
collapse so *if* the problem is to avoid congestion collapse in an emergency
(as defined by the way the PSTN collapses in an emergency) then the answer
probably is that the Internet already does that. " 


So, what do I mean by "PSTN collapse"?  I mean that many of the population
in a disaster area cannot get dial tone or receives busy signals.  I am not
assuming physical damage to the lines.  I am assuming that people are
blocked because there aren't enough resources to share.  If this were not a
genuine problem during emergencies, then the U.S. government has spent an
enormous amount of money on a service called GETS that was completely
unneeded.  I assure you they have not created GETS on the whim that maybe
such a problem could occur.  

Note that I *never* said that the PSTN is bad, nor do I believe such a
thing.  We are not here to have an Internet vs. PSTN battle.  Both networks
are good; one happy family!

My point is that this working group needs to create concise problem
statements about exactly what problem(s) they are attempting to solve.  It
is also to examine if the problems encountered are really the same as the
PSTN (e.g., call blocking) or if they have new problems.

Kimberly


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 08:14:24 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA05577
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 08:14:24 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBADGpB27823
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 08:16:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBADGov27820
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 08:16:50 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA05574
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 08:13:53 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBADF8v27738;
	Tue, 10 Dec 2002 08:15:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBADEhv27699
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 08:14:43 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA05457
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 08:11:46 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Tue, 10 Dec 2002 08:14:07 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121008142817749
 ; Tue, 10 Dec 2002 08:14:28 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPH6D06>; Tue, 10 Dec 2002 08:16:00 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A1C@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Gunn, Janet'" <Janet.Gunn@DynCorp.com>,
        "'King, Kimberly  S.'" <KIMBERLY.S.KING@saic.com>,
        "'ieprep@ietf.org'" <ieprep@ietf.org>
Cc: "'Henry.Sinnreich@wcom.com'" <Henry.Sinnreich@wcom.com>
Subject: RE: [Ieprep] commercial SIP service
Date: Tue, 10 Dec 2002 08:14:23 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Janet,

> Does this mean that only the 800 based "alternate GETS 
> numbers" would work,
> or would the 710 based numbers work too?

I asked "If you dial 710 now on your SIP phone, do you reach the PSTN and
one of the IXCs?"

and Henry's answer was "yes" along with the more detailed explanation that I
provided.

Kimberly

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 09:40:55 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07910
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 09:40:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAEhMX00914
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 09:43:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAEhMv00911
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 09:43:22 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07827
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 09:40:23 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAEfbv00526;
	Tue, 10 Dec 2002 09:41:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAEdov32701
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 09:39:50 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07509
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 09:36:52 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id E1EAC67108; Tue, 10 Dec 2002 06:06:36 -0500 (EST)
Date: Tue, 10 Dec 2002 09:39:46 -0500
Subject: Re: [Ieprep] Re: PSTN collapse
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: <ieprep@ietf.org>
To: "Dvorak, Charles A (Chuck), ALASO" <cdvorak@att.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <456FD0E7B215B24ABBAD711614E4A7A20184252B@OCCLUST03EVS1.ugd.att.com>
Message-Id: <3A4BEDA5-0C4D-11D7-9E28-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 9, 2002, at 23:59 America/Montreal, Dvorak, Charles A 
(Chuck), ALASO wrote:
> The real issue is when people still have access (e.g., wireless) but
> there is severe network congestion, say due to a disaster, what do you
> do in the network?

During 9/11, the Internet did NOT suffer significant network congestion 
inside
the core (primarily because of massive over-provisioning, which ISPs 
find is
more cost-effective than deploying QoS).   As various people have been 
noting,
there were particular access links that were cut or congested, but that 
is
a separate issue.  And in many cases, the Internet was able to route 
around
fibre cuts that did cause problems for the RBOCs.

Nor was 9/11 an isolated case.  A long list of natural disasters (e.g.
Kobe, Loma Prieta, Hurricane Andrew, Northridge, various twisters) are
good examples where the telephone system did suffer significant 
congestion
(resulting in inability to place calls) but the Internet did NOT suffer
significant congestion (resulting in people using the Internet 
successfully
to communicate even though their phone service was unavailable due to 
switch
congestion).

> You've got a bottleneck,

First, folks need to demonstrate that the Internet backbones have 
suffered
or are likely to suffer severe congestion.  So far, all the evidence, 
including
the recent NAS study, indicates that this has not happened so far -- 
and there is
nil evidence that it is likely to happen in the near future.

> and while the PSTN and the
> Internet deal with bottlenecks differently, the issue for this group
> should be: When some deserving (emergency-related) traffic merits
> priority treatment, and authorized users so authenticate, how does the
> network increase the probabilty of completion over non-prioritized
> traffic? (Note I did not say guarantee...this is only done in private
> networks)

IP has packets, and does not have "calls" so there is no concept in IP
of "call completion".  Basically, some folks are trying to apply a 
circuit-switched
network concept to a packet-switched network -- which a bunch of folks 
here
think is a broken approach.  To be fair, it is as broken as trying to 
apply
a packet-switched network concept to a circuit-switched network.

So if anyone wants to make headway with one's agenda, one needs to 
first convince
most folks here that one's perception of the problem is reasonable.  
Until
someone does that in this group, the group will remain stalled 
indefinitely,
because "rough consensus" is required to make progress in IETF (and we
don't have rough consensus right now on what issues this group ought to 
focus on).

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 09:56:59 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA08900
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 09:56:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAExQS02892
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 09:59:26 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAExQv02889
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 09:59:26 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA08846
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 09:56:28 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAEw1v02623;
	Tue, 10 Dec 2002 09:58:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAEv1v02502
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 09:57:01 -0500
Received: from multicasttech.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA08703
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 09:54:02 -0500 (EST)
Received: from [140.221.56.5] (account marshall_eubanks HELO localhost)
  by multicasttech.com (CommuniGate Pro SMTP 3.4.8)
  with ESMTP-TLS id 1531829; Tue, 10 Dec 2002 09:56:57 -0500
Date: Tue, 10 Dec 2002 09:56:58 -0500
Subject: Re: [Ieprep] RE: PSTN collapse
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v482)
Cc: "'Mpierce1@aol.com '" <Mpierce1@aol.com>,
        "'saq66@umkc.edu '" <saq66@umkc.edu>,
        "'woody@pch.net '" <woody@pch.net>,
        "'dhc@dcrocker.net '" <dhc@dcrocker.net>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: Marshall Eubanks <tme@multicasttech.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A1A@mcl-its-exs01.mail.saic.com>
Message-Id: <A1E990FA-0C4F-11D7-9B57-003065BA697E@multicasttech.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.482)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

When I lived in California we went through two significant (not major) 
earthquakes.
Both times we could not complete an outbound call for some hours - but 
once
we got through, the connection was fine.

That sounds like a PSTN collapse to me.

I heard from a Japanese colleague that the PSTN collapse after the Kobe
earthquake was particularly bad.

Regards
Marshall Eubanks

On Tuesday, December 10, 2002, at 05:39  AM, King, Kimberly S. wrote:

> Mike said:
>> As usual, I see that you can not answer a direct question regarding a
>> wild claim about how bad the PSTN is.
>
> Mike,
>
> I started this "PSTN collapse" talk when I made the following statement
>
> "At the crux of the matter, the Internet is not centrally controlled, is
> extremely survivable, and was specifically engineered to avoid 
> congestion
> collapse so *if* the problem is to avoid congestion collapse in an 
> emergency
> (as defined by the way the PSTN collapses in an emergency) then the 
> answer
> probably is that the Internet already does that. "
>
>
> So, what do I mean by "PSTN collapse"?  I mean that many of the 
> population
> in a disaster area cannot get dial tone or receives busy signals.  I am 
> not
> assuming physical damage to the lines.  I am assuming that people are
> blocked because there aren't enough resources to share.  If this were 
> not a
> genuine problem during emergencies, then the U.S. government has spent 
> an
> enormous amount of money on a service called GETS that was completely
> unneeded.  I assure you they have not created GETS on the whim that 
> maybe
> such a problem could occur.
>
> Note that I *never* said that the PSTN is bad, nor do I believe such a
> thing.  We are not here to have an Internet vs. PSTN battle.  Both 
> networks
> are good; one happy family!
>
> My point is that this working group needs to create concise problem
> statements about exactly what problem(s) they are attempting to solve.  
> It
> is also to examine if the problems encountered are really the same as 
> the
> PSTN (e.g., call blocking) or if they have new problems.
>
> Kimberly
>
>
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep
>
                                  Regards
                                  Marshall Eubanks

This e-mail may contain confidential and proprietary information of
Multicast Technologies, Inc, subject to Non-Disclosure Agreements

T.M. Eubanks
Multicast Technologies, Inc.
10301 Democracy Lane, Suite 410
Fairfax, Virginia 22030
Phone : 703-293-9624       Fax     : 703-293-9609
e-mail : tme@multicasttech.com
http://www.multicasttech.com

Test your network for multicast :
http://www.multicasttech.com/mt/
  Status of Multicast on the Web  :
  http://www.multicasttech.com/status/index.html

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 10:14:03 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA09905
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 10:14:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAFGUl05138
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 10:16:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFGUv05135
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 10:16:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA09857
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 10:13:31 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFF3v04976;
	Tue, 10 Dec 2002 10:15:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFEZv04908
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 10:14:35 -0500
Received: from cs.columbia.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA09736
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 10:11:36 -0500 (EST)
Received: from opus.cs.columbia.edu (opus.cs.columbia.edu [128.59.20.100])
	by cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gBAFEUR3013406
	(version=TLSv1/SSLv3 cipher=EDH-RSA-DES-CBC3-SHA bits=168 verify=NOT);
	Tue, 10 Dec 2002 10:14:31 -0500 (EST)
Received: from cs.columbia.edu (tallgrass.netlab.uky.edu [204.198.76.66])
	(authenticated bits=0)
	by opus.cs.columbia.edu (8.12.6/8.12.6) with ESMTP id gBAFETdG024822
	(version=TLSv1/SSLv3 cipher=RC4-MD5 bits=128 verify=NOT);
	Tue, 10 Dec 2002 10:14:30 -0500 (EST)
Message-ID: <3DF60549.9010505@cs.columbia.edu>
Date: Tue, 10 Dec 2002 10:16:25 -0500
From: Henning Schulzrinne <hgs@cs.columbia.edu>
User-Agent: Mozilla/5.0 (Windows; U; Windows NT 5.1; en-US; rv:1.2.1) Gecko/20021130
X-Accept-Language: en-us, en
MIME-Version: 1.0
To: Marshall Eubanks <tme@multicasttech.com>
CC: "King, Kimberly S." <KIMBERLY.S.KING@saic.com>,
        "'Mpierce1@aol.com '" <Mpierce1@aol.com>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
Subject: Re: [Ieprep] RE: PSTN collapse
References: <A1E990FA-0C4F-11D7-9B57-003065BA697E@multicasttech.com>
In-Reply-To: <A1E990FA-0C4F-11D7-9B57-003065BA697E@multicasttech.com>
Content-Type: text/plain; charset=us-ascii; format=flowed
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Does anybody have real, numeric data on how much additional Internet 
traffic was generated during these events?

There are really two separate dealing-with-collapse questions here that 
seem to be conflated constantly:

- how well does a system deal with a sudden, unplanned flood of 
additional (emergency-induced) traffic?

- how well does a system cope with partial destruction of the underlying 
infrastructure

At least from what I can tell, except for places like links to CNN, 
overall Internet traffic does not increase significantly during 
emergencies (or has not in the past). CNN.com and similar sites 
obviously did "collapse" until they could be reconfigured (e.g., by 
converting to text-only content). People send more email during 
emergencies, but they probably worry less about ordering books from 
Amazon, exchanging music files or playing networked games.

Thus, unless somebody has verifiable proof that the Internet had, say, 
sustained traffic several times larger than usual during those 
emergencies, these experiences say very little should a significant 
fraction of the existing interactive communications traffic move to the 
Internet. After all, except for downtown, the problem in Manhattan was 
mostly a sudden surge in volume. (Downtown, there was no Internet, 
either...)

A separate question is whether emergencies are likely to induce such 
floods of traffic. However, I think it is more honest to separate these 
issues.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 10:19:16 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10172
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 10:19:16 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAFLii05688
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 10:21:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFLiv05685
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 10:21:44 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10136
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 10:18:45 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFK4v05584;
	Tue, 10 Dec 2002 10:20:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFJbv05486
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 10:19:37 -0500
Received: from imo-d02.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10037
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 10:16:38 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d02.mx.aol.com (mail_out_v34.13.) id j.149.49a6485 (16335);
	Tue, 10 Dec 2002 10:19:24 -0500 (EST)
Message-ID: <149.49a6485.2b275ffc@aol.com>
Date: Tue, 10 Dec 2002 10:19:24 EST
Subject: Re: [Ieprep] identification of problem
To: rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
X-Mailer: AOL 6.0 for Windows XP US sub 51
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

In a message dated 12/9/2002 2:19:05 PM Eastern Standard Time, 
rja@extremenetworks.com writes:


And you seem to think that "local congestion" doesn't qualify as collapsed
for the PSTN (which I disagree with), but why do you have such a blatent 
double
standard that it does qualify as collapsed in the IP network case ?



[MAP] I have never said or implied that I thought that a "local congestion" 
qualifies as "collapsed" in the IP network case. The whole point I've been 
trying to make is the need to develop ways to deal with "local congestion", 
i.e., congestion on a single link which the emergency traffic is trying to 
use. (I've been told that no IP routing protocol can route around such 
congestion and it will result in all calls getting equally poor quality.) 
When claims have been made that there is no problem, I've asked for specific 
solutions to specific cases, but have gotten nothing but sarcastic responses, 
either that the requirement doesn't exist, or that the case I'm describing is 
stupid.

I'd like to hear a specific solution to the congestion case I detailed in my 
e-mail of 6 Dec. I have to take the absence of serious answers as an 
indication that there are none.

Mike Pierce
Artel
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 10:19:18 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10185
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 10:19:18 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAFLkW05707
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 10:21:46 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFLkv05704
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 10:21:46 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10141
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 10:18:47 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFK1v05543;
	Tue, 10 Dec 2002 10:20:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFJ9v05469
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 10:19:09 -0500
Received: from imo-m06.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10010
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 10:16:11 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m06.mx.aol.com (mail_out_v34.13.) id f.70.27734718 (16335);
	Tue, 10 Dec 2002 10:18:30 -0500 (EST)
Message-ID: <70.27734718.2b275fc6@aol.com>
Date: Tue, 10 Dec 2002 10:18:30 EST
To: saq66@umkc.edu, woody@pch.net
CC: KIMBERLY.S.KING@saic.com, dhc@dcrocker.net, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
X-Mailer: AOL 6.0 for Windows XP US sub 51
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] Re: PSTN collapse
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

In a message dated 12/9/2002 5:57:59 PM Eastern Standard Time, saq66@umkc.edu 
writes:


Please provide reference to PSTN collapse (since i dont have any
access to ITU-T documents.) If local meltdown is not collapse,
then what is your definition of PSTN collapse? 



[MAP] I didn't start the reference to "PSTN collapse". I have no idea what it 
means, since it has never happened as far as I know, given any definition 
that I could think up. There is no ITU-T definition.

"Local meldown" within the PSTN, as you call, it is common due to unforeseen 
events (backhoes digging up thousand pair telephone cables, central offices 
burning, etc.) and local customers can not make phone calls. There has never 
been an argument on that. That is not "PSTN collapse". (Remember, the "N" 
means network.) In fact, the PSTN contains specific protocols to deal with 
these local ituations. But such "local meltdowns" are also common in the 
Internet. If the server that I am connected to goes down (or is "taken down 
for maintenance" as seems to happen weekly) I lose service and would not get 
a call through on my SIP phone either. Of coures, the number of "subscribers" 
affected by such a failure of a server is typically small, unlike the 100 
thousand or so when the Hinsdale central office burned. That isn't the point. 
The performance measure for this is the probability of any one "subscriber" 
being out of service (no dial tone) for how long per year. The overall 
performance for the PSTN (at least in the US not counting trees over wires in 
the local neighborhood) is on the order of minutes per year per subscriber 
(the infamous "5 nines").

For those who missed it, during the week of the IETF I was watching CNN news 
one night and at the end of the newscast, the announcer said "Send us email 
except when the server is down which is Thursday." Unfortunately, that is the 
common impression people have of Internet service.

Mike Pierce
Artel
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 10:20:04 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10261
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 10:20:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAFMVT05783
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 10:22:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFMVv05780
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 10:22:31 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10234
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 10:19:32 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFL1v05654;
	Tue, 10 Dec 2002 10:21:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFK8v05597
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 10:20:08 -0500
Received: from imo-r06.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10063
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 10:17:09 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r06.mx.aol.com (mail_out_v34.13.) id j.189.124aff8c (16335);
	Tue, 10 Dec 2002 10:19:54 -0500 (EST)
Message-ID: <189.124aff8c.2b27601a@aol.com>
Date: Tue, 10 Dec 2002 10:19:54 EST
Subject: Re: [Ieprep] Re: PSTN collapse
To: rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_189.124aff8c.2b27601a_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_189.124aff8c.2b27601a_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/9/2002 6:30:52 PM Eastern Standard Time, 
rja@extremenetworks.com writes:


Mike,

    [RJA] You are seriously inconsistent with yourself.  Access links are 
entirely
comparable to "wires to your house" -- because those are common kinds of 
access
links.  Yet, you are totally worked up about IP congestion on access links 
even
though most access links only have one site (e.g. house in the case of DSL)
behind the link.  If local outages don't count for the PSTN, then it is only
fair, reasonable, and logical that they don't count against the Internet.


[MAP] Yes, I understand that is one case, "access" links are like phone wires 
coming into my house. "Local outages" count in both cases, but that again is 
not the real issue here, rather congestion is. In the PSTN, congestion is 
dealt with on my phone wire since only 1 call can be made at a time. The CO 
won't connect a second call to my line when one is there, thus not causing 
the existing call to become of such poor quality that I can't talk. And for 
the emergency case, whether it is a 911 call or an authorized GETS user, they 
can use the line by disconnecting the current call or telling the other 
person to get off the line (with penalties for not giving it up). In the 
Internet case, it has been clealy identified that in the case of a local 
access line (into my house or serving a group of end users) that this is not 
true. Many distant users, can establish sessions (SIP) to devices at my end 
of that "access" link and nothing stops them from all trying to send media 
until all of them have such degraded quality that none of them can talk. The 
whole point here has been to find a solution to this "problem".


    [RJA] You also selectively reply to email.  Several folks gave specific 
examples
where access wires were in place, yet dial tone did not exist for long periods
of time (e.g. Bill Woodcock's Loma Prieta example, among several posted here
recently).  You seem to have a kind of selective amnesia to facts that you
find inconvenient.



[MAP] I was not aware that there was a question related to the above. Of 
course there are many cases in which access wires are in place and no dial 
tone for long periods. No one has diputed that. And I freely admit that there 
have been some massive failues in the PSTN caused, for example, by a Central 
Offic burning. (I just don't think "PSTN collapse" is the right way to 
describe such an event.) I don't know what the point of the above quite wa or 
the question was.

    [RJA] Your hypocritical behaviour is impressively bad -- accusing others 
of
not providing a direct response while you fail to acknowledge that numerous
specific PSTN collapse events HAVE been presented here recently -- and
presented by several different folks.


[MAP] I'll let others who read this list judge who is being hypocritcal.

Mike Pierce
Artel


--part1_189.124aff8c.2b27601a_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/9/2002 6:30:52 PM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR>Mike,
<BR>
<BR> &nbsp;&nbsp;&nbsp;[RJA] You are seriously inconsistent with yourself. &nbsp;Access links are entirely
<BR>comparable to "wires to your house" -- because those are common kinds of access
<BR>links. &nbsp;Yet, you are totally worked up about IP congestion on access links even
<BR>though most access links only have one site (e.g. house in the case of DSL)
<BR>behind the link. &nbsp;If local outages don't count for the PSTN, then it is only
<BR>fair, reasonable, and logical that they don't count against the Internet.
<BR>
<BR>
<BR>[MAP] Yes, I understand that is one case, "access" links are like phone wires coming into my house. "Local outages" count in both cases, but that again is not the real issue here, rather congestion is. In the PSTN, congestion is dealt with on my phone wire since only 1 call can be made at a time. The CO won't connect a second call to my line when one is there, thus not causing the existing call to become of such poor quality that I can't talk. And for the emergency case, whether it is a 911 call or an authorized GETS user, they can use the line by disconnecting the current call or telling the other person to get off the line (with penalties for not giving it up). In the Internet case, it has been clealy identified that in the case of a local access line (into my house or serving a group of end users) that this is not true. Many distant users, can establish sessions (SIP) to devices at my end of that "access" link and nothing stops them from all trying to send media until!
  all of them have such degraded quality that none of them can talk. The whole point here has been to find a solution to this "problem".
<BR>
<BR>
<BR> &nbsp;&nbsp;&nbsp;[RJA] You also selectively reply to email. &nbsp;Several folks gave specific examples
<BR>where access wires were in place, yet dial tone did not exist for long periods
<BR>of time (e.g. Bill Woodcock's Loma Prieta example, among several posted here
<BR>recently). &nbsp;You seem to have a kind of selective amnesia to facts that you
<BR>find inconvenient.
<BR>
<BR>
<BR>
<BR>[MAP] I was not aware that there was a question related to the above. Of course there are many cases in which access wires are in place and no dial tone for long periods. No one has diputed that. And I freely admit that there have been some massive failues in the PSTN caused, for example, by a Central Offic burning. (I just don't think "PSTN collapse" is the right way to describe such an event.) I don't know what the point of the above quite wa or the question was.
<BR>
<BR> &nbsp;&nbsp;&nbsp;[RJA] Your hypocritical behaviour is impressively bad -- accusing others of
<BR>not providing a direct response while you fail to acknowledge that numerous
<BR>specific PSTN collapse events HAVE been presented here recently -- and
<BR>presented by several different folks.
<BR>
<BR>
<BR>[MAP] I'll let others who read this list judge who is being hypocritcal.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_189.124aff8c.2b27601a_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 11:01:22 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA11538
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 11:01:22 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAG3o207934
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 11:03:50 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAG3ov07931
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 11:03:50 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA11505
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 11:00:51 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFxGv07767;
	Tue, 10 Dec 2002 10:59:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAFwMv07736
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 10:58:22 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA11289
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 10:55:22 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAG9D6>; Tue, 10 Dec 2002 10:55:48 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149D3@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        "Dvorak, Charles A (Chuck), ALASO" <cdvorak@att.com>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] Re: PSTN collapse
Date: Tue, 10 Dec 2002 10:55:39 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>



-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]

 As various people have been 
noting,
there were particular access links that were cut or congested, but that 
is
a separate issue. 

[JPG] I don't think that is "a separate issue".  I think that is the most
pressing part of the overall issue IEPREP is supposed to be addressing. (

[JPG] I don't think it is useful to be debating the different kinds of
"collapse" in the circuit switched vs. packet switched networks.  I agree
that it is important to understand how they are different.  But the "my
network is better than your network" doesn't help.  That being said, I am
sorry, but I can't resist adding my 2 cents.  Sometime in the last decade I
was consulting with the IT department of a university in California.  One of
the tangents of the project was to look at whether it would be a good time
to consider migrating from the PBX to VoIP.  One of the  arguments against
VoIP on the campus then was "In the last earthquake, the LAN went down, but
I still had dial tone on the PBX".

[JPG] Again, we come back to "the last mile".

IP has packets, and does not have "calls" so there is no concept in IP
of "call completion".  

[JPG] If you mean the IP protocol by itself, OK.  But the term "IP" is often
used to apply to the entire protocol suite (IP, UDP, TCP, RTP, SIP, etc.),
and some parts of the protocol suite DO have "calls".


because "rough consensus" is required to make progress in IETF (and we
don't have rough consensus right now on what issues this group ought to 
focus on).

[JPG] Maybe I am being naive, but it looks to me as if we have "rough
consensus" that "access and gateways" are the issues the group should focus
on.
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 11:10:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA11897
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 11:10:30 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAGCvq09002
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 11:12:57 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAGCvv08999
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 11:12:57 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA11885
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 11:09:59 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAG83v08808;
	Tue, 10 Dec 2002 11:08:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAG7xv08773
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 11:07:59 -0500
Received: from imo-m08.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA11663
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 11:05:00 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m08.mx.aol.com (mail_out_v34.13.) id j.185.1339863f (18251);
	Tue, 10 Dec 2002 11:06:55 -0500 (EST)
Message-ID: <185.1339863f.2b276b1f@aol.com>
Date: Tue, 10 Dec 2002 11:06:55 EST
Subject: Re: [Ieprep] Re: PSTN collapse
To: rja@extremenetworks.com, cdvorak@att.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_185.1339863f.2b276b1f_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_185.1339863f.2b276b1f_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/10/2002 9:42:36 AM Eastern Standard Time, 
rja@extremenetworks.com writes:


> Nor was 9/11 an isolated case.  A long list of natural disasters (e.g.
> Kobe, Loma Prieta, Hurricane Andrew, Northridge, various twisters) are
> good examples where the telephone system did suffer significant congestion
> (resulting in inability to place calls) but the Internet did NOT suffer
> significant congestion (resulting in people using the Internet successfully
> to communicate even though their phone service was unavailable due to 
> switch congestion).
> 

[MAP] I fail to see how these events (excpet 9/11) are accurate indications 
of how VoIP might function in a major event today. They all (except 9/11) 
occurred before there was any VoIP (of any significant amount). How will an 
emergency worker get access after the next big event?

Incidentally, I was in San Francisco during the really big one in 1989 and 
made about 10 phone calls out of the area right after with no problem. People 
may not have been able to call into the area, but that is one of the 
congestion management procedures used in the PSTN. The only problems I heard 
of in that case, as well as after Andrew, were the usual ones of not being 
able to get fuel to some Central Office generators. (How many Internet 
routers have generators anyway?)

Mike Pierce
Artel


--part1_185.1339863f.2b276b1f_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/10/2002 9:42:36 AM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">Nor was 9/11 an isolated case. &nbsp;A long list of natural disasters (e.g.
<BR>Kobe, Loma Prieta, Hurricane Andrew, Northridge, various twisters) are
<BR>good examples where the telephone system did suffer significant congestion
<BR>(resulting in inability to place calls) but the Internet did NOT suffer
<BR>significant congestion (resulting in people using the Internet successfully
<BR>to communicate even though their phone service was unavailable due to 
<BR>switch congestion).
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>[MAP] I fail to see how these events (excpet 9/11) are accurate indications of how VoIP might function in a major event today. They all (except 9/11) occurred before there was any VoIP (of any significant amount). How will an emergency worker get access after the next big event?
<BR>
<BR>Incidentally, I was in San Francisco during the really big one in 1989 and made about 10 phone calls out of the area right after with no problem. People may not have been able to call into the area, but that is one of the congestion management procedures used in the PSTN. The only problems I heard of in that case, as well as after Andrew, were the usual ones of not being able to get fuel to some Central Office generators. (How many Internet routers have generators anyway?)
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_185.1339863f.2b276b1f_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 11:10:32 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA11910
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 11:10:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAGCxD09020
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 11:12:59 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAGCxv09017
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 11:12:59 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA11890
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 11:10:01 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAG81v08791;
	Tue, 10 Dec 2002 11:08:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAG7sv08769
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 11:07:54 -0500
Received: from imo-d09.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA11651
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 11:04:54 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d09.mx.aol.com (mail_out_v34.13.) id c.198.11f3feb6 (18251);
	Tue, 10 Dec 2002 11:07:28 -0500 (EST)
Message-ID: <198.11f3feb6.2b276b40@aol.com>
Date: Tue, 10 Dec 2002 11:07:28 EST
Subject: Re: [Ieprep] Re: PSTN collapse
To: fred@cisco.com, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_198.11f3feb6.2b276b40_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_198.11f3feb6.2b276b40_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/9/2002 6:26:33 PM Eastern Standard Time, fred@cisco.com 
writes:


> I have an amusing email from an engineer who reports that he is working 
> from home today and has call-forwarded his home phone to his desktop VoIP 
> phone. He had a call this morning from the phone company, who wanted to 
> know when his service had been restored. They didn't like being told that 
> their dial tone was still unavailable and he was being serviced by VoIP 
> from 3000 miles away. 
> 


Great story, and I believe it would be true. Perfect example of having an 
aternative for the last mile - that hunk of aerial telephone wire that always 
comes down in an ice-storm. I wonder how that individual gets his Internet 
service. Maybe by a TV cable that happens to go a different way (route 
diversity). In my case, we have no TV cable, but two phone lines. When they 
get to the pole outside, they go opposite directions up and down the road to 
separate concentrators to different Central Offices.

When all is done in NC, I think you'll find that the "PSTN", that is the 
switches and backbone, did not fail. All the problems are in the access and 
are not congestion problems, but rather broken physical layer, a problem 
shared by both traditional telephony and Internet access.

Mike Pierce
Artel


--part1_198.11f3feb6.2b276b40_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/9/2002 6:26:33 PM Eastern Standard Time, fred@cisco.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">I have an amusing email from an engineer who reports that he is working 
<BR>from home today and has call-forwarded his home phone to his desktop VoIP 
<BR>phone. He had a call this morning from the phone company, who wanted to 
<BR>know when his service had been restored. They didn't like being told that 
<BR>their dial tone was still unavailable and he was being serviced by VoIP 
<BR>from 3000 miles away. 
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>Great story, and I believe it would be true. Perfect example of having an aternative for the last mile - that hunk of aerial telephone wire that always comes down in an ice-storm. I wonder how that individual gets his Internet service. Maybe by a TV cable that happens to go a different way (route diversity). In my case, we have no TV cable, but two phone lines. When they get to the pole outside, they go opposite directions up and down the road to separate concentrators to different Central Offices.
<BR>
<BR>When all is done in NC, I think you'll find that the "PSTN", that is the switches and backbone, did not fail. All the problems are in the access and are not congestion problems, but rather broken physical layer, a problem shared by both traditional telephony and Internet access.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_198.11f3feb6.2b276b40_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 13:18:55 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA16606
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 13:18:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAILNa18056
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 13:21:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAILNv18053
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 13:21:23 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA16595
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 13:18:24 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAIIGv17692;
	Tue, 10 Dec 2002 13:18:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAIH2v17645
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 13:17:02 -0500
Received: from sj-msg-core-2.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA16399
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 13:14:01 -0500 (EST)
Received: from imop.cisco.com (IDENT:mirapoint@imop.cisco.com [171.69.11.44])
	by sj-msg-core-2.cisco.com (8.12.2/8.12.2) with ESMTP id gBAIGXfm000714;
	Tue, 10 Dec 2002 10:16:33 -0800 (PST)
Received: from cisco.com (ssh-sjc-1.cisco.com [171.68.225.134])
	by imop.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABK72328;
	Tue, 10 Dec 2002 10:13:28 -0800 (PST)
Date: Mon, 9 Dec 2002 23:01:32 -0800
Subject: Re: [Ieprep] Problem?
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'Mpierce1@aol.com'" <Mpierce1@aol.com>, ieprep@ietf.org
To: "Gunn, Janet" <Janet.Gunn@dyncorp.com>
From: Rohan Mahy <rohan@cisco.com>
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149CD@chntex02.is.dyncorp.com>
Message-Id: <36BCD8CA-0C0D-11D7-8C59-0003938AF740@cisco.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBAIH2v17646
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


On Monday, December 9, 2002, at 02:52 PM, Gunn, Janet wrote:

> Mike,
>  
> My point was to "back off" and restrict the problem to a really simple 
> example that COULD be addressed- then see how far we can expand it.
>  
> I agree that enhanced service between two pre-defined locations isn't 
> particularly valuable in addressing the overall problem- but I am 
> trying to explore WHERE the line between "doable/practical", and "not 
> possible/not practical" is. 
>  
> While "any to any" opportunistic priority service (like GETS) is 
> desirable, it is not a hard and fast "requirement". When we expanded 
> GETS to WPS, we gave up part of the opportunistic element- it will 
> only work on pre-subscribed cell phones, not any cell phone.  But that 
> is still a big improvement over the "no wireless priority" that 
> existed before.
>  
> Opportunistic "any to any" priority service on the Internet would be 
> highly desirable.  But if it can't be done from a practical 
> perspective (I am not giving up on it, just conceding it for the time 
> being to see what CAN be done), even priority service between 100 
> pre-defined sites (or IP domains) could definitely be an improvement 
> over the status quo.

Right, Improving the ability to do opportunistic priority on one edge 
is useful, incrementally better than what we have, and might be 
deployable. There is no way we are going to be able to do "any to any" 
all at once.

> Bill referred to a government program to fund "bigger pipes" for 
> NGOs.  That would presumably cover the Red Cross in my example, but 
> wouldn't cover the "critical infrastructure" side of things (oil. gas, 
> electric utilities, transportation companies, drug companies, etc.)
>  
> We DON'T know in advance where the emergency will be.  But we probably 
> CAN predict in advance where AT LEAST SOME of the emergency traffic 
> will be.

absolutely.
thanks,
-rohan

> -----Original Message-----
> From: Mpierce1@aol.com [mailto:Mpierce1@aol.com]
> Sent: Monday, December 09, 2002 5:18 PM
> To: ieprep@ietf.org
> Subject: [Ieprep] Problem?
>
> To all,
>
> The following quote, taken from a recent e-mail, certainly shows why 
> we are having such difficulty on this list in getting anywhere.
>
> "If we know _in advance_ that the emergency traffic is going to be 
> between
> A and B, ..."
>
> There seem to be some (maybe only one?) who still do not grasp the 
> problem - emergencies are not preplanned.
>
> Mike Pierce
> Artel
>

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 13:40:20 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA17193
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 13:40:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAIgn919688
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 13:42:49 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAIgnv19685
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 13:42:49 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA17182
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 13:39:49 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAIf4v19646;
	Tue, 10 Dec 2002 13:41:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAIenv19630
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 13:40:49 -0500
Received: from lakemtao08.cox.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA17166
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 13:37:50 -0500 (EST)
From: gregking@cox.net
Received: from smtp.central.cox.net ([172.18.52.51]) by lakemtao08.cox.net
          (InterMail vM.5.01.04.05 201-253-122-122-105-20011231) with SMTP
          id <20021210184045.IHNS2213.lakemtao08.cox.net@smtp.central.cox.net>
          for <ieprep@ietf.org>; Tue, 10 Dec 2002 12:40:45 -0600
To: ieprep@ietf.org
Date: Tue, 10 Dec 2002 13:40:45 -0500
MIME-Version: 1.0
Content-Type: multipart/mixed; boundary="----=____1039545645111_4xM99zXlqO"
Message-Id: <20021210184045.IHNS2213.lakemtao08.cox.net@smtp.central.cox.net>
Subject: [Ieprep] "The Problem" (was Re: PSTN collapse)
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

This is a multi-part message in MIME format.

------=____1039545645111_4xM99zXlqO
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit


If this is "the problem"... have yet to hear of a single ISP that **prohibits** their customers from using CPE devices (e.g. SIP aware firewalls and/or QoS appliances) to do per flow call admission control....

Greg

P.S. Having said that though, I don't believe that the NRC report concludes that this was in fact the 9/11 "problem" 

Mike Pierce said:

> true. Many distant users, can establish sessions (SIP) to devices at my end 
> of that "access" link and nothing stops them from all trying to send media 
> until all of them have such degraded quality that none of them can talk. The 
> whole point here has been to find a solution to this "problem".
> 


 

------=____1039545645111_4xM99zXlqO
Content-Type: text/html;
	name="replyAll"
Content-Disposition: inline;
	filename="replyAll"

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/9/2002 6:30:52 PM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR>Mike,
<BR>
<BR> &nbsp;&nbsp;&nbsp;[RJA] You are seriously inconsistent with yourself. &nbsp;Access links are entirely
<BR>comparable to "wires to your house" -- because those are common kinds of access
<BR>links. &nbsp;Yet, you are totally worked up about IP congestion on access links even
<BR>though most access links only have one site (e.g. house in the case of DSL)
<BR>behind the link. &nbsp;If local outages don't count for the PSTN, then it is only
<BR>fair, reasonable, and logical that they don't count against the Internet.
<BR>
<BR>
<BR>[MAP] Yes, I understand that is one case, "access" links are like phone wires coming into my house. "Local outages" count in both cases, but that again is not the real issue here, rather congestion is. In the PSTN, congestion is dealt with on my phone wire since only 1 call can be made at a time. The CO won't connect a second call to my line when one is there, thus not causing the existing call to become of such poor quality that I can't talk. And for the emergency case, whether it is a 911 call or an authorized GETS user, they can use the line by disconnecting the current call or telling the other person to get off the line (with penalties for not giving it up). In the Internet case, it has been clealy identified that in the case of a local access line (into my house or serving a group of end users) that this is not true. Many distant users, can establish sessions (SIP) to devices at my end of that "access" link and nothing stops them from all trying to send media until!
  all of them have such degraded quality that none of them can talk. The whole point here has been to find a solution to this "problem".
<BR>
<BR>
<BR> &nbsp;&nbsp;&nbsp;[RJA] You also selectively reply to email. &nbsp;Several folks gave specific examples
<BR>where access wires were in place, yet dial tone did not exist for long periods
<BR>of time (e.g. Bill Woodcock's Loma Prieta example, among several posted here
<BR>recently). &nbsp;You seem to have a kind of selective amnesia to facts that you
<BR>find inconvenient.
<BR>
<BR>
<BR>
<BR>[MAP] I was not aware that there was a question related to the above. Of course there are many cases in which access wires are in place and no dial tone for long periods. No one has diputed that. And I freely admit that there have been some massive failues in the PSTN caused, for example, by a Central Offic burning. (I just don't think "PSTN collapse" is the right way to describe such an event.) I don't know what the point of the above quite wa or the question was.
<BR>
<BR> &nbsp;&nbsp;&nbsp;[RJA] Your hypocritical behaviour is impressively bad -- accusing others of
<BR>not providing a direct response while you fail to acknowledge that numerous
<BR>specific PSTN collapse events HAVE been presented here recently -- and
<BR>presented by several different folks.
<BR>
<BR>
<BR>[MAP] I'll let others who read this list judge who is being hypocritcal.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>


------=____1039545645111_4xM99zXlqO--

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 15:33:21 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20044
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 15:33:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAKZp726068
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 15:35:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAKZpv26065
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 15:35:51 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20038
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 15:32:50 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAKW7v25927;
	Tue, 10 Dec 2002 15:32:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAKVev25894
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 15:31:40 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA19936
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 15:28:39 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 0FD7667108; Tue, 10 Dec 2002 11:58:27 -0500 (EST)
Date: Tue, 10 Dec 2002 15:31:33 -0500
Subject: Re: [Ieprep] Re: PSTN collapse
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <70.27734718.2b275fc6@aol.com>
Message-Id: <5F5C0FAA-0C7E-11D7-9E28-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 10, 2002, at 10:18 America/Montreal, Mpierce1@aol.com 
wrote:
> For those who missed it, during the week of the IETF I was watching 
> CNN news
> one night and at the end of the newscast, the announcer said "Send us 
> email
> except when the server is down which is Thursday." Unfortunately, that 
> is the
> common impression people have of Internet service.

Again you make misleading comments, confusing scheduled downtime for one
organisation's mail server with availability of Internet service.

Its just not helpful to the WG progress when you keep doing that 
stuff...

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 15:45:49 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20406
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 15:45:49 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAKmJF27163
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 15:48:19 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAKmJv27160
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 15:48:19 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20402
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 15:45:18 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAKk2v27045;
	Tue, 10 Dec 2002 15:46:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAKjSv27002
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 15:45:28 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20316
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 15:42:26 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 2DA4467108; Tue, 10 Dec 2002 12:12:14 -0500 (EST)
Date: Tue, 10 Dec 2002 15:45:20 -0500
Subject: Re: [Ieprep] Re: PSTN collapse
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149D3@chntex02.is.dyncorp.com>
Message-Id: <4C5AB1EE-0C80-11D7-9E28-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 10, 2002, at 10:55 America/Montreal, Gunn, Janet wrote:
> -----Original Message-----
> From: RJ Atkinson [mailto:rja@extremenetworks.com]
>
>  As various people have been
> noting,
> there were particular access links that were cut or congested, but that
> is
> a separate issue.
>
> [JPG] I don't think that is "a separate issue".  I think that is the 
> most
> pressing part of the overall issue IEPREP is supposed to be 
> addressing. (

Well, at least Mike Pierce seems to think that IEprep is solving the
non-problem of congestion in the core and does not seem to think that
the various approaches to access link congestion are relevant.

> IP has packets, and does not have "calls" so there is no concept in IP
> of "call completion".
>
> [JPG] If you mean the IP protocol by itself, OK.  But the term "IP" is 
> often
> used to apply to the entire protocol suite (IP, UDP, TCP, RTP, SIP, 
> etc.),
> and some parts of the protocol suite DO have "calls".

Hmm.  Your definition of "calls" isn't clear to me.
Clarity would help.

UDP is connectionless, so it probably is not what you mean.
TCP and SCTP have connection state, so might be what you mean.
RTP runs on UDP but does have some session state within RTP.
SIP clearly has the concept of a phone call.

Part of the challenge is that *routers* are IP-layer devices, not
(for example) SIP or RTP or TCP devices.  So the router, which is
the widget controlling which packets go where, which packets are
queued within the router using what queuing discipline, and which
packets get dropped if there were an overload, does not have the
concept of a "call".  So it isn't clear to me how one can apply
the concept of "call" to preferential packet handling.

Applying the concept of "call" to stuff implemented within SIP
(and not reaching down below the application layer) has seemed
straight-forward (to me at least).  I think this portion of stuff
has been addressed by Henning's document, but maybe I misunderstood.

> [JPG] Maybe I am being naive, but it looks to me as if we have "rough
> consensus" that "access and gateways" are the issues the group should 
> focus
> on.

Certainly, I'd agree we should focus on individual access links,
not the core, and not inter-domain QoS, but others here have been
taking issue with that notion.

I am unclear what you mean by "gateways", since some folks might
mean a "SIP gateway", others might mean some kind of middle-box,
others might mean "IP router", and there might be other definitions
that I haven't imagined.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 15:59:29 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20833
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 15:59:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAL1xg27672
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 16:01:59 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAL1xv27669
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 16:01:59 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20823
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 15:58:58 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAL04v27602;
	Tue, 10 Dec 2002 16:00:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAKxev27548
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 15:59:40 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20746
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 15:56:38 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 5DE8A67108; Tue, 10 Dec 2002 12:26:25 -0500 (EST)
Date: Tue, 10 Dec 2002 15:59:31 -0500
Subject: Re: [Ieprep] Re: PSTN collapse & ISP facilities
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <185.1339863f.2b276b1f@aol.com>
Message-Id: <47B054E3-0C82-11D7-9E28-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 10, 2002, at 11:06 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] I fail to see how these events (excpet 9/11) are accurate 
> indications of how VoIP might function in a major event today. They 
> all (except 9/11) occurred before there was any VoIP (of any 
> significant amount). How will an emergency worker get access after the 
> next big event?

	It is not at all clear to me that VoIP is necessary or sufficient.  A 
lot of us,
particularly those who think about the world beyond the US (e.g. Japan, 
where
VoIP is NOT part of their IEprep model), think that it would be pretty 
darned clever
to use also other non-voice applications (e.g. email, web, Jabber, 
other) for
communications during an emergency.  Voice might be part of the mix, 
but certainly
is not the only application that is useful and relevant to IEprep.  So 
the incidents
prior to 9/11/01 are in fact relevant (at least as examples of 
non-voice applications
working fine -- and maybe also as voice applications working fine [1]).

> The only problems I heard of in that case, as well as after Andrew, 
> were the usual ones of not being able to get fuel to some Central 
> Office generators. (How many Internet routers have generators anyway?)

At least Bill Woodcock has indicated that he was unable to get dial 
tone from
within SFBA to SFBA during such a natural disaster.  So the mileage 
varies.

Most ISP core routers are located in the same quality of facilities 
that long-haul
voice carriers use for long-haul voice switches.  For example, @Home's 
core routers
were all located inside the same AT&T Long Lines facilities that had 
their long-haul
voice systems.  @Home's NOC was on its own Siemens generator and there 
were also
cold-spares, hot-spares, offsite backups of data, backup NOC site, etc. 
  NDAs preclude
my listing which specific big backbone ISPs that I know do all this, 
but I can say
that a significant number of them do such as normal commercial 
practice. Some
access ISPs (e.g. Cox Cable) use that quality of facilities even for 
their
head-end equipment.

So most core routers in most large backbone networks have generators, 
have in-depth
backups, spares, and all that other good stuff.

Ran
rja@extremenetworks.com

[1]  Voice over IP applications (written by others, I don't have the 
right clues
for such applications) have been in my paws since the early 90s and 
WERE used
during Loma Prieta and several other natural disasters.

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 15:59:29 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20842
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 15:59:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAL20N27687
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 16:02:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAL20v27684
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 16:02:00 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20825
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 15:58:58 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAL02v27586;
	Tue, 10 Dec 2002 16:00:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAKx6v27539
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 15:59:06 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20741
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 15:56:04 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAG04B>; Tue, 10 Dec 2002 15:56:29 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149DE@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>
Cc: ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: PSTN collapse
Date: Tue, 10 Dec 2002 15:56:20 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>



-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Tuesday, December 10, 2002 3:45 PM



> IP has packets, and does not have "calls" so there is no concept in IP
> of "call completion".
>
> [JPG] If you mean the IP protocol by itself, OK.  But the term "IP" is 
> often
> used to apply to the entire protocol suite (IP, UDP, TCP, RTP, SIP, 
> etc.),
> and some parts of the protocol suite DO have "calls".

Hmm.  Your definition of "calls" isn't clear to me.
Clarity would help.

[JPG] Well, you are the one who said "IP...does not have "calls"...", so
what did YOU mean.

The example protocols were only intended to illustrate what I meant by
"protocol suite", and yes, SIP is the one that most obviously has "calls".

UDP is connectionless, so it probably is not what you mean.
TCP and SCTP have connection state, so might be what you mean.
RTP runs on UDP but does have some session state within RTP.
SIP clearly has the concept of a phone call.


> [JPG] Maybe I am being naive, but it looks to me as if we have "rough
> consensus" that "access and gateways" are the issues the group should 
> focus
> on.

Certainly, I'd agree we should focus on individual access links,
not the core, and not inter-domain QoS, but others here have been
taking issue with that notion.

I am unclear what you mean by "gateways", since some folks might
mean a "SIP gateway", others might mean some kind of middle-box,
others might mean "IP router", and there might be other definitions
that I haven't imagined.

[JPG] Anything that isn't "core", and might be a bottleneck/congestion
point.

Ran
rja@extremenetworks.com
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 16:01:38 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20951
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 16:01:38 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBAL49O27744
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 16:04:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAL49v27741
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 16:04:09 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA20938
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 16:01:07 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAL21v27709;
	Tue, 10 Dec 2002 16:02:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBAL13v27643
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 16:01:03 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20789
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 15:58:02 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 9151467108; Tue, 10 Dec 2002 12:27:49 -0500 (EST)
Date: Tue, 10 Dec 2002 16:00:56 -0500
Subject: Re: [Ieprep] Re: PSTN collapse
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <198.11f3feb6.2b276b40@aol.com>
Message-Id: <79D6B1D7-0C82-11D7-9E28-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 10, 2002, at 11:07 America/Montreal, Mpierce1@aol.com 
wrote:
> When all is done in NC, I think you'll find that the "PSTN", that is 
> the switches and backbone, did not fail. All the problems are in the 
> access and are not congestion problems, but rather broken physical 
> layer, a problem shared by both traditional telephony and Internet 
> access.

This is one reason why microwave links are popular (for both PSTN and 
Internet)
as backups to physical links.

Ran


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 16:31:36 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA21522
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 16:31:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBALY7m29361
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 16:34:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBALY7v29358
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 16:34:07 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA21512
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 16:31:05 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBALW7v29204;
	Tue, 10 Dec 2002 16:32:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBALVLv29167
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 16:31:21 -0500
Received: from bells.cs.ucl.ac.uk (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id QAA21441
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 16:28:18 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.16300-0@bells.cs.ucl.ac.uk>; Tue, 10 Dec 2002 21:31:10 +0000
To: ieprep@ietf.org
Date: Tue, 10 Dec 2002 21:31:08 +0000
Message-ID: <28286.1039555868@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Subject: [Ieprep] 9/11 stats
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


> From: Henning Schulzrinne
>
> Does anybody have real, numeric data on how much additional Internet 
> traffic was generated during these events?
>
> <snip>
>
> At least from what I can tell, except for places like links to CNN, 
> overall Internet traffic does not increase significantly during 
> emergencies (or has not in the past). 

the following is just food for thought. it doesn't provide any conclusive
evidence one way or the other -- in part because I can't provide a
more detailed breakdown on smaller time scales (hourly or less).  all it
provides is single day stats from traffic to/from the states and JANET 
(the UK academic network backbone).

one will notice that there was a larger natural increase in traffic
at the start of the week versus the actual day of 9/11, which seems to
follow Henning's second comment above.

-ken
                 *************************************



Total      Mbytes    %misc TCP  %misc UDP  %FTP  %Mail  %DNS  %Web
---------------------------------------------------------------------
From US  2178476.28						      9/3 Mon
To US    1097069.29

From US  2191774.81						      9/4 Tue
To US    1259389.94

From US  2176650.90						      9/5 Wed
To US    1437022.95

From US  2169924.62						      9/6 Thu
To US    1127189.78

From US  2153254.88						      9/7 Fri
To US    1202812.76

From US  1095932.47						      9/8 Sat
To US    1027896.08

From US  932540.53    46.37      6.38      6.92    1.75  0.59   35.93 9/9 Sun
To US    980891.84    74.54      0.73      8.95    0.49  0.33   13.32

From US  2102666.86   38.79      3.52      4.65    1.63   .46   49.34 9/10 Mon
To US     900256.89   68.40       .66      7.97    1.20   .47   18.55
-------------------
From US  2261806.90   38.90      5.69      4.37    1.46   .44   47.79 9/11 Tue
To US    1265781.69   68.79      1.49      8.64    0.95   .44   14.80
-------------------
From US  2355819.24   36.48      9.53      4.18    1.55   .45   46.78 9/12 Wed
To US     974910.82   65.39      3.49     10.03    0.99   .47   13.76

From US  2339243.28   35.87      8.01      3.73    1.50   .47   49.32 9/13 Thu
To US    1270224.81   64.83      0.68      8.13    1.01   .45   16.37

From US  2194372.74   9/14 Fri
To US    1267408.09

From US  1054788.35   9/15 Sat
To US    1116076.01

From US   995756.39   9/16 Sun
To US    1110935.14

From US  2310819.52   9/17 Mon
To US    1266942.94

From US  2410681.70   9/18 Tue
To US    1326568.09

From US  2345710.97   9/19 Wed
To US    1058718.87


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 16:31:50 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA21535
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 16:31:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBALYLe29387
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 16:34:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBALYLv29384
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 16:34:21 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA21516
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 16:31:19 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBALWCv29227;
	Tue, 10 Dec 2002 16:32:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBALVmv29188
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 16:31:48 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA21447
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 16:28:45 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id C1ED667108; Tue, 10 Dec 2002 12:58:31 -0500 (EST)
Date: Tue, 10 Dec 2002 16:31:37 -0500
Subject: Re: [Ieprep] Re: architectural constraints & WG focus
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149DE@chntex02.is.dyncorp.com>
Message-Id: <C3BA9FE3-0C86-11D7-9E28-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 10, 2002, at 15:56 America/Montreal, Gunn, Janet wrote:
> [JPG] Well, you are the one who said "IP...does not have "calls"...", 
> so
> what did YOU mean.
>
> The example protocols were only intended to illustrate what I meant by
> "protocol suite", and yes, SIP is the one that most obviously has 
> "calls".

The crux of my point was:
	- routers operate only at the IP-layer
	- the IP-layer does not have the concept of a call
	- routers are what might implement special queuing of the
		set of packets that are trying to get onto an access uplink
	- routers are what might drop packets if the offered load
		is greater than the access link capacity

So unclear to me how the Internet is really capable of "prioritising 
calls".
The Internet can (within specific constraints omitted here) prioritise 
packets
(several examples were described in various earlier notes about your 
scenario example).

> [JPG] Anything that isn't "core", and might be a bottleneck/congestion
> point.

OK, but I remain confused on what sort of widget you mean by "gateway".
Is a "gateway" an IP router ?  Or is it some kind of middlebox ?
Or some kind of SIP application-layer-only widget ?

If the WG wants to focus on "approaches to using IP-layer routers to 
manage
potential issues with packet congestion on access links", then that 
sounds fine
to me.  I'm not sure if that is exactly what you intended.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 16:55:07 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA22164
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 16:55:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBALvcW30794
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 16:57:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBALvcv30791
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 16:57:38 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA22160
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 16:54:35 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBALu3v30748;
	Tue, 10 Dec 2002 16:56:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBALtnv30723
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 16:55:49 -0500
Received: from multicasttech.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA22122
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 16:52:46 -0500 (EST)
Received: from float56005.mcs.anl-external.org (account marshall_eubanks [140.221.56.5] verified)
  by multicasttech.com (CommuniGate Pro SMTP 3.4.8)
  with ESMTP-TLS id 1532238; Tue, 10 Dec 2002 16:55:42 -0500
Date: Tue, 10 Dec 2002 16:55:25 -0500
Subject: Re: [Ieprep] RE: PSTN collapse
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v482)
Cc: "King, Kimberly S." <KIMBERLY.S.KING@saic.com>,
        "'Mpierce1@aol.com '" <Mpierce1@aol.com>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
To: Henning Schulzrinne <hgs@cs.columbia.edu>
From: Marshall Eubanks <tme@multicasttech.com>
In-Reply-To: <3DF60549.9010505@cs.columbia.edu>
Message-Id: <16EEB9C2-0C8A-11D7-B5A0-003065BA697E@multicasttech.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.482)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Henning;

Did you see the NANOG-23 presentations on this ? That is the best start 
I can recommend :

http://www.nanog.org/mtg-0110/agenda.html

(The Monday morning talks -  They are all on line I believe.)

Regards
Marshall Eubanks


On Tuesday, December 10, 2002, at 10:16  AM, Henning Schulzrinne wrote:

> Does anybody have real, numeric data on how much additional Internet 
> traffic was generated during these events?
>
> There are really two separate dealing-with-collapse questions here that 
> seem to be conflated constantly:
>
> - how well does a system deal with a sudden, unplanned flood of 
> additional (emergency-induced) traffic?
>
> - how well does a system cope with partial destruction of the 
> underlying infrastructure
>
> At least from what I can tell, except for places like links to CNN, 
> overall Internet traffic does not increase significantly during 
> emergencies (or has not in the past). CNN.com and similar sites 
> obviously did "collapse" until they could be reconfigured (e.g., by 
> converting to text-only content). People send more email during 
> emergencies, but they probably worry less about ordering books from 
> Amazon, exchanging music files or playing networked games.
>
> Thus, unless somebody has verifiable proof that the Internet had, say, 
> sustained traffic several times larger than usual during those 
> emergencies, these experiences say very little should a significant 
> fraction of the existing interactive communications traffic move to the 
> Internet. After all, except for downtown, the problem in Manhattan was 
> mostly a sudden surge in volume. (Downtown, there was no Internet, 
> either...)
>
> A separate question is whether emergencies are likely to induce such 
> floods of traffic. However, I think it is more honest to separate these 
> issues.
>
>
>

T.M. Eubanks
Multicast Technologies, Inc.
10301 Democracy Lane, Suite 410
Fairfax, Virginia 22030
Phone : 703-293-9624       Fax     : 703-293-9609
e-mail : tme@multicasttech.com
http://www.multicasttech.com

Test your network for multicast :
http://www.multicasttech.com/mt/
  Status of Multicast on the Web  :
  http://www.multicasttech.com/status/index.html

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 21:49:46 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA29212
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 21:49:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBB2qJE15375
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 21:52:19 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBB2qJv15372
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 21:52:19 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA29194
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 21:49:15 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBB2oMv15272;
	Tue, 10 Dec 2002 21:50:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBB2nhv15232
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 21:49:43 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA29150
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 21:46:36 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBB2h9o12127;
	Tue, 10 Dec 2002 18:43:09 -0800 (PST)
Date: Tue, 10 Dec 2002 18:43:09 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
cc: "'Gunn, Janet'" <Janet.Gunn@DynCorp.com>,
        "'ieprep@ietf.org'" <ieprep@ietf.org>,
        "'Henry.Sinnreich@wcom.com'" <Henry.Sinnreich@wcom.com>
Subject: RE: [Ieprep] commercial SIP service
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A1C@mcl-its-exs01.mail.saic.com>
Message-ID: <Pine.GSO.4.44.0212101841470.12078-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Tue, 10 Dec 2002, King, Kimberly  S. wrote:
    > I asked "If you dial 710 now on your SIP phone, do you reach the
    > PSTN and one of the IXCs?"

That would depend on the dial-plan:

% whois -h whois.ripe.net as710

aut-num:      AS710
as-name:      BENT-NET-AS
descr:        Royal Air Base
              Bentwaters
              GB


                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 10 21:53:00 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA29277
	for <ieprep-archive@odin.ietf.org>; Tue, 10 Dec 2002 21:53:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBB2tY715490
	for ieprep-archive@odin.ietf.org; Tue, 10 Dec 2002 21:55:34 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBB2tYv15487
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 10 Dec 2002 21:55:34 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA29261
	for <ieprep-web-archive@ietf.org>; Tue, 10 Dec 2002 21:52:29 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBB2s1v15457;
	Tue, 10 Dec 2002 21:54:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBB2rUv15430
	for <ieprep@optimus.ietf.org>; Tue, 10 Dec 2002 21:53:30 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA29233
	for <ieprep@ietf.org>; Tue, 10 Dec 2002 21:50:26 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBB2kwo12149;
	Tue, 10 Dec 2002 18:46:59 -0800 (PST)
Date: Tue, 10 Dec 2002 18:46:58 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: Henning Schulzrinne <hgs@cs.columbia.edu>
cc: Marshall Eubanks <tme@multicasttech.com>,
        "King, Kimberly S." <KIMBERLY.S.KING@saic.com>,
        "'Mpierce1@aol.com '" <Mpierce1@aol.com>,
        "'ieprep@ietf.org '" <ieprep@ietf.org>
Subject: Re: [Ieprep] RE: PSTN collapse
In-Reply-To: <3DF60549.9010505@cs.columbia.edu>
Message-ID: <Pine.GSO.4.44.0212101844460.12078-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Tue, 10 Dec 2002, Henning Schulzrinne wrote:
    > At least from what I can tell, except for places like links to CNN,
    > overall Internet traffic does not increase significantly during
    > emergencies.

Correct.  Traffic does not increase during infrastructure collapse, but
capacity and diversity are typically somewhat reduced.  Rarely to the
point that any actual decreased performance could be observed by a user,
however.

What increases traffic is events like the Mars lander.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 11 06:39:21 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA01827
	for <ieprep-archive@odin.ietf.org>; Wed, 11 Dec 2002 06:39:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBBBfkg21319
	for ieprep-archive@odin.ietf.org; Wed, 11 Dec 2002 06:41:46 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBBfkv21316
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 11 Dec 2002 06:41:46 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA01822
	for <ieprep-web-archive@ietf.org>; Wed, 11 Dec 2002 06:38:50 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBBeAv21277;
	Wed, 11 Dec 2002 06:40:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBBdQv21217
	for <ieprep@optimus.ietf.org>; Wed, 11 Dec 2002 06:39:26 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA01803
	for <ieprep@ietf.org>; Wed, 11 Dec 2002 06:36:31 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Wed, 11 Dec 2002 06:39:00 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121106392132661
 for <ieprep@ietf.org>; Wed, 11 Dec 2002 06:39:21 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPJMK22>; Wed, 11 Dec 2002 06:39:08 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A37@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
Date: Wed, 11 Dec 2002 06:39:21 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] National Organization for Victim Assistance--Internet Use
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Folks,

I am asking various disaster recovery groups for information about how they
use the Internet and how we might help them.  See below for one response and
please, if for some reason you contact directly, be really nice and polite.
:-)  I have asked them for their help and they aren't signing up to be "beat
up" (not that anyone here would be rude--it's just a gentle reminder).  

Kimberly

-----Original Message-----
From: Cheryl
To: kimberly.s.king@saic.com
Sent: 12/10/2002 3:43 PM
Subject: FW: [voadnet] Response Requested

Our agency, NOVA, uses the internet during disasters for:

1.	obtaining donations (money, services, volunteers, some goods)
2.	providing information about our National Crisis Response Team
activities
3.	educational purposes, generic referrals for victims and
survivors
4.	posting messages for victims/survivors or allowing people to
express
thoughts and opinions (i.e., memorial board)

We have not used Voice option; don't know how.  Have limited IT
knowledge
and expertise among our staff.  Would use other apps if we were aware;
don't
even know what to consider as options.

Please don't hesitate to foward.  As I was thinking about this more, one 
thing that occurred to me, as NVOAD board president, that we have lots of 
agencies who could use the internet more effectively if they had ideas of 
what the possibilities were.  For example, I'd like to be able to do "crisis

intervention" via the Internet, like I've done on the radio. 

Thanks for asking,
Cheryl


Cheryl Guidry Tyiska
Director of Victim Services
National Organization for Victim Assistance
1730 Park Road, N.W.
Washington, D.C.  20010
202-232-6682 x 105
202-426-2255 fax
www.try-nova.org
cheryl@try-nova.org

Justice for All . . . Even the Victim
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 11 08:57:32 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA05287
	for <ieprep-archive@odin.ietf.org>; Wed, 11 Dec 2002 08:57:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBBDxxn28439
	for ieprep-archive@odin.ietf.org; Wed, 11 Dec 2002 08:59:59 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBDxxv28435
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 11 Dec 2002 08:59:59 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA05250
	for <ieprep-web-archive@ietf.org>; Wed, 11 Dec 2002 08:57:01 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBDwIv28373;
	Wed, 11 Dec 2002 08:58:18 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBDukv28297
	for <ieprep@optimus.ietf.org>; Wed, 11 Dec 2002 08:56:46 -0500
Received: from pfwhqs1.ncr.disa.mil (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA05156
	for <ieprep@ietf.org>; Wed, 11 Dec 2002 08:53:48 -0500 (EST)
Received: from mtahqs3.ncr.disa.mil by pfwhqs1.ncr.disa.mil
          via smtpd (for ietf-mx.ietf.org [132.151.6.1]) with SMTP; 11 Dec 2002 13:53:13 UT
Received: by mtahqs3.ncr.disa.mil with Internet Mail Service (5.5.2653.19)
	id <Y4GWD5P5>; Wed, 11 Dec 2002 08:58:21 -0500
Message-ID: <7F18415E4D63CB45BB9B3A591F68D12D5B3DDB@emshqs1.ncr.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'Bill Woodcock '" <woody@pch.net>,
        "'King, Kimberly  S. '"
	 <KIMBERLY.S.KING@saic.com>
Cc: "''Gunn, Janet' '" <Janet.Gunn@DynCorp.com>,
        "''ieprep@ietf.org' '"
	 <ieprep@ietf.org>,
        "''Henry.Sinnreich@wcom.com' '"
	 <Henry.Sinnreich@wcom.com>
Subject: RE: [Ieprep] commercial SIP service
Date: Wed, 11 Dec 2002 08:56:36 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

I think Kimberly meant the GETS area code, not the AS number.

An

-----Original Message-----
From: Bill Woodcock
To: King, Kimberly  S.
Cc: 'Gunn, Janet'; 'ieprep@ietf.org'; 'Henry.Sinnreich@wcom.com'
Sent: 12/10/02 9:43 PM
Subject: RE: [Ieprep] commercial SIP service

      On Tue, 10 Dec 2002, King, Kimberly  S. wrote:
    > I asked "If you dial 710 now on your SIP phone, do you reach the
    > PSTN and one of the IXCs?"

That would depend on the dial-plan:

% whois -h whois.ripe.net as710

aut-num:      AS710
as-name:      BENT-NET-AS
descr:        Royal Air Base
              Bentwaters
              GB


                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 11 11:22:32 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA10924
	for <ieprep-archive@odin.ietf.org>; Wed, 11 Dec 2002 11:22:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBBGP0205266
	for ieprep-archive@odin.ietf.org; Wed, 11 Dec 2002 11:25:00 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBGP0v05263
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 11 Dec 2002 11:25:00 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA10894
	for <ieprep-web-archive@ietf.org>; Wed, 11 Dec 2002 11:22:00 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBGN4v05202;
	Wed, 11 Dec 2002 11:23:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBGMKv05176
	for <ieprep@optimus.ietf.org>; Wed, 11 Dec 2002 11:22:20 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA10827
	for <ieprep@ietf.org>; Wed, 11 Dec 2002 11:19:20 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAHBDL>; Wed, 11 Dec 2002 11:19:46 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149E1@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'Nguyen, An'" <nguyena@ncs.gov>, "'Bill Woodcock '" <woody@pch.net>,
        "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>
Cc: "''ieprep@ietf.org' '" <ieprep@ietf.org>,
        "''Henry.Sinnreich@wcom.com' '" <Henry.Sinnreich@wcom.com>
Subject: RE: [Ieprep] commercial SIP service
Date: Wed, 11 Dec 2002 11:19:28 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

As in 710 xxx yyyy, in a context where a 10 digit North American Numbering
Plan number is appropriate input.

-----Original Message-----
From: Nguyen, An [mailto:nguyena@ncs.gov]
Sent: Wednesday, December 11, 2002 8:57 AM
To: 'Bill Woodcock '; 'King, Kimberly S. '
Cc: Gunn, Janet; ''ieprep@ietf.org' '; ''Henry.Sinnreich@wcom.com' '
Subject: RE: [Ieprep] commercial SIP service


I think Kimberly meant the GETS area code, not the AS number.

An

-----Original Message-----
From: Bill Woodcock
To: King, Kimberly  S.
Cc: 'Gunn, Janet'; 'ieprep@ietf.org'; 'Henry.Sinnreich@wcom.com'
Sent: 12/10/02 9:43 PM
Subject: RE: [Ieprep] commercial SIP service

      On Tue, 10 Dec 2002, King, Kimberly  S. wrote:
    > I asked "If you dial 710 now on your SIP phone, do you reach the
    > PSTN and one of the IXCs?"

That would depend on the dial-plan:

% whois -h whois.ripe.net as710

aut-num:      AS710
as-name:      BENT-NET-AS
descr:        Royal Air Base
              Bentwaters
              GB


                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 11 14:08:25 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA16729
	for <ieprep-archive@odin.ietf.org>; Wed, 11 Dec 2002 14:08:25 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBBJAsE15235
	for ieprep-archive@odin.ietf.org; Wed, 11 Dec 2002 14:10:54 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBJAsv15232
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 11 Dec 2002 14:10:54 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA16704
	for <ieprep-web-archive@ietf.org>; Wed, 11 Dec 2002 14:07:54 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBJ9Uv15180;
	Wed, 11 Dec 2002 14:09:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBBJ5Lv14433
	for <ieprep@optimus.ietf.org>; Wed, 11 Dec 2002 14:05:21 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA16579
	for <ieprep@ietf.org>; Wed, 11 Dec 2002 14:02:21 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Wed, 11 Dec 2002 14:04:43 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121114050313041
 for <ieprep@ietf.org>; Wed, 11 Dec 2002 14:05:03 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YW1T9RDG>; Wed, 11 Dec 2002 14:04:50 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A48@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
Date: Wed, 11 Dec 2002 14:04:58 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] Southern Baptist Disaster Relief Ministry--Internet Use
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Folks,

Below is another response from a disaster relief organization.  I am hopeful
that by talking to actual emergency relief organizations that we can gage
some of their concerns and how we can be of assistance.

Kimberly

________________

I will be glad to respond to your questions on behalf of the Southern
Baptist Disaster Relief Ministry.

1) How does your organization use the Internet during emergency
situations or for emergency recovery?

Southern Baptist use the Internet to post a general disaster relief
report at http://namb.net/dr during a response.  We also use our 800 ras
number to gain access to the Internet to pass email, reports and pictures
from the field to national HQ.

2) Is there a way that the Internet fails to meet your needs in these
situations?  Will you describe the symptoms of the problem? 

Our biggest problem is gaining reliable phone or data transfer service from
the disaster area to national HQ.  Once reliable phone service is
established we are able to transfer information directly.

3) Do you envision using Voice over IP applications in such situations?
(For example, Microsoft NetMeeting can support audio conversations between
people like a traditional telephone call except using their computers with
microphones and speakers) 

I could see a use for this kind of applications especially for conference
calls or NOVAD meetings which could take place on the web.  However, the
problem is again having the reliable ISP from the field

4) Have you already tried such voice applications over the Internet to
aid emergency recovery? 

No, but I would be interested in trying.


5) In other words, how would you like to use the Internet during an
emergency or how do you think it could be of assistance?

I would like to use the Internet to help with live reporting and data
collection.  If reliable ISP could be provided from the field then we
would be interested in a web based application for disaster management and
reporting.  However, since during a disaster we do not have a reliable
ISP from the field we have not attempted a web based application.


Because HE Lives,
Joel W. Phillips
NAMB DR Off-Site Coordinator <//><
jophillips@namb.net <mailto:jophillips@namb.net> 
800-462-8657 Ext. 6133
 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 11 20:07:25 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA27968
	for <ieprep-archive@odin.ietf.org>; Wed, 11 Dec 2002 20:07:25 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBC19vG01936
	for ieprep-archive@odin.ietf.org; Wed, 11 Dec 2002 20:09:57 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBC19vv01933
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 11 Dec 2002 20:09:57 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA27942
	for <ieprep-web-archive@ietf.org>; Wed, 11 Dec 2002 20:06:54 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBC189v01879;
	Wed, 11 Dec 2002 20:08:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBC17ov01864
	for <ieprep@optimus.ietf.org>; Wed, 11 Dec 2002 20:07:50 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA27910
	for <ieprep@ietf.org>; Wed, 11 Dec 2002 20:04:47 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBC11Do21317;
	Wed, 11 Dec 2002 17:01:13 -0800 (PST)
Date: Wed, 11 Dec 2002 17:01:13 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "Nguyen, An" <nguyena@ncs.gov>
cc: "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>,
        "''Gunn, Janet' '" <Janet.Gunn@DynCorp.com>,
        "''ieprep@ietf.org' '" <ieprep@ietf.org>,
        "''Henry.Sinnreich@wcom.com' '" <Henry.Sinnreich@wcom.com>
Subject: RE: [Ieprep] commercial SIP service
In-Reply-To: <7F18415E4D63CB45BB9B3A591F68D12D5B3DDB@emshqs1.ncr.disa.mil>
Message-ID: <Pine.GSO.4.44.0212111700070.21216-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Wed, 11 Dec 2002, Nguyen, An wrote:
    > I think Kimberly meant the GETS area code, not the AS number.

And I repeat...  Dialing that would produce different results (GETS, RAF,
or something else entirely) depending what dial-plan you were using.  That
was exactly my point.

    > That would depend on the dial-plan:
    >
    > % whois -h whois.ripe.net as710
    >
    > aut-num:      AS710
    > as-name:      BENT-NET-AS
    > descr:        Royal Air Base
    >               Bentwaters
    >               GB
    >
    >
    >                                 -Bill
    >
    >
    > _______________________________________________
    > Ieprep mailing list
    > Ieprep@ietf.org
    > https://www1.ietf.org/mailman/listinfo/ieprep
    > _______________________________________________
    > Ieprep mailing list
    > Ieprep@ietf.org
    > https://www1.ietf.org/mailman/listinfo/ieprep
    >


                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 03:38:12 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA16760
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 03:38:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBC8emE02493
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 03:40:48 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBC8elv02490
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 03:40:47 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA16750
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 03:37:40 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBC8cQv02343;
	Thu, 12 Dec 2002 03:38:26 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBC8bev01858
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 03:37:43 -0500
Received: from kc-msxproto4.kc.umkc.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA16676
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 03:34:24 -0500 (EST)
Received: from KC-MAIL2.kc.umkc.edu ([134.193.143.162] RDNS failed) by kc-msxproto4.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.5329);
	 Thu, 12 Dec 2002 02:37:16 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Date: Thu, 12 Dec 2002 02:37:15 -0600
Message-ID: <A29143A40AEAE3459FF829D1DD43316101D9B091@KC-MAIL2.kc.umkc.edu>
Thread-Topic: Guam - Typhoon 180 MPH winds
Thread-Index: AcKhq/UmBik7NvSKQrStkT+xjEGhlwAAgwkA
From: "Ayyasamy, Senthilkumar  (UMKC-Student)" <saq66@umkc.edu>
To: <ieprep@ietf.org>
X-OriginalArrivalTime: 12 Dec 2002 08:37:16.0138 (UTC) FILETIME=[ACE314A0:01C2A1B9]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBC8bxv01952
Subject: [Ieprep] real problem: sequel to emergency response team feedback
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


Sean Donelan post to nanog list on Dec 09, 2002 
> Guam was struck by Typhoon Pongona with sustanted winds of 
> 150 mph, and
> gusts of 184 mph.  Electricity is out on the entire island.  Most
> telephone service is down, although some calls have been 
> received.  The
> University of Guam, the the ccTLD manager for .GU, is offline.  Backup
> name service for .GU is being provided by NS.RIPE.NET, 
> although I haven't
> found working Internet service to the island.


Sean Donelan post to nanog list on Dec 12, 2002 
> The local telephone company has restored partial commercial service.
> Commercial electric service is still down. The cable landing 
> stations and
> networks with direct connections remained available throughout the
> typhoon.  Other locations lost service when intermediate transmission
> facilities on the island lost power or were destroyed.  University of
> Guam's systems stayed up throughout the typhon, but lost connectivity
> through their upstream provider. Most locations are still operating on
> backup power.  A fire at the island's fuel storage farm has caused the
> authorities to restrict access to fuel supplies.

A quick browse through http://news.google.com/ gave some facts:

- Like 9/11, power management was the biggest problem and some loss in
  Business establishments.

- Ham Radio users were wonderful source of information (like any other emergency
  situation.)

- As Govt declared it as state emergency, FEMA/PHS/USACE was in service.

- http://www.computerworld.com/securitytopics/security/recovery/story/0,10801,76559,00.html
  "WorldCom Inc., which provides some of the cable connections that link Guam to the 
  outside world, was ready to go as soon as the local telephone provider re-energized 
  its lines.
  ....
  Tony Das, managing director of Startec+PCI, a cellular and Internet service provider, 
  said it may be weeks before power is restored. "I told my chief engineer here to make 
  sure the generator is in good enough condition to last at least two months if 
  necessary," he said. 

  Startec maintained its connectivity with the island's two main cable stations and is 
  supplying Internet services to its customers who still have working lines. For those 
  who can't connect from their homes or offices, Das said he plans to set up computers 
  and telephones at his offices and invite customers to use them"

-  http://triangle.bizjournals.com/triangle/stories/2002/12/09/daily23.html
   "TWR operates five shortwave transmitters on Guam. The antenna array suffered severe damage, 
   although all the towers remained erect and apparently undamaged. Three of the five antenna 
   curtains were "shredded," according to TWR's staff on the island. 
   
   The other two were damaged, but the staff hopes to be able to repair and re-install 
   them later this week."


- http://www.guampdn.com/news/stories/20021212/localnews/553616.html
  "All University of Guam students and faculty members must meet during their scheduled exam 
  times during exam week, Dec. 16 to 18, to complete the semester, according to a UOG press 
  release. But all classes scheduled this week have been canceled. "
   But , even during typhoon students have to work.

  


Kimberly said:
> Fred was absolutely right in discouraging discussions of  solutions 
> space  before requirements.  However, something even more  fundamental
> is missing--a  concise statement of the problem to be solved.  
> 
> Over two *years* ago, Sean Donelan asked some probing questions.
> ( http://www.iepscheme.net/archive/0045.html )
> "What problems have emergency service users actually had using the
> Internet?"
> .....
> In order to discuss requirements, there has to be a crisp 
> definition of the  problem to be solved.

A 10 minutes browse through the news articles listed the following as various emergency
rescue teams deployed at guam: "Red cross, FEMA, Puublic Health services, memorial 
hospital volunteers,Guam Salvation Army, US Navy, GEPA, GETS(?)" (might be more....)

Lets try to define the problem here: 
 How can Internet help to co-ordinate between these emergeny response teams? 

Points to be considered:

* Cell phones, PDA and IP phones will be very useful in a place like 
  guam where their is no complete Internet coverage. 

* IM, email service will be very useful. But, this requires no special work 
  in ieprep.

* Service like IAA will be very useful. But, we have to look into security issues..
  DoS attacks on such databases.

* we need to have wireless LAN to help local emergency response operation.

* more ...


- How can Internet actually help ieprep community in a place where their is no/minimal
  Internet connectivity? 

- What does the emergency response community expect from IETF in situation like this?
  ( Kimberly can help in getting feedback from the emergency response community)



Note: Please limit Deep solutions discussion for this one. If you want to get into deep
solution discussion, change the subject likely.Just try to define the actual 
problem here and what IETF can do to help emergency response community? 













_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 07:31:21 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA20762
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 07:31:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCCXlr15282
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 07:33:47 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCCXlv15279
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 07:33:47 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA20746
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 07:30:50 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCCW4v15246;
	Thu, 12 Dec 2002 07:32:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCCVTv15221
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 07:31:29 -0500
Received: from bells.cs.ucl.ac.uk (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id HAA20705
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 07:28:32 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.29183-0@bells.cs.ucl.ac.uk>; Thu, 12 Dec 2002 12:30:45 +0000
To: ieprep@ietf.org
Date: Thu, 12 Dec 2002 12:30:42 +0000
Message-ID: <21639.1039696242@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Subject: [Ieprep] updated requirements documents
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


an updated version of the draft-carlberg-ets-*-01.txt requirements
documents have been placed in the I-D repositories.  they reflect
comments/responses discussed about the -00.txt version.

an administrative objective of the authors is to move these documents to
official working group items.  if there are concerns about this, please
let the group and/or chairs know.

regards,

-ken

http://www.ietf.org/internet-drafts/draft-carlberg-ets-general-01.txt
http://www.ietf.org/internet-drafts/draft-carlberg-ets-telephony-01.txt

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 10:28:23 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA27304
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 10:28:23 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCFUqD26061
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 10:30:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCFUqv26058
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 10:30:52 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA27270
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 10:27:52 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCFQ6v25798;
	Thu, 12 Dec 2002 10:26:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCFPmv25766
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 10:25:48 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA27077
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 10:22:48 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAHDSS>; Thu, 12 Dec 2002 10:23:14 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149E6@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'Bill Woodcock'" <woody@pch.net>, "Nguyen, An" <nguyena@ncs.gov>
Cc: "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>,
        "''ieprep@ietf.org' '" <ieprep@ietf.org>,
        "''Henry.Sinnreich@wcom.com' '"
	 <Henry.Sinnreich@wcom.com>
Subject: RE: [Ieprep] commercial SIP service
Date: Thu, 12 Dec 2002 10:23:04 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

And I repeat...
In a context where the NORTH AMERICAN NUMBERING PLAN (that sure sounds like
a "dial-plan" to me) is expected.

We all recognize that if you dial 710 xxx yyyy in a context that is
expecting SOMETHING OTHER THAN THE NORTH AMERICAN NUMBERING PLAN, you aren't
going to get the GETS behavior,  That is true in the PSTN too.

-----Original Message-----
From: Bill Woodcock [mailto:woody@pch.net]
Sent: Wednesday, December 11, 2002 8:01 PM
To: Nguyen, An
Cc: 'King, Kimberly S. '; Gunn, Janet; ''ieprep@ietf.org' ';
''Henry.Sinnreich@wcom.com' '
Subject: RE: [Ieprep] commercial SIP service


      On Wed, 11 Dec 2002, Nguyen, An wrote:
    > I think Kimberly meant the GETS area code, not the AS number.

And I repeat...  Dialing that would produce different results (GETS, RAF,
or something else entirely) depending what dial-plan you were using.  That
was exactly my point.

    > That would depend on the dial-plan:
    >
    > % whois -h whois.ripe.net as710
    >
    > aut-num:      AS710
    > as-name:      BENT-NET-AS
    > descr:        Royal Air Base
    >               Bentwaters
    >               GB
    >
    >
    >                                 -Bill
    >
    >
    > _______________________________________________
    > Ieprep mailing list
    > Ieprep@ietf.org
    > https://www1.ietf.org/mailman/listinfo/ieprep
    > _______________________________________________
    > Ieprep mailing list
    > Ieprep@ietf.org
    > https://www1.ietf.org/mailman/listinfo/ieprep
    >


                                -Bill

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 11:14:03 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA29347
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 11:14:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCGGX729595
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 11:16:33 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCGGXv29592
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 11:16:33 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA29321
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 11:13:32 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCGE3v29432;
	Thu, 12 Dec 2002 11:14:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCGDbv29371
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 11:13:37 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA29206
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 11:10:37 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Thu, 12 Dec 2002 11:12:57 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121211131800418
 for <ieprep@ietf.org>; Thu, 12 Dec 2002 11:13:18 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YW14BAM7>; Thu, 12 Dec 2002 11:13:05 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A50@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "Ieprep (E-mail)" <ieprep@ietf.org>
Date: Thu, 12 Dec 2002 11:13:17 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] Web-based logistics software planned for Red Cross and Red Crescent Societies
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

FYI: Example Internet use in support of emergency relief operations

"A Fritz Institute team of logistics and software experts is at work at the
Federation Secretariat in Geneva to tailor the software, which will enable
the organization and its 178 member Red Cross and Red Crescent Societies to
provide unparalleled support to emergency relief operations anywhere in the
world.

"What we are getting is the very latest in logistics technology, a Web-based
system that will take our current processes and connect all steps of the
logistics chain, from the launch of an emergency appeal through procurement,
donations, warehousing and distribution to disaster victims," said Didier
Cherpitel, Federation secretary general."

more at http://www.ifrc.org/docs/news/02/052301/ and 
http://www.fritzinstitute.org/index.html
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 13:27:11 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA05835
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 13:27:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCITeu06009
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 13:29:40 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCITev06006
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 13:29:40 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA05825
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 13:26:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCIR4v05909;
	Thu, 12 Dec 2002 13:27:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCIQ9v05819
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 13:26:09 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA05697
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 13:23:08 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP id A831867106
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 13:26:07 -0500 (EST)
Date: Thu, 12 Dec 2002 13:26:02 -0500
Mime-Version: 1.0 (Apple Message framework v548)
Content-Type: text/plain; delsp=yes; charset=US-ASCII; format=flowed
From: RJ Atkinson <rja@extremenetworks.com>
To: ieprep <ieprep@ietf.org>
Content-Transfer-Encoding: 7bit
Message-Id: <2B39E360-0DFF-11D7-8C00-00039357A82A@extremenetworks.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] An alternative concept of IEprep
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


	Personally, I find the Japanese deployment of IEprep to be very
interesting.  Trying to take their concept and generalise it a bit,
here is an alternative (non-voice) concept for IEprep for discussion.

Ran
rja@extremenetworks.com

------------------------------------------------------------------------ 
-

OBJECTIVE:
	Provide enhanced ability for emergency workers to communicate
among themselves during an (unplanned) emergency.

OPERATING ENVIRONMENT:
	This service should work from a laptop deployed in the field,
that would be connected via dialup PSTN, wireless, Ethernet, or
whatever other connectivity existed in the field.

	This service should work equally well for an emergency service
worker who happens across an Internet cafe that offers access to
the world wide web.


APPLICATIONS:
	Only standard Internet applications protocols would be used.  The
primary applications would be all be based around HTTP and TLS [1].
The applications would be chosen in part for their ability to operate
robustly in challenging environments -- meaning that the applications
should still be able to work just fine in circumstances of packet loss
on access links, variable latency and jitter, and so forth.  The  
applications
should also all be well-behaved from a congestion avoidance and control
perspective.

	Specific supported applications would include web-based email, both
secure web sites (using TLS and login/password) and non-secure web sites
(e.g. http://news.bbc.co.uk), Internet Relay Chat (IRC), Instant  
Messaging (IM),
and possibly also tailored web-based applications.

	Web-mail could be used for nearly any sort of communications,
including requests, situation reports, and so forth.  For improved
availability, users that would predictably need access during an
emergency (e.g. US FEMA staff, Red Cross, local emergency services)
could obtain an account on a dedicated webmail cluster that is built
in a high-availability and distributed manner.  The URLs, login id,
and password could all be pre-printed on a wallet-sized card, for  
example.

	The secure web sites could be used for uploading/downloading various
kinds of documents (e.g. situation reports, requests for equipment or
other assistance, standard operating procedures, lists of victims,
lists of known survivors, etc.).  There might be  
pre-planned/pre-deployed
IEprep web server clusters (similar to the above webmail cluster  
concept),
in addition to using whatever other servers were suitable in an ad hoc
manner.

DEPLOYMENT CONCEPT:

	While dedicated servers are not required for this concept to work well
and be very useful, dedicated servers could help improve service  
availability
and reduce communications latency.

	A good strategy for the servers would be to build several  
high-availability
server clusters.  High-availability meaning that if a component failed,  
the
system could quickly recover and the non-failed components would  
continue to
work together, though possibility with diminished storage, I/O, or CPU
capabilities.  HA server clusters are widely available COTS products,  
though
they are often not cheap.

	Clusters would be placed in ISP co-location facilities, preferably in  
the
same bunker-style facilities used for the ISP's core routers.  The goal  
would
be to have at least one server cluster inside each large ISP's core.   
Contracts
could be signed with the ISPs to give service restoration to those  
clusters
priority -- and if co-located in the ISP's core router POPs, there is  
high
probability that the ISP could get technical staff on site quickly if  
needed.
Being inside an ISP's core router POP also helps ensure service  
availability,
because the server cluster could easily have redundant connectivity to  
multiple
core routers.

	In addition, the operator of the server clusters would endeavor to get  
the
ISPs to agree that service restoration of links among the contracted  
ISPs
would get preference in the event of a fibre cut (or similar).

	Each server cluster would be configured to automatically remain in sync
with each other server cluster.  So that if one lost any single  
cluster, there
would be a high probability that most of the data on that one cluster  
would
also be replicated on some other cluster.

	Applications would need to be deployed such that login to any cluster
would be approximately equal to logging into any other cluster.  This  
way,
users need only login to the nearest or most reachable cluster.

	Anycast might or might not be useful for addressing the servers.  An
alternative to anycast is to list the URL for each cluster on a  
wallet-sized
card, so that users could try each one in turn from the field.

SECURITY

	These server clusters would be a high-value target.  So they should
probably have high-grade IP encryptors at each cluster to provide secure
connectivity to other clusters.  If operated by a government, one could
imagine using a military-grade IPsec encryptor [2] or a formally  
evaluated
commercial-grade IPsec encryptor.  The server software (OS and  
applications)
should be hardened.  The servers should have a high-availability system
configuration, including redundant hardware and HA-capable OS software.
The web services might well use X.509v3 certificates (or equivalent) to
provide strong authentication of identity.  The web services should use
strong cryptography (e.g. 3DES-CBC + HMAC SHA-1 or better) with their
TLS implementation.  Login database data should be replicated at each
site rather than centralised in order to improve availability of login
services.  Disk systems should be RAID with more capacity than one  
thinks
is needed in the worst case (disk is cheap).  The systems should  
probably
be behind some sort of firewall device.  The servers ought NOT be behind
a NAT because that makes IPsec and other security applications difficult
or impossible to deploy

NOTES:

[1] IETF's Transport-Layer Security (TLS) specification is the primary  
standard
	for security with HTTP.  TLS is derived from SSL 3.0, but contains a  
variety
	of architectural and technical fixes, including more solid security.
[2] For data rates up to 1 Gbps, these either exist GOTS/COTS today or  
are going
	to be available near term.  For data rates up to 10 Gbps, one might  
need
	to wait up to 24 months.  If the speed were critical, one could use  
evaluated
	commercial-grade products on an interim basis.  The real trick is  
having a
	solid key management plan, of course.

------------------------------------------------------------------------ 
------------

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 13:42:40 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06482
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 13:42:40 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCIjAQ07443
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 13:45:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCIjAv07440
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 13:45:10 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06454
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 13:42:09 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCIg1v07362;
	Thu, 12 Dec 2002 13:42:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCIeNv07255
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 13:40:23 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06290
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 13:37:21 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP id 00B9367106
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 13:40:09 -0500 (EST)
Date: Thu, 12 Dec 2002 13:40:01 -0500
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
From: RJ Atkinson <rja@extremenetworks.com>
To: ieprep <ieprep@ietf.org>
Content-Transfer-Encoding: 7bit
In-Reply-To: <148.445fa65.2b2244d2@aol.com>
Message-Id: <1F42AE93-0E01-11D7-8C00-00039357A82A@extremenetworks.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] Re: Mike's scenario
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Dec 6, 2002, at 13:22 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] I'll try one more time. The question presumes the following 
> situation:
>
> 1. There is a government agency that "authorizes" indiviuduals to get 
> priority treatment. (How or why is not the issue here)
> 2. There is a service provider who has agreed by contract to support 
> this capabilitity. The service provider does not "authorize" anyone, 
> but presume that they must "authenticate" its use. (Whether they are 
> "paid" to do this is not the issue.)
> 3. There is a single 384 kbit/s path that must carry all the traffic 
> from the area (the redundant paths may have been taken out by the 
> terrorist act). There is no other "PVC" to route traffic onto. (Why 
> the service provider didn't have sufficient backup is not the issue.)
> 4. There is excessive congestion on the 384 kbit/s path, caused by too 
> many voice calls, all of which are getting unacceptable quality. (Any 
> data or elastic traffic has already been squeezed out.)
>
> If you don't accept the above as the case being discussed, please do 
> not respond.

	The scenario you describe does not sound realistic for IEprep.
In particular, the assumptions do not all seem reasonable to me (and
some few other folks here).

> How is the 1 authorized individual able to get their call through when 
> the link is congested by users without the special authorization from 
> the government?

	Authorisation from NCS (or whoever) is substantially less relevant 
than authorisation
from the organisation that is paying for the access link.  In 
particular, if it
is not NCS's access link on NCS's budget, then NCS does NOT get to set 
the policies
for that link.

	Instead, the end-user organisation who is paying for the access link 
is the
only organisation with authority to set policy for those links.  The 
ISP does NOT
have that authority since it is not their link, so a NCS contract with 
the ISP is
NOT sufficient or terribly relevant.  In this regard, consider that an 
IP access link
is very much like last-mile PSTN to a home.  If there is an emergency, 
that does not
mean that some stranger gets to use one's home telephone unless one 
says they can.
(And I would not suggest that the stranger try to force the issue in 
places like Texas,
where there are decent chances the homeowner is well armed.)

	Note well that ISPs are NOT "common carriers" in the US and in most 
other
countries, unlike PSTN operators.

	For the person who has authorisation from *the end-user organisation 
that is
paying for the access link*, that organisation can configure routers, 
switches,
and other devices to provide such preferential treatment (e.g. 
Class-based queuing
in the routers, switches, or other devices).  This can be configured in 
a variety
of ways, including access-lists, filters, and other 
implementation-specific methods.
A variety of implementation-specific, link-layer, and intra-domain QoS 
methods
(several of which have been described on this list in the past) can be 
used.

	Please note the absence of inter-domain QoS as part of the solution.

Ran
rja@extremenetworks.com


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 13:45:58 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06631
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 13:45:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCImS607564
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 13:48:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCImSv07561
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 13:48:28 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06595
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 13:45:27 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCIlCv07512;
	Thu, 12 Dec 2002 13:47:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCIk4v07467
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 13:46:04 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA06505
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 13:43:04 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAH1ST>; Thu, 12 Dec 2002 13:43:28 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149E8@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>,
        "'Bill Woodcock'"
	 <woody@pch.net>, "Nguyen, An" <nguyena@ncs.gov>
Cc: "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>,
        "''ieprep@ietf.org' '" <ieprep@ietf.org>,
        "''Henry.Sinnreich@wcom.com' '"
	 <Henry.Sinnreich@wcom.com>
Subject: RE: [Ieprep] commercial SIP service
Date: Thu, 12 Dec 2002 13:43:17 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

I have been corrected in my terminology.

In the context of a dial-plan which expects to se a NANP number, and at the
place in the dial-plan which expects to see a NANP number.

-----Original Message-----
From: Gunn, Janet 
Sent: Thursday, December 12, 2002 10:23 AM
To: 'Bill Woodcock'; Nguyen, An
Cc: 'King, Kimberly S. '; ''ieprep@ietf.org' ';
''Henry.Sinnreich@wcom.com' '
Subject: RE: [Ieprep] commercial SIP service


And I repeat...
In a context where the NORTH AMERICAN NUMBERING PLAN (that sure sounds like
a "dial-plan" to me) is expected.

We all recognize that if you dial 710 xxx yyyy in a context that is
expecting SOMETHING OTHER THAN THE NORTH AMERICAN NUMBERING PLAN, you aren't
going to get the GETS behavior,  That is true in the PSTN too.

-----Original Message-----
From: Bill Woodcock [mailto:woody@pch.net]
Sent: Wednesday, December 11, 2002 8:01 PM
To: Nguyen, An
Cc: 'King, Kimberly S. '; Gunn, Janet; ''ieprep@ietf.org' ';
''Henry.Sinnreich@wcom.com' '
Subject: RE: [Ieprep] commercial SIP service


      On Wed, 11 Dec 2002, Nguyen, An wrote:
    > I think Kimberly meant the GETS area code, not the AS number.

And I repeat...  Dialing that would produce different results (GETS, RAF,
or something else entirely) depending what dial-plan you were using.  That
was exactly my point.

    > That would depend on the dial-plan:
    >
    > % whois -h whois.ripe.net as710
    >
    > aut-num:      AS710
    > as-name:      BENT-NET-AS
    > descr:        Royal Air Base
    >               Bentwaters
    >               GB
    >
    >
    >                                 -Bill
    >
    >
    > _______________________________________________
    > Ieprep mailing list
    > Ieprep@ietf.org
    > https://www1.ietf.org/mailman/listinfo/ieprep
    > _______________________________________________
    > Ieprep mailing list
    > Ieprep@ietf.org
    > https://www1.ietf.org/mailman/listinfo/ieprep
    >


                                -Bill

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 14:37:20 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA09302
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 14:37:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCJdpw11351
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 14:39:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCJdpv11348
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 14:39:51 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA09275
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 14:36:49 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCJc6v11209;
	Thu, 12 Dec 2002 14:38:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCJb4v10652
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 14:37:04 -0500
Received: from halt-in.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA09134
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 14:34:02 -0500 (EST)
Received: from cisco.com (10.32.253.236)
  by halt-in.cisco.com with SMTP; 12 Dec 2002 11:38:02 +0000
Message-Id: <1qcd67$stm9@halt-in.cisco.com>
Message-Id: <5.2.0.9.2.20021212112328.03832528@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.2.0.9
Date: Thu, 12 Dec 2002 11:33:29 -0800
To: RJ Atkinson <rja@extremenetworks.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Re: Mike's scenario
Cc: ieprep <ieprep@ietf.org>
In-Reply-To: <1F42AE93-0E01-11D7-8C00-00039357A82A@extremenetworks.com>
References: <148.445fa65.2b2244d2@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

At 01:40 PM 12/12/2002 -0500, RJ Atkinson wrote:

>On Friday, Dec 6, 2002, at 13:22 America/Montreal, Mpierce1@aol.com wrote:
>>[MAP] I'll try one more time. The question presumes the following situation:
>>
>>1. There is a government agency that "authorizes" indiviuduals to get 
>>priority treatment. (How or why is not the issue here)
>>2. There is a service provider who has agreed by contract to support this 
>>capabilitity. The service provider does not "authorize" anyone, but 
>>presume that they must "authenticate" its use. (Whether they are "paid" 
>>to do this is not the issue.)
>>3. There is a single 384 kbit/s path that must carry all the traffic from 
>>the area (the redundant paths may have been taken out by the terrorist 
>>act). There is no other "PVC" to route traffic onto. (Why the service 
>>provider didn't have sufficient backup is not the issue.)
>>4. There is excessive congestion on the 384 kbit/s path, caused by too 
>>many voice calls, all of which are getting unacceptable quality. (Any 
>>data or elastic traffic has already been squeezed out.)
>>
>>If you don't accept the above as the case being discussed, please do not 
>>respond.
>
>         The scenario you describe does not sound realistic for IEprep.

I'm not sure what you're arguing, Ran.

Are you arguing that it is unlikely that a service like GETS might be 
deployed using VoIP technology? If so, you are arguing that the stated 
plans of known government agencies are irrelevant.

Are you arguing that a service provider, perhaps one owned by a government 
agency and perhaps a public one, would choose to not write a contract for 
the capability? Every service provider I know of is killing itself to get 
new revenue.

Are you arguing that there would never be a situation where there is a 
stated bandwidth which is less than infinite when compared to the load that 
might be placed on it at a point in time? If so, a great number of people 
in the Internet will be happy to know that you have decided to fund the 
increase in their available bandwidth.

>         Authorisation from NCS (or whoever) is substantially less 
> relevant than authorisation
>from the organisation that is paying for the access link.  In particular, 
>if it
>is not NCS's access link on NCS's budget, then NCS does NOT get to set the 
>policies
>for that link.

Wonderful, Ran. The point is that the service on the link would have to 
intersect *two* policies: there has to be local service authorization (try 
placing a GETS call from a phone whse owner doesn't pay his bill; I'll bet 
it doesn't work at all), and in addition it needs to have the relevant 
governmental policy authorization.

To be very frank, your response sounds obstructionist. The items Mike put 
forward represent the situation envisioned in the charter of this working 
group. If you don't like the charter, go to the AD. It is chartered; it 
would be nice if you would allow it to be discussed.
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 14:59:39 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA10224
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 14:59:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCK2AP12338
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 15:02:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCK2Av12335
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 15:02:10 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA10213
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 14:59:08 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCK02v12211;
	Thu, 12 Dec 2002 15:00:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCJvjv12096
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 14:57:45 -0500
Received: from halt-in.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA10006
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 14:54:43 -0500 (EST)
Received: from cisco.com (10.32.253.236)
  by halt-in.cisco.com with SMTP; 12 Dec 2002 11:58:44 +0000
Message-Id: <1qcd67$t233@halt-in.cisco.com>
Message-Id: <5.2.0.9.2.20021212114409.082ff508@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.2.0.9
Date: Thu, 12 Dec 2002 11:44:59 -0800
To: Mpierce1@aol.com
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Re: PSTN collapse
Cc: ieprep@ietf.org
In-Reply-To: <185.1339863f.2b276b1f@aol.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

At 11:06 AM 12/10/2002 -0500, Mpierce1@aol.com wrote:
>(How many Internet routers have generators anyway?)

most of the ones in ISP POPs do. They get pretty serious about that. They 
like revenue. 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 15:04:53 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA10455
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 15:04:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCK7OD13099
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 15:07:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCK7Ov13096
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 15:07:24 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA10441
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 15:04:22 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCK63v12495;
	Thu, 12 Dec 2002 15:06:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCK5Tv12456
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 15:05:29 -0500
Received: from sith.maoz.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA10353
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 15:02:27 -0500 (EST)
Received: (from dmm@localhost)
	by sith.maoz.com (8.12.6/8.12.6) id gBCK5ukn027526;
	Thu, 12 Dec 2002 12:05:56 -0800
Date: Thu, 12 Dec 2002 12:05:56 -0800
From: David Meyer <dmm@sprint.net>
To: Fred Baker <fred@cisco.com>
Cc: Mpierce1@aol.com, ieprep@ietf.org
Subject: Re: [Ieprep] Re: PSTN collapse
Message-ID: <20021212120556.A27516@sprint.net>
References: <185.1339863f.2b276b1f@aol.com> <5.2.0.9.2.20021212114409.082ff508@mira-sjcm-4.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.2.5.1i
In-Reply-To: <5.2.0.9.2.20021212114409.082ff508@mira-sjcm-4.cisco.com>; from fred@cisco.com on Thu, Dec 12, 2002 at 11:44:59AM -0800
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

>> At 11:06 AM 12/10/2002 -0500, Mpierce1@aol.com wrote:
>> >(How many Internet routers have generators anyway?)
>> 
>> most of the ones in ISP POPs do. They get pretty serious about that. They 
>> like revenue. 

I would guess that most ISPs of any size have redundance power
(from redundant grids, with redundant entrance), as well as other
backup power (e.g. diesel generator); heck, UO even has that
much. In addition, frequently the entire HVAC systems are
redundant, so the change of this kind of failure is pretty
minimized (again, at least for the carriers that I know about).

Dave
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 15:09:42 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA10737
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 15:09:42 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCKCCg13326
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 15:12:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCKCCv13323
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 15:12:12 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA10718
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 15:09:10 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCJtCv11992;
	Thu, 12 Dec 2002 14:55:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCJsdv11954
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 14:54:39 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA09907
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 14:51:37 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAH19A>; Thu, 12 Dec 2002 14:52:02 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149EB@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>, ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: Mike's scenario
Date: Thu, 12 Dec 2002 14:51:50 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

OK, I'll add another assumption
Assume that "organisation that is paying for the access link" authorizes the
user.
Or assume that "an authorizing body " (NCS or otherwise) has an agreement
(in advance) with the "organisation that is paying for the access link" to
recognize that authorization. (YES- I KNOW that to be  PRACTICAL/USEFUL you
would need a LARGE NUMBER of such agreements.)

-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Thursday, December 12, 2002 1:40 PM
To: ieprep
Subject: [Ieprep] Re: Mike's scenario



On Friday, Dec 6, 2002, at 13:22 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] I'll try one more time. The question presumes the following 
> situation:
>
> 1. There is a government agency that "authorizes" indiviuduals to get 
> priority treatment. (How or why is not the issue here)
> 2. There is a service provider who has agreed by contract to support 
> this capabilitity. The service provider does not "authorize" anyone, 
> but presume that they must "authenticate" its use. (Whether they are 
> "paid" to do this is not the issue.)
> 3. There is a single 384 kbit/s path that must carry all the traffic 
> from the area (the redundant paths may have been taken out by the 
> terrorist act). There is no other "PVC" to route traffic onto. (Why 
> the service provider didn't have sufficient backup is not the issue.)
> 4. There is excessive congestion on the 384 kbit/s path, caused by too 
> many voice calls, all of which are getting unacceptable quality. (Any 
> data or elastic traffic has already been squeezed out.)
>
> If you don't accept the above as the case being discussed, please do 
> not respond.

	The scenario you describe does not sound realistic for IEprep.
In particular, the assumptions do not all seem reasonable to me (and
some few other folks here).

> How is the 1 authorized individual able to get their call through when 
> the link is congested by users without the special authorization from 
> the government?

	Authorisation from NCS (or whoever) is substantially less relevant 
than authorisation
from the organisation that is paying for the access link.  In 
particular, if it
is not NCS's access link on NCS's budget, then NCS does NOT get to set 
the policies
for that link.

	Instead, the end-user organisation who is paying for the access link

is the
only organisation with authority to set policy for those links.  The 
ISP does NOT
have that authority since it is not their link, so a NCS contract with 
the ISP is
NOT sufficient or terribly relevant.  In this regard, consider that an 
IP access link
is very much like last-mile PSTN to a home.  If there is an emergency, 
that does not
mean that some stranger gets to use one's home telephone unless one 
says they can.
(And I would not suggest that the stranger try to force the issue in 
places like Texas,
where there are decent chances the homeowner is well armed.)

	Note well that ISPs are NOT "common carriers" in the US and in most 
other
countries, unlike PSTN operators.

	For the person who has authorisation from *the end-user organisation

that is
paying for the access link*, that organisation can configure routers, 
switches,
and other devices to provide such preferential treatment (e.g. 
Class-based queuing
in the routers, switches, or other devices).  This can be configured in 
a variety
of ways, including access-lists, filters, and other 
implementation-specific methods.
A variety of implementation-specific, link-layer, and intra-domain QoS 
methods
(several of which have been described on this list in the past) can be 
used.

	Please note the absence of inter-domain QoS as part of the solution.

Ran
rja@extremenetworks.com


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 16:33:19 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA13664
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 16:33:19 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCLZpL17497
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 16:35:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLZpv17494
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 16:35:51 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA13661
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 16:32:48 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLY1v17401;
	Thu, 12 Dec 2002 16:34:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLXIv17385
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 16:33:18 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA13547
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 16:30:15 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 5E53F67106; Thu, 12 Dec 2002 16:33:18 -0500 (EST)
Date: Thu, 12 Dec 2002 16:33:11 -0500
Subject: Re: [Ieprep] Re: Mike's scenario
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149EB@chntex02.is.dyncorp.com>
Message-Id: <503070DE-0E19-11D7-8C00-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Thursday, Dec 12, 2002, at 14:51 America/Montreal, Gunn, Janet wrote:
> OK, I'll add another assumption
> Assume that "organisation that is paying for the access link" 
> authorizes the
> user.
> Or assume that "an authorizing body " (NCS or otherwise) has an 
> agreement
> (in advance) with the "organisation that is paying for the access 
> link" to
> recognize that authorization. (YES- I KNOW that to be  
> PRACTICAL/USEFUL you
> would need a LARGE NUMBER of such agreements.)

Thanks.  You read more carefully than Fred, and its appreciated.

The answer to your followup question is at the bottom of my original 
message,
which I'll quote, trimming out the other bits of that note:

> -----Original Message-----
> From: RJ Atkinson [mailto:rja@extremenetworks.com]
> Sent: Thursday, December 12, 2002 1:40 PM
> To: ieprep
> Subject: [Ieprep] Re: Mike's scenario


> 	For the person who has authorisation from *the end-user organisation
> that is paying for the access link*, that organisation can configure 
> routers,
> switches, and other devices to provide such preferential treatment 
> (e.g.
> Class-based queuing in the routers, switches, or other devices).  This 
> can be
> configured in  a variety of ways, including access-lists, filters, and 
> other
> implementation-specific methods.
> A variety of implementation-specific, link-layer, and intra-domain QoS
> methods (several of which have been described on this list in the past)
> can be used.
>
> 	Please note the absence of inter-domain QoS as part of the solution.
>
> Ran
> rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 16:33:59 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA13693
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 16:33:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCLaVM17529
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 16:36:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLaVv17526
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 16:36:31 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA13684
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 16:33:28 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLW4v17299;
	Thu, 12 Dec 2002 16:32:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLV3v17277
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 16:31:03 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA13460
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 16:28:00 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 3EEE067122; Thu, 12 Dec 2002 16:31:02 -0500 (EST)
Date: Thu, 12 Dec 2002 16:30:55 -0500
Subject: Re: [Ieprep] Re: Mike's scenario
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Fred Baker <fred@cisco.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <1qcd67$stm9@halt-in.cisco.com>
Message-Id: <FEF8B0A5-0E18-11D7-8C00-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Thursday, Dec 12, 2002, at 14:33 America/Montreal, Fred Baker wrote:
>>         The scenario you describe does not sound realistic for IEprep.
>
> I'm not sure what you're arguing, Ran.

I'm arguing that the assumptions that Mike has made are not all
equally realistic assumptions to make.

>>         Authorisation from NCS (or whoever) is substantially less 
>> relevant than authorisation
>> from the organisation that is paying for the access link.  In 
>> particular, if it
>> is not NCS's access link on NCS's budget, then NCS does NOT get to 
>> set the policies
>> for that link.
>
> Wonderful, Ran. The point is that the service on the link would have 
> to intersect *two* policies: there has to be local service 
> authorization (try placing a GETS call from a phone whse owner doesn't 
> pay his bill; I'll bet it doesn't work at all), and in addition it 
> needs to have the relevant governmental policy authorization.

Nope.  If (A) pays for an IP access link, then (A) gets to set the 
policy on their
link.  A's ISP does not get the power to set the policy on that link 
unilaterally
(generally speaking; the details of the ISP's contract with A obviously 
matters here).
The government does not get the power to set the policy on that link 
unilaterally either.

If the government wants to have influence over A's access link(s), it 
seems to several
of us that the government needs to have agreements with both A and A's 
ISP(s).  Absent
specific agreements with A *and* A's ISP(s), then the government simply 
does not have
any right to influence the policy deployed on A's access link.  An 
agreement
between a government and an ISP might impact how that ISP does things 
*inside
the ISP's own network*, but that contract(with the ISP)'s scope could 
not legitimately
include anything belonging to an ISP's customer (e.g. CPE access router 
on the access
link that is owned/operated by the customer).

This is very very basic contract law, at least in places using English 
Common Law
as their basis.  In the US case, there is actually specific 
constitutional
prohibitions from the government just appropriating other folks' 
property.
In other countries, laws probably vary.

Mike's scenario was specific in saying that the government had a 
contract with
the ISP.  It did NOT say there was a contract or other agreement with 
the end-user,
yet Mike's scenario has "government-authorised" folks who might not be
authorised by the end-user (A) consuming bandwidth that is paid for 
solely
by A.

And that *absolutely* does matter.  It is a very important point that 
several
of us have been trying to make for a long while now:

	NCS (or whoever) needs to be thinking about LOTS of contracts,
	encompassing not just the big ISPs, but also the end-users whose
	access links that NCS might want to have NCS's user-base access.

Instead, the repeated misunderstanding from NCS and its contractors is 
that its
sufficient to just have contracts with a few big ISPs.  With the 
Internet's
de-centralised architecture (opposite to PSTN's centralised 
architecture),
the set of agreements needed is pretty significantly larger than with 
the
existing PSTN.

Is that more clear to you ?

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 16:50:07 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA14281
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 16:50:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCLqdW18721
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 16:52:39 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLqdv18718
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 16:52:39 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA14257
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 16:49:36 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLm3v18552;
	Thu, 12 Dec 2002 16:48:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLlxv18514
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 16:47:59 -0500
Received: from imo-d06.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA14069
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 16:44:55 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d06.mx.aol.com (mail_out_v34.13.) id i.167.1855e15d (4312);
	Thu, 12 Dec 2002 16:47:42 -0500 (EST)
Message-ID: <167.1855e15d.2b2a5dfe@aol.com>
Date: Thu, 12 Dec 2002 16:47:42 EST
Subject: Re: [Ieprep] updated requirements documents (general)
To: K.Carlberg@cs.ucl.ac.uk, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_167.1855e15d.2b2a5dfe_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_167.1855e15d.2b2a5dfe_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

Comments on draft -01:

This generally covers the requirements quite well.

1.1 List of existing standards:

H.GEF.4 is now H.460.4

3. General Requirements

2) The statement that "labels for different layers are NOT required to be 
related to each other" could be misinterpreted by some (regardless of the 
exact grammatical parsing of the words). The fundamental requirement detailed 
in draft-pierce-ieprep-assured-service-00 for one application presumes that 
the "label" which identifies the call level priority may affect the choice of 
a "label" which identifies the packet layer treatment. (Otherwise why have 
it?) This statement in the draft could be read by some to mean that this is 
not required (or even that it is not allowed). In any case, the upper case 
"NOT" by itself is not one of the key words defined in RFC 2119. Suggest that 
this sentence be reworded as:

"Labels for different layers are not required to be the same, but MAY be 
related to each other."

I believe this is consistent with what others have said it is intended to 
mean and would help avoid the possibility of mis-interpretation.

3) In the 4th paragraph, the sentence beginning "The relationship between 
unrelated labels..." seems a little strange. Perhaps it should say "The 
meanings of unrelated labels..."

4. Issues

1) Accounting: Following my presentation in Atlanta at which no objection was 
raised, this should be moved to the Requirements section. I suggest the 
following text, which I believe correctly reflects the need and avoids any 
bad implications:

9) Accounting (auditing)

Accounting represents a method of tracking actual usage of a service for 
purposes of capacity and trend analysis, cost allocation, auditing, and 
billing [RFC2975]. We assume that the usage of any service better than best 
effort will be tracked and may subsequently be billed to the user. Accounting 
for billing purposes is not addressed as a general requirement for ETS. 
However, solutions used to realize ETS should include the capability of an 
accounting mechanism to support auditing as defined in [RFC2975].


6. References should be corrected as follows:

2. Designation should be "ANSI T1.631-1993 (R1999)" It was reaffirmed in 1999.
3. Date should be "E.106, March 2000"
4. F.706 has not yet been approved, so list it as "draft" without a date. We 
hope it will be approved in May.
5. This has been approved as "Call Priority Designation for H.323 Calls", 
H.460.4, November 2002.
Add: "Introduction to Accounting Management", RFC 2975, October 2000.

Mike Pierce
Artel


--part1_167.1855e15d.2b2a5dfe_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>Comments on draft -01:
<BR>
<BR>This generally covers the requirements quite well.
<BR>
<BR>1.1 List of existing standards:
<BR>
<BR>H.GEF.4 is now H.460.4
<BR>
<BR>3. General Requirements
<BR>
<BR>2) The statement that "labels for different layers are NOT required to be related to each other" could be misinterpreted by some (regardless of the exact grammatical parsing of the words). The fundamental requirement detailed in draft-pierce-ieprep-assured-service-00 for one application presumes that the "label" which identifies the call level priority may affect the choice of a "label" which identifies the packet layer treatment. (Otherwise why have it?) This statement in the draft could be read by some to mean that this is not required (or even that it is not allowed). In any case, the upper case "NOT" by itself is not one of the key words defined in RFC 2119. Suggest that this sentence be reworded as:
<BR>
<BR>"Labels for different layers are not required to be the same, but MAY be related to each other."
<BR>
<BR>I believe this is consistent with what others have said it is intended to mean and would help avoid the possibility of mis-interpretation.
<BR>
<BR>3) In the 4th paragraph, the sentence beginning "The relationship between unrelated labels..." seems a little strange. Perhaps it should say "The meanings of unrelated labels..."
<BR>
<BR>4. Issues
<BR>
<BR>1) Accounting: Following my presentation in Atlanta at which no objection was raised, this should be moved to the Requirements section. I suggest the following text, which I believe correctly reflects the need and avoids any bad implications:
<BR>
<BR>9) Accounting (auditing)
<BR>
<BR>Accounting represents a method of tracking actual usage of a service for purposes of capacity and trend analysis, cost allocation, auditing, and billing [RFC2975]. We assume that the usage of any service better than best effort will be tracked and may subsequently be billed to the user. Accounting for billing purposes is not addressed as a general requirement for ETS. However, solutions used to realize ETS should include the capability of an accounting mechanism to support auditing as defined in [RFC2975].
<BR>
<BR>
<BR>6. References should be corrected as follows:
<BR>
<BR>2. Designation should be "ANSI T1.631-1993 (R1999)" It was reaffirmed in 1999.
<BR>3. Date should be "E.106, March 2000"
<BR>4. F.706 has not yet been approved, so list it as "draft" without a date. We hope it will be approved in May.
<BR>5. This has been approved as "Call Priority Designation for H.323 Calls", H.460.4, November 2002.
<BR>Add: "Introduction to Accounting Management", RFC 2975, October 2000.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_167.1855e15d.2b2a5dfe_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 16:50:11 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA14294
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 16:50:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCLqhs18736
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 16:52:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLqhv18733
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 16:52:43 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA14261
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 16:49:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLm2v18536;
	Thu, 12 Dec 2002 16:48:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLlxv18510
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 16:47:59 -0500
Received: from imo-d10.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA14068
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 16:44:55 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d10.mx.aol.com (mail_out_v34.13.) id i.16b.1870d08c (4312);
	Thu, 12 Dec 2002 16:47:45 -0500 (EST)
Message-ID: <16b.1870d08c.2b2a5e00@aol.com>
Date: Thu, 12 Dec 2002 16:47:44 EST
Subject: Re: [Ieprep] updated requirements documents (telephony)
To: K.Carlberg@cs.ucl.ac.uk, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_16b.1870d08c.2b2a5e00_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_16b.1870d08c.2b2a5e00_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

I believe that the one "issue" in Section 4 needs to be described more fully.

1) The point in alternate routing for ETS is not that existing flows are 
automatically rerouted if a router becomes unavailable (even though this is 
useful to maintain a connection when a failure occurs). It is that, while the 
packets for a normal traffic might follow a specific, efficient path (to 
setup a new connection), the packets for an ETS call should attempt 
additional routes (even if very inefficient) which the normal call does not 
if there is congestion on the normal route. (Yes, I know all the arguments 
about "congestion can't happen in the core", but it's been clearly admitted 
that it can occur on some access links where alternate routing could also be 
used.)

The current second paragraph seems to limit the issue of "alternate routing" 
at the application layer to "choosing .. egress points to the PSTN". A 
significant issue for survivability and congestion handling in some networks 
could be the desire for abilities to alternate route around congestion within 
the IP portion. (The ability to reroute around failures is already assumed.)

I suggest this section on Alternate Paths be split into two separate 
sub-parts to address the two "layers" within IP and relate each to what the 
PSTN did (or couldn't do) as follows:

1) Alternate paths

Experience with GETS over the PSTN has shown the utility of alternate paths 
to a destination to help facilitate emergency-related communications. This is 
done at the application layer, independent of any lower layer rerouting 
(SONET rings, etc) that naturally applies to all traffic using the link, 
whether emergency or normal. Application of emergency telephony services to 
the Internet (and IP in general) means that there are two separate issues to 
be addressed.

a) IP Packet Transport layer: Unlike the PSTN, which creates a fixed path 
(circuit) during call setup phase, the Internet uses dynamic routing for IP 
packets. This dynamic routing capability automatically causes IP packets to 
travel the best current path, which may change during the life of the "call" 
due to failures of infrastructure components. However, at the time of "call" 
set up (at the application layer), the route choice for the packet transport 
layer has a single choice (the best current one). An issue to be addressed is 
the utility and possibility of providing additional routes (even is less 
efficient) for packet flows for higher priority "calls".

b) Call application layer:  The Internet infrastructure itself does not have 
the concept of a "call" or the concept of "call setup", although IP telephony 
applications using it would have application-layer awareness of calls or the 
call setup concept. Alternate path routing at the application layer may be 
useful in choosing one of possibly several egress points to the PSTN from the 
Internet.


Mike Pierce
Artel



--part1_16b.1870d08c.2b2a5e00_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>I believe that the one "issue" in Section 4 needs to be described more fully.
<BR>
<BR>1) The point in alternate routing for ETS is not that existing flows are automatically rerouted if a router becomes unavailable (even though this is useful to maintain a connection when a failure occurs). It is that, while the packets for a normal traffic might follow a specific, efficient path (to setup a new connection), the packets for an ETS call should attempt additional routes (even if very inefficient) which the normal call does not if there is congestion on the normal route. (Yes, I know all the arguments about "congestion can't happen in the core", but it's been clearly admitted that it can occur on some access links where alternate routing could also be used.)
<BR>
<BR>The current second paragraph seems to limit the issue of "alternate routing" at the application layer to "choosing .. egress points to the PSTN". A significant issue for survivability and congestion handling in some networks could be the desire for abilities to alternate route around congestion within the IP portion. (The ability to reroute around failures is already assumed.)
<BR>
<BR>I suggest this section on Alternate Paths be split into two separate sub-parts to address the two "layers" within IP and relate each to what the PSTN did (or couldn't do) as follows:
<BR>
<BR>1) Alternate paths
<BR>
<BR>Experience with GETS over the PSTN has shown the utility of alternate paths to a destination to help facilitate emergency-related communications. This is done at the application layer, independent of any lower layer rerouting (SONET rings, etc) that naturally applies to all traffic using the link, whether emergency or normal. Application of emergency telephony services to the Internet (and IP in general) means that there are two separate issues to be addressed.
<BR>
<BR>a) IP Packet Transport layer: Unlike the PSTN, which creates a fixed path (circuit) during call setup phase, the Internet uses dynamic routing for IP packets. This dynamic routing capability automatically causes IP packets to travel the best current path, which may change during the life of the "call" due to failures of infrastructure components. However, at the time of "call" set up (at the application layer), the route choice for the packet transport layer has a single choice (the best current one). An issue to be addressed is the utility and possibility of providing additional routes (even is less efficient) for packet flows for higher priority "calls".
<BR>
<BR>b) Call application layer: &nbsp;The Internet infrastructure itself does not have the concept of a "call" or the concept of "call setup", although IP telephony applications using it would have application-layer awareness of calls or the call setup concept. Alternate path routing at the application layer may be useful in choosing one of possibly several egress points to the PSTN from the Internet.
<BR>
<BR>
<BR>Mike Pierce
<BR>Artel
<BR>
<BR></FONT></HTML>

--part1_16b.1870d08c.2b2a5e00_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 16:59:59 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA14637
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 16:59:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCM2V119631
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 17:02:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCM2Vv19628
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 17:02:31 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA14611
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 16:59:28 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLw2v18957;
	Thu, 12 Dec 2002 16:58:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCLvgv18941
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 16:57:42 -0500
Received: from halt-in.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA14411
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 16:54:39 -0500 (EST)
Received: from cisco.com (10.32.253.236)
  by halt-in.cisco.com with SMTP; 12 Dec 2002 13:58:44 +0000
Message-Id: <1qcd67$tqth@halt-in.cisco.com>
Message-Id: <5.2.0.9.2.20021212134855.05b00170@mira-sjcm-4.cisco.com>
X-Sender: fred@mira-sjcm-4.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 5.2.0.9
Date: Thu, 12 Dec 2002 13:54:19 -0800
To: RJ Atkinson <rja@extremenetworks.com>
From: Fred Baker <fred@cisco.com>
Subject: Re: [Ieprep] Re: Mike's scenario
Cc: ieprep <ieprep@ietf.org>
In-Reply-To: <FEF8B0A5-0E18-11D7-8C00-00039357A82A@extremenetworks.com>
References: <1qcd67$stm9@halt-in.cisco.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

At 04:30 PM 12/12/2002 -0500, RJ Atkinson wrote:
>This is very very basic contract law, at least in places using English 
>Common Law as their basis.  In the US case, there is actually specific 
>constitutional prohibitions from the government just appropriating other 
>folks' property.

Fine. As far as I know, nobody on this list is a lawyer, and this working 
group is not chartered to discuss law. This working group is chartered to 
discuss technology under the assumption that the appropriate legal vehicles 
are in place.

>the repeated misunderstanding from NCS and its contractors is that its 
>sufficient to just have contracts with a few big ISPs.  With the 
>Internet's de-centralised architecture (opposite to PSTN's centralised 
>architecture), the set of agreements needed is pretty significantly larger 
>than with the existing PSTN.

OK, so your issue here is that the policy has to apply, by appropriate law 
or contract, to both ends of the link. I'll buy that. Please insert it as 
an unstated assumption in Mike's comments.

Can we get back to dialog on the topic, rather than attempts to derail the 
working group? I can't say I have seen useful dialog on this list since the 
IETF, and I haven't seen comments from you on the particular cases in point 
where Mikes concerns apply and documentation has been sent to the list to 
substantiate them. 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 17:33:17 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15553
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 17:33:17 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCMZn621386
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 17:35:49 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCMZnv21383
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 17:35:49 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15550
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 17:32:46 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCMY5v21286;
	Thu, 12 Dec 2002 17:34:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCMXZv21253
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 17:33:35 -0500
Received: from [140.32.132.66] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA15438
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 17:30:31 -0500 (EST)
Received: from mail.nps.navy.mil by [140.32.132.66]
          via smtpd (for [132.151.6.1]) with ESMTP; Thu, 12 Dec 2002 14:28:50 -0800
Received: from [131.120.179.253] (budden.ro.nps.navy.mil [131.120.179.253])
	by capella.nps.navy.mil (8.12.2/8.12.2) with ESMTP id gBCMTESI025619;
	Thu, 12 Dec 2002 14:29:15 -0800 (PST)
Mime-Version: 1.0
X-Sender: budden@essex.ad.nps.navy.mil
Message-Id: <a05100311ba1ebcc655a4@[131.120.179.253]>
In-Reply-To: <20021212120556.A27516@sprint.net>
References: <185.1339863f.2b276b1f@aol.com>
 <5.2.0.9.2.20021212114409.082ff508@mira-sjcm-4.cisco.com>
 <20021212120556.A27516@sprint.net>
Date: Thu, 12 Dec 2002 14:31:41 -0800
To: David Meyer <dmm@sprint.net>, Fred Baker <fred@cisco.com>
From: Rex Buddenberg <budden@nps.navy.mil>
Subject: Re: [Ieprep] Re: PSTN collapse
Cc: Mpierce1@aol.com, ieprep@ietf.org
Content-Type: text/plain; charset="us-ascii" ; format="flowed"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

At 12:05 PM -0800 12/12/02, David Meyer wrote:
>  >> At 11:06 AM 12/10/2002 -0500, Mpierce1@aol.com wrote:
>>>  >(How many Internet routers have generators anyway?)
>>>
>>>  most of the ones in ISP POPs do. They get pretty serious about that. They
>>>  like revenue.
>
>I would guess that most ISPs of any size have redundance power
>(from redundant grids, with redundant entrance), as well as other
>backup power (e.g. diesel generator); heck, UO even has that
>much. In addition, frequently the entire HVAC systems are
>redundant, so the change of this kind of failure is pretty
>minimized (again, at least for the carriers that I know about).


You've got the Achilles Heel of network high availability engineering.

We can design a very resilient network structure -- lots of alternate 
routes and lots of redundant routers.  But if the volts go away, none 
of it works.  Today, our power grid is quite centralized (much like 
the PSTN).
      We all know about putting UPSs under routers and switches ... 
but they're endurance tends to be very limited.  Backup generators 
help, but they're expensive enough that you're unlikely to have one 
for each router.  (I've had students track the fuel cell developments 
to see if there's offset there -- so far, mostly promises).
      When USS Cole was bombed in Yemen two years ago, it was without 
power for 8 days -- the bomb destroyed the engine room that contained 
the generators.  So if it had a network in it to hold up the damage 
control system, the network would have failed.   Not because the 
network equipment was broken, but because it didn't have power.  (In 
reality, most of the damage control network is still sneakernet).
      I think if you look at the contributions that Kimberly is 
digging up, they're pretty much all saying they need high 
availability and my Navy-specific example is pretty close to the same 
kinds of issues they have -- they need comms and indirectly, they 
need the volts to hold the comms up.



-- 
b
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 18:10:34 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA16979
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 18:10:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCND7N24304
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 18:13:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCND7v24301
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 18:13:07 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA16970
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 18:10:02 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCNBCv24191;
	Thu, 12 Dec 2002 18:11:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCNA3v24060
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 18:10:03 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA16818
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 18:06:58 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Thu, 12 Dec 2002 18:09:17 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121218093820104
 ; Thu, 12 Dec 2002 18:09:38 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YW14CM92>; Thu, 12 Dec 2002 18:09:25 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A60@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson '" <rja@extremenetworks.com>
Cc: "'ieprep '" <ieprep@ietf.org>
Date: Thu, 12 Dec 2002 18:09:32 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBCNA3v24061
Subject: [Ieprep] Summary of IEPREP work and VoIP
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

>Thanks.  You read more carefully than Fred, and its appreciated.

To be fair, Fred may not have read all the thousands of emails that have
been written in ieprep lately.  ;-)

Mike's scenario has been discussed, in various forms, many times. The below
cases represent our main findings in my own words.  We assume an ETS
location is connected via an access link to the Internet.

Case 1:  There is a predefined relationship between SIP provider and ISP
providing access to the ETS location and this is known in advance of an
emergency.  (The model might be like an enhanced GETS-like WorldCom SIP
service).  In this case, there are two main things that can be done.  

o	Call signaling where some new resource priority mechanism (this
mechanism will hopefully to be created in SIPPING as a result of Henning's
requirements document) is in SIP that can be used to identify calls for
preferential access to ports on PSTN gateways or other end systems.  
o	If the provider runs stateful SIP proxies, the new resource priority
mechanism could also allow preferential treatment of proxy resources.  
o	Depending upon how this provider engineers their service (suppose
they want to be very stateful and track how many calls are coming to/from
which locations), a traditional call admission control could possibly be
supported giving preferential treatment to emergency calls by allowing them
a greater probably of call setup than regular calls.
o	We take the assumption (largely substantiated for many U.S. regions)
that network congestion, if it exists, is on the access link.  Hence, to
increase the likelihood of good sounding calls over an access link, the
number of calls can be controlled locally at the customer location.  This
can be achieved through the use of devices (e.g., QoS appliances or SIP
proxy firewalls) that are enhanced to only setup the highest priority calls
over the access link, up to a predetermined maximum influenced by the speed
of the access link, possibly even supporting preemption.

·	The second thing that is possible is some kind of CoS on the access
link.  
o	In all cases, the customer may use filters, QoS appliances, etc. to
prioritize voice over data at the CPE (as Ran suggested).  In some cases,
the ISP will also perform this service at the network end of that link
(e.g., Sprint CoS service).

Case 2:  The ISP provides a CoS service but any SIP services come from a
provider whose service is distributed in other places in the Internet.
Basically, here the ISP provides transport with some CoS but isn't coupled
with a SIP service.
o	The resource priority mechanism would still be useful if gateways or
end systems support such a mechanism.  
o	However, since the SIP provider has no information about how the
user is obtaining access, it doesn't really make sense to do "call admission
control".  (For example, 10 users at the same location could rely on 10
different SIP service providers - none of the service provider can determine
what is flowing over the access link they happen to all be using.)  
o	Note that users may still retain control over which connections get
setup by local control (several QoS appliances or SIP proxy firewalls could
be modified to support this).

o	The second thing that is possible is some kind of CoS on the access
link.  Again, the customer may use filters, QoS appliances, etc. to
prioritize voice over data at the CPE (as Ran suggested).  In this case, the
ISP will also perform this service at the network end of that link (e.g.,
Sprint CoS service).

Case 3:  The ISP provides a best effort service and any SIP services come
from a provider whose service is distributed in many places in the Internet.


This is just like case 2 above except we don't expect a CoS treatment at the
network end of the access link.

In terms of what needs to be done for SIP protocol changes, I think we have
achieved our objectives.  I haven't shared this with the group yet, but I
believe Megaco is okay as is.  I'll save my research there for another
email. Now, Rohan had a nice idea that could also supplement the above from
say an Internet Café but no one has expressed an interest.  I think all the
pieces, at least surrounding preferential treatment and SIP can be glued
together to form preferential call service.  But, the working group remains
open to new ideas.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 12 18:42:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17690
	for <ieprep-archive@odin.ietf.org>; Thu, 12 Dec 2002 18:42:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBCNj5T26299
	for ieprep-archive@odin.ietf.org; Thu, 12 Dec 2002 18:45:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCNj5v26296
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 12 Dec 2002 18:45:05 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17687
	for <ieprep-web-archive@ietf.org>; Thu, 12 Dec 2002 18:42:00 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCNh6v26243;
	Thu, 12 Dec 2002 18:43:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBCNgYv26205
	for <ieprep@optimus.ietf.org>; Thu, 12 Dec 2002 18:42:34 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17594
	for <ieprep@ietf.org>; Thu, 12 Dec 2002 18:39:30 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAHF6M>; Thu, 12 Dec 2002 18:39:56 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149F0@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'King, Kimberly  S.'" <KIMBERLY.S.KING@saic.com>,
        "'RJ Atkinson '"
	 <rja@extremenetworks.com>
Cc: "'ieprep '" <ieprep@ietf.org>
Subject: Rohan's idea RE: [Ieprep] Summary of IEPREP work and VoIP
Date: Thu, 12 Dec 2002 18:39:41 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBCNgYv26207
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

I am interested.

-----Original Message-----
From: King, Kimberly S. [mailto:KIMBERLY.S.KING@saic.com]

 Now, Rohan had a nice idea that could also supplement the above from
say an Internet Café but no one has expressed an interest.  I think all the
pieces, at least surrounding preferential treatment and SIP can be glued
together to form preferential call service.  But, the working group remains
open to new ideas.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 07:31:07 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA21811
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 07:31:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDCXYE13305
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 07:33:34 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDCXXv13302
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 07:33:33 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA21794
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 07:30:34 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDCUSv13200;
	Fri, 13 Dec 2002 07:30:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDCTLv13162
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 07:29:21 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA21760
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 07:26:21 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBDCN1o05603;
	Fri, 13 Dec 2002 04:23:01 -0800 (PST)
Date: Fri, 13 Dec 2002 04:23:01 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: Fred Baker <fred@cisco.com>
cc: Mpierce1@aol.com, <ieprep@ietf.org>
Subject: Re: [Ieprep] Re: PSTN collapse
In-Reply-To: <1qcd67$t233@halt-in.cisco.com>
Message-ID: <Pine.GSO.4.44.0212130421410.5482-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Thu, 12 Dec 2002, Fred Baker wrote:
    > >(How many Internet routers have generators anyway?)
    > most of the ones in ISP POPs do. They get pretty serious about that. They
    > like revenue.

Any customer of any size has a battery string and generators.  As a
Californian, I can't imagine trying to keep an Internet connection up
under this administration without one.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 08:10:08 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22527
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 08:10:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDDCaM15881
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 08:12:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDCZv15878
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 08:12:35 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22520
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 08:09:37 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDBCv15803;
	Fri, 13 Dec 2002 08:11:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDAhv15784
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 08:10:43 -0500
Received: from bells.cs.ucl.ac.uk (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA22484
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 08:07:45 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.04348-0@bells.cs.ucl.ac.uk>; Fri, 13 Dec 2002 13:10:30 +0000
To: Mpierce1@aol.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] updated requirements documents (telephony)
In-reply-to: Your message of "Thu, 12 Dec 2002 16:47:44 EST." <16b.1870d08c.2b2a5e00@aol.com>
Date: Fri, 13 Dec 2002 13:10:28 +0000
Message-ID: <1200.1039785028@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>



> I believe that the one "issue" in Section 4 needs to be described more
> fully.
> 
> 1) The point in alternate routing for ETS is not that existing flows are
> automatically rerouted if a router becomes unavailable (even though this
> is useful to maintain a connection when a failure occurs). It is that,
> while the packets for a normal traffic might follow a specific,
> efficient path (to setup a new connection), the packets for an ETS call
> should attempt additional routes (even if very inefficient) which the
> normal call does not if there is congestion on the normal route. (Yes, I
> know all the arguments about "congestion can't happen in the core", but
> it's been clearly admitted that it can occur on some access links where
> alternate routing could also be used.)

the original passage in section 4 does not make a reference to "existing
flows" versus new flows or calls -- ie, its not that specific.  and so
taken at a higher level, i would say that the original passage would
apply to both (new and existing flows) without citing the subject of
congestion.  I'd prefer not to get into specifics in this document
and instead push more extended discussions on these "Issues" to the
Framework document.

> The current second paragraph seems to limit the issue of "alternate
> routing" at the application layer to "choosing .. egress points to the
> PSTN". A significant issue for survivability and congestion handling in
> some networks could be the desire for abilities to alternate route
> around congestion within the IP portion. (The ability to reroute around
> failures is already assumed.)
>
> I suggest this section on Alternate Paths be split into two separate 
> sub-parts to address the two "layers" within IP and relate each to what the 
> PSTN did (or couldn't do) as follows:

ok, the intent was to present a point of reference as opposed to limit
the issue to the application layer.  it may be easier and simpler to
just remove the second paragraph than to rewrite the section and discuss
in more depth the areas that altenate path routing does/can exist. Again,
lets push this to the Framework document.

-ken

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 08:10:55 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22575
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 08:10:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDDDMs15925
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 08:13:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDDMv15922
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 08:13:22 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22544
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 08:10:24 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDC1v15847;
	Fri, 13 Dec 2002 08:12:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDBwv15828
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 08:11:58 -0500
Received: from bells.cs.ucl.ac.uk (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA22506
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 08:09:00 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.04455-0@bells.cs.ucl.ac.uk>; Fri, 13 Dec 2002 13:11:53 +0000
To: Mpierce1@aol.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] updated requirements documents (general)
In-reply-to: Your message of "Thu, 12 Dec 2002 16:47:42 EST." <167.1855e15d.2b2a5dfe@aol.com>
Date: Fri, 13 Dec 2002 13:11:52 +0000
Message-ID: <1228.1039785112@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


> 1.1 List of existing standards:
> 
> H.GEF.4 is now H.460.4

right, will do.

> 3. General Requirements
> 
> 2) The statement that "labels for different layers are NOT required to
> be related to each other" could be misinterpreted by some (regardless of
> the exact grammatical parsing of the words). The fundamental requirement
> detailed in draft-pierce-ieprep-assured-service-00 for one application
> presumes that the "label" which identifies the call level priority may
> affect the choice of a "label" which identifies the packet layer
> treatment. (Otherwise why have it?) This statement in the draft could be
> read by some to mean that this is not required (or even that it is not
> allowed). In any case, the upper case "NOT" by itself is not one of the
> key words defined in RFC 2119. Suggest that this sentence be reworded
> as:
> 
> "Labels for different layers are not required to be the same, but MAY be
> related to each other."

ok, I can make the change to the original passage so that it is similar
to your above sentence.

> 3) In the 4th paragraph, the sentence beginning "The relationship
> between unrelated labels..." seems a little strange. Perhaps it should
> say "The meanings of unrelated labels..."

I'll clarify the passage in the next version.

> 4. Issues
> 
> 1) Accounting: Following my presentation in Atlanta at which no
> objection was raised, this should be moved to the Requirements
> section. I suggest the following text, which I believe correctly
> reflects the need and avoids any bad implications:

I disagree with listing it as a requirement -- especially in the General
Requirements draft.  I have listed my objections in the past, as has
Henning and others on the list.  I have also stated on this list and
in the General requirements document that...

   Subsequent documents that articulate a more specific set of
   requirements for a particular service may make a statement about the
   following issues.

...and you've done that with the Pierce draft as it relates to MLPP.
Without revisiting the previous arguements, I'll leave it at that.

As for your presentation in Atlanta, I have to apologize that while i
was present, I was engaged in another conversation.  It was my
impression that your presentation was specifically scoped at MLPP and
not ETS as a whole, which may explain in part why there were no
objections.  And finally, working group consensus is predominantly
reached on the list.  I didn't hear a call for humms when you spoke :-)


> 6. References should be corrected as follows:
> 
> 2. Designation should be "ANSI T1.631-1993 (R1999)" It was reaffirmed 
> in 1999.
> 3. Date should be "E.106, March 2000"
> 4. F.706 has not yet been approved, so list it as "draft" without a date. 
> We hope it will be approved in May.
> 5. This has been approved as "Call Priority Designation for H.323 Calls", 
> H.460.4, November 2002.

ok.  these will be changed.

> Add: "Introduction to Accounting Management", RFC 2975, October 2000.

I may add it in.

-ken

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 08:21:58 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22748
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 08:21:58 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDDOPH16251
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 08:24:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDOPv16248
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 08:24:25 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22741
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 08:21:26 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDN1v16201;
	Fri, 13 Dec 2002 08:23:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDM1v16156
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 08:22:01 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22670
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 08:19:03 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.98])
	by gnat.inet.org (Postfix) with ESMTP
	id 3D07667104; Fri, 13 Dec 2002 08:22:09 -0500 (EST)
Date: Fri, 13 Dec 2002 08:21:52 -0500
Subject: Re: [Ieprep] updated requirements documents (general)
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: K.Carlberg@cs.ucl.ac.uk, Ran Atkinson <rja@extremenetworks.com>
To: ieprep <ieprep@ietf.org>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <167.1855e15d.2b2a5dfe@aol.com>
Message-Id: <D80FC790-0E9D-11D7-BE1C-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Thursday, Dec 12, 2002, at 16:47 America/Montreal, Mpierce1@aol.com 
wrote:
> 2) The statement that "labels for different layers are NOT required to 
> be related to each other" could be misinterpreted by some (regardless 
> of the exact grammatical parsing of the words). The fundamental 
> requirement detailed in draft-pierce-ieprep-assured-service-00 for one 
> application presumes that the "label" which identifies the call level 
> priority may affect the choice of a "label" which identifies the 
> packet layer treatment. (Otherwise why have it?)

One reason to have a label in SIP, even if lower-layer labels don't 
happen to be in use,
might be to ensure that a telephone voice-call whose signalling path is 
PSTN--SIP--PSTN
does not lose any labels that originated on the PSTN side (where they 
might be
more meaningful) just because it transited a SIP segment.

>  This statement in the draft could be read by some to mean that this 
> is not required (or even that it is not allowed). In any case, the 
> upper case "NOT" by itself is not one of the key words defined in RFC 
> 2119. Suggest that this sentence be reworded as:

(There is precedence for capitalising NOT in standards-track RFCs, so 
that
is not at issue here.)

> 1) Accounting: Following my presentation in Atlanta at which no 
> objection was raised,

Disagree.  There were several objections from operators during the 
meeting.  I was
in the room.  Moreover, the list discussion takes precedence over 
in-person meetings
under IETF process rules (unlike with some other standards groups).  
And there are
numerous unresolved objections here on the list -- mainly because you 
haven't made
it clear what exactly you want to account for.

It would help further discussion a great deal if you could be more 
clear and very
specific in what objects you want.  Depending on what they are, folks 
here might
well agree with the specifics.

The root problem with your proposal remains the same.  Your proposal 
does not make clear
what sorts of objects you want accounting to be performed on.  A number 
of us might
withdraw our objections if you could be a lot more clear about this.  I 
noted this
on list specifically before the presentation and (in vain) asked that 
the presentation
make this clear.  The presentation did not make it clear.

>  this should be moved to the Requirements section. I suggest the 
> following text, which I believe correctly reflects the need and avoids 
> any bad implications:
>
> 9) Accounting (auditing)
>
> Accounting represents a method of tracking actual usage of a service 
> for purposes of capacity and trend analysis, cost allocation, 
> auditing, and billing [RFC2975]. We assume that the usage of any 
> service better than best effort will be tracked and may subsequently 
> be billed to the user. Accounting for billing purposes is not 
> addressed as a general requirement for ETS. However, solutions used to 
> realize ETS should include the capability of an accounting mechanism 
> to support auditing as defined in [RFC2975].

Disagree.  I'd be happy to add text saying that a solution "must not 
preclude
an operator from performing accounting", but not text making it a 
requirement.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 08:23:00 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22771
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 08:23:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDDPRP16298
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 08:25:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDPRv16295
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 08:25:27 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22764
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 08:22:29 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDO1v16238;
	Fri, 13 Dec 2002 08:24:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDNqv16217
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 08:23:52 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22723
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 08:20:54 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.98])
	by gnat.inet.org (Postfix) with ESMTP
	id D44EF67104; Fri, 13 Dec 2002 08:24:00 -0500 (EST)
Date: Fri, 13 Dec 2002 08:23:44 -0500
Subject: Re: [Ieprep] Re: Mike's scenario
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Fred Baker <fred@cisco.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <1qcd67$tqth@halt-in.cisco.com>
Message-Id: <1ACEEDA5-0E9E-11D7-BE1C-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Thursday, Dec 12, 2002, at 16:54 America/Montreal, Fred Baker wrote:
> I can't say I have seen useful dialog on this list since the IETF, and 
> I haven't seen comments from you on the particular cases in point 
> where Mikes concerns apply and documentation has been sent to the list 
> to substantiate them.

Try reading the responses to Janet Gunn's scenario.
Try also reading the lower portion of the email of mine that you 
misread.

If those got lost in transit to your mailbox, send me a private note and
I'll happily resend them to you privately.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 08:29:04 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22856
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 08:29:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDDVVv16470
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 08:31:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDVVv16467
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 08:31:31 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22849
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 08:28:33 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDU1v16415;
	Fri, 13 Dec 2002 08:30:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDTLv16389
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 08:29:21 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22830
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 08:26:22 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.98])
	by gnat.inet.org (Postfix) with ESMTP
	id 3BD5C67104; Fri, 13 Dec 2002 08:29:29 -0500 (EST)
Date: Fri, 13 Dec 2002 08:29:13 -0500
Subject: Re: [Ieprep] updated requirements documents (telephony)
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: K.Carlberg@cs.ucl.ac.uk, ieprep@ietf.org
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <16b.1870d08c.2b2a5e00@aol.com>
Message-Id: <DEA75F0A-0E9E-11D7-BE1C-00039357A82A@extremenetworks.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBDDTLv16390
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


On Thursday, Dec 12, 2002, at 16:47 America/Montreal, Mpierce1@aol.com 
wrote:
> 1) The point in alternate routing for ETS is not that existing flows 
> are automatically rerouted if a router becomes unavailable (even 
> though this is useful to maintain a connection when a failure occurs). 
> It is that, while the packets for a normal traffic might follow a 
> specific, efficient path (to setup a new connection), the packets for 
> an ETS call should attempt additional routes (even if very 
> inefficient) which the normal call does not if there is congestion on 
> the normal route. (Yes, I know all the arguments about "congestion 
> can't happen in the core", but it's been clearly admitted that it can 
> occur on some access links where alternate routing could also be > used.)
>
> The current second paragraph seems to limit the issue of "alternate 
> routing" at the application layer to "choosing .. egress points to the 
> PSTN". A significant issue for survivability and congestion handling 
> in some networks could be the desire for abilities to alternate route 
> around congestion within the IP portion. (The ability to reroute 
> around failures is already assumed.)

The clarity is helpful and appreciated, but please see Joel Halpern's 
previous
commentary on QoS-based routing.  There is significant negative 
experience with
routing schemes that try to include congestion as part of the routing 
metric.
What happens is an oscillation where traffic switches from link A to 
link B
because A is getting full, then link B gets full, and the traffic all 
switches
back.  A better approach, that is widely available today, is to use 
other
techniques (within a domain, ECMP is one example or MPLS-TE might be 
another).

> I suggest this section on Alternate Paths be split into two separate 
> sub-parts to address the two "layers" within IP and relate each to 
> what the PSTN did (or couldn't do) as follows:
>
> 1) Alternate paths
>
> Experience with GETS over the PSTN has shown the utility of alternate 
> paths to a destination to help facilitate emergency-related 
> communications. This is done at the application layer, independent of 
> any lower layer rerouting (SONET rings, etc) that naturally applies to 
> all traffic using the link, whether emergency or normal. Application 
> of emergency telephony services to the Internet (and IP in general) 
> means that there are two separate issues to be addressed.

What do you mean by "application-layer" and "alternate paths" ?

In the case of the Internet, path selection is an IP-layer function 
performed
by the routing system, not an application-layer function.

> a) IP Packet Transport layer: Unlike the PSTN, which creates a fixed 
> path (circuit) during call setup phase, the Internet uses dynamic 
> routing for IP packets. This dynamic routing capability automatically 
> causes IP packets to travel the best current path, which may change 
> during the life of the "call" due to failures of infrastructure 
> components. However, at the time of "call" set up (at the application 
> layer), the route choice for the packet transport layer has a single 
> choice (the best current one). An issue to be addressed is the utility 
> and possibility of providing additional routes (even is less 
> efficient) for packet flows for higher priority "calls".

I definitely don't follow what you are trying to get at above.  Please 
try to
explain in more detail what you are trying to do here.

> b) Call application layer:  The Internet infrastructure itself does 
> not have the concept of a "call" or the concept of "call setup", 
> although IP telephony applications using it would have 
> application-layer awareness of calls or the call setup concept. 
> Alternate path routing at the application layer may be useful in 
> choosing one of possibly several egress points to the PSTN from the 
> Internet.

Please explain what "alternate path routing at the application layer" 
is or
might be, with reference to the usual Internet Architecture.

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 08:45:06 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA23043
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 08:45:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDDlYJ17422
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 08:47:34 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDlYv17419
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 08:47:34 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA23031
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 08:44:35 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDk4v17377;
	Fri, 13 Dec 2002 08:46:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDDj9v17347
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 08:45:09 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA22991
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 08:42:10 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.98])
	by gnat.inet.org (Postfix) with ESMTP
	id ECEB267104; Fri, 13 Dec 2002 08:45:16 -0500 (EST)
Date: Fri, 13 Dec 2002 08:45:00 -0500
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep '" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A60@mcl-its-exs01.mail.saic.com>
Message-Id: <135A9FDE-0EA1-11D7-BE1C-00039357A82A@extremenetworks.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBDDj9v17348
Subject: [Ieprep] Re: Summary of IEPREP work and VoIP
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


On Thursday, Dec 12, 2002, at 18:09 America/Montreal, King, Kimberly S. 
wrote:
> Case 1:  There is a predefined relationship between SIP provider and 
> ISP
> providing access to the ETS location and this is known in advance of an
> emergency.  (The model might be like an enhanced GETS-like WorldCom SIP
> service).  In this case, there are two main things that can be done.
>
> o	Call signaling where some new resource priority mechanism (this
> mechanism will hopefully to be created in SIPPING as a result of 
> Henning's
> requirements document) is in SIP that can be used to identify calls for
> preferential access to ports on PSTN gateways or other end systems.
> o	If the provider runs stateful SIP proxies, the new resource priority
> mechanism could also allow preferential treatment of proxy resources.
> o	Depending upon how this provider engineers their service (suppose
> they want to be very stateful and track how many calls are coming 
> to/from
> which locations), a traditional call admission control could possibly 
> be
> supported giving preferential treatment to emergency calls by allowing 
> them
> a greater probably of call setup than regular calls.
> o	We take the assumption (largely substantiated for many U.S. regions)
> that network congestion, if it exists, is on the access link.  Hence, 
> to
> increase the likelihood of good sounding calls over an access link, the
> number of calls can be controlled locally at the customer location.  
> This
> can be achieved through the use of devices (e.g., QoS appliances or SIP
> proxy firewalls)

or SIP proxy firewalls or IP routers that are configured to enable 
CBQ/CoS
or link-layer QoS mechanisms)

>  that are enhanced to only setup the highest priority calls
> over the access link, up to a predetermined maximum influenced by the 
> speed
> of the access link, possibly even supporting preemption.

That phrasing is awkward and ambiguous since the IP-layer still does not
have the concept of a "call".  Presumably what one meant is something
more like:

...that are enhanced to provide preferred forwarding over the access 
link
to IP datagrams that are associated with specific authorised 
emergency-related
flows, up to some predetermined maximim influenced by the speed of the
access link.  This preferred forwarding might involve CBQ in routers,
lower-drop probability for authorised emergency-related IP packets,
or other mechanisms.  Some such mechanisms might be inside the 
link-layer,
rather than in the IP-layer or above.

> ·	The second thing that is possible is some kind of CoS on the access
> link.
> o	In all cases, the customer may use filters, QoS appliances, etc. to

s/filters,/router filters, router CoS, switch filters, link-specific 
AoS,/

> prioritize voice over data at the CPE (as Ran suggested).

Or, if a policy and deployment is agreed upon *by the operators of that 
link*,
one might be able to prioritise selected IP flows (of any application 
protocol).

> In some cases,
> the ISP will also perform this service at the network end of that link
> (e.g., Sprint CoS service).
>
> Case 2:  The ISP provides a CoS service but any SIP services come from 
> a
> provider whose service is distributed in other places in the Internet.
> Basically, here the ISP provides transport with some CoS but isn't 
> coupled
> with a SIP service.
> o	The resource priority mechanism would still be useful if gateways or
> end systems support such a mechanism.
> o	However, since the SIP provider has no information about how the
> user is obtaining access, it doesn't really make sense to do "call 
> admission
> control".  (For example, 10 users at the same location could rely on 10
> different SIP service providers - none of the service provider can 
> determine
> what is flowing over the access link they happen to all be using.)
> o	Note that users may still retain control over which connections get
> setup by local control (several QoS appliances or SIP proxy firewalls 
> could
> be modified to support this).

At least some existing switches and routers can be configured to do 
this,
within various implementation-specific constraints.

> o	The second thing that is possible is some kind of CoS on the access
> link.  Again, the customer may use filters, QoS appliances, etc. to

s/filters,/filters, routers, switches,/

> prioritize voice over data at the CPE (as Ran suggested).  In this 
> case, the
> ISP will also perform this service at the network end of that link 
> (e.g.,
> Sprint CoS service).

s/ISP will also/ISP might also/

> Case 3:  The ISP provides a best effort service and any SIP services 
> come
> from a provider whose service is distributed in many places in the 
> Internet.
>
>
> This is just like case 2 above except we don't expect a CoS treatment 
> at the
> network end of the access link.


What is missing in each scenario above is discussion of codec selection.

A valid and productive response to a pair of VoIP end systems becoming
aware of congestion someplace along the path between them would be to
switch to a lower-data-rate-codec.  This kind of behaviour is needed
regarless of IEprep, as part of normal Internet congestion avoidance
and control.   And there is certainly room to use codecs other than 
G.711,
as others have discussed here.  And either of those changes could well
lead to a reduction in the congestion on the link causing the two humans
to still be able to understand each other clearly (which is the 
functional
need), even though one might not be able to hear a pin drop at the 
remote
end.

Also missing is discussion of the desirability of using link-layer QoS
mechanisms (e.g. IEEE 802.1p) where they exist.

But overall, a helpful draft summary.

Ran
rja@extremenetworks.com



_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 09:24:11 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24189
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 09:24:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDEQds19906
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 09:26:39 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEQdv19903
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 09:26:39 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24144
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 09:23:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEP6v19850;
	Fri, 13 Dec 2002 09:25:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEObv19810
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 09:24:37 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24079
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 09:21:38 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.102])
	by gnat.inet.org (Postfix) with ESMTP id F025767106
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 09:24:44 -0500 (EST)
Date: Fri, 13 Dec 2002 09:21:27 -0500
Mime-Version: 1.0 (Apple Message framework v548)
Content-Type: text/plain; charset=US-ASCII; format=flowed
From: RJ Atkinson <rja@extremenetworks.com>
To: ieprep <ieprep@ietf.org>
Content-Transfer-Encoding: 7bit
Message-Id: <2AD3E836-0EA6-11D7-BE1C-00039357A82A@extremenetworks.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] Yet another scenario
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


Here is some discussion of a specific deployment strategy that relates
to the situation where one has authorised access to a cable modem
and the CATV firm associated with that cable modem deployment has signed
suitable agreements.  I think it requires some unreasonable assumptions
(for today), so I'll try to sort my assumptions into 2 buckets.  I'm
certain that there are other issues and assumptions, so this won't be
a sufficient discussion...

Reasonable Assumptions:
	- NCS (or whoever) and CATV firm that owns the HFC and the cable 
routers
	  that are deployed have signed some agreements.
	- End-user (e.g. owner of cable modem) that is customer of the CATV 
firm
	  agrees to let their link get used.


Unreasonable Assumptions:
	- DOCSIS 1.1 (or later) is widely deployed, so is present in the 
scenario,
		both in the cable router and in the cable modem -- and the equipment
		has been configured to enable the 1.1 feature set  [This might happen
	  	in future, but is not generally true today.]
	- The emergency site has power and the entire path of HFC to the 
emergency
	  site has power (e.g. for lasers and RF amps) and is undamaged, the
	  headend is undamaged, and an IP path exists from the cable modem to 
the
	  destination.  Also, the CATV firm's DOCSIS provisioning system (and 
any
	  related necessary servers) are up and fully operational.
	- The CATV firm's DOCSIS provisioning system supports configuration of 
all
	  DOCSIS 1.1 parameters on a modem-by-modem basis. [Today, I'm not 
aware of
	  any that do this.  Probably fixable by applying lots of money to the 
issue.]
	- The CATV firm has built a user-login web page that permits a user to 
login,
	  authenticate themselves, and request improved QoS.  Conceptually 
this might
	  tie back to the CATV firm's list of customers who paid more for 
their service
	  in order to be able to request QoS, to some magic list from the 
government,
	  or other such lists of users and their permissions.

Background:
	- With DOCSIS cable modems, there is a shared downstream.  In typical 
deployments,
	  64QAM modulation is used on a 6 MHz carrier.  Allowing for overhead, 
this
	  means about 27 Mbps of aggregate (shared) downstream capacity is 
available.
	- With DOCSIS cable modems, there are typically N upstreams (for N 
usually
	  in the range of 4 to 8) for each downstream.  These are normally 
space-division
	  multiplexed, so only one upstream is typically useable by a given 
cable modem.
	- In typical HFC plant in North America [1], one can only use QPSK 
modulation.
	  Normal upstream channel width is 1.5 MHz.  This yields about 1.5 
Mbps of
	  aggregate (shared) upstream capacity.
	- A typical DOCSIS cable modem has one RJ-45 10baseT Ethernet jack and 
one
	  RF interface.
	- DOCSIS 1.0 (and later) includes support for upstream and downstream 
rate-limits
	  on a per-modem basis.  In the better CM implementations, this is 
traffic-shaped.
	  In the lowest-cost implementations, clipping harmful to TCP goodput 
occurs.
	- DOCSIS 1.1 includes support for link-layer QoS (on a per-modem 
basis) that
	  is designed to support voice services ("PacketCable").
	- Depending on the HFC topology, HFC build parameters, and whatnot, 
there
	  might typically be hundreds of cable modems physically connected to 
a single
	  DOCSIS upstream.  Operational experience to date is that most of 
those CMs
	  are sending no user traffic most of the time.  Even during peak hours
	  (e.g. 4pm-9pm local time), a large percentage of them are idle at any
	  given moment in time.  Similarly, there might typically be hundreds 
to
	  a couple thousand modems connected to a single downstream.
	- In the DOCSIS standard, the operator of the cable router (CMTS) has 
total
	  control over the configuration of each individual cable modem.  The 
end user
	  does not have the ability to change QoS parameters or rate-limits or
	  what have you.

Concept of Operations:
	- The authorised emergency worker (with permission of whoever's cable 
modem
	  it is), connects a device with a web browser to the Ethernet.
	- Using that web browser, the worker types in a URL for the CATV firm's
	  magic QoS-enabling web page.   S/he then logs into the web page using
	  whichever method was pre-arranged (I'm guessing all this needs to be 
on
	  a wallet-sized card because its a lot to remember).  At that point, 
the
	  person requests that that particular cable modem be granted improved
	  service quality.  Improved service quality might mean removing an 
upstream
	  and/or downstream rate limit, enabling some of the DOCSIS 1.1 
features
	  designed to improve voice service quality, or something else 
entirely.
	- At this point the authorised emergency worker (with permission of 
whoever's
	  cable modem it is), connects their VoIP phone to the Cable modem.
	- From here, they'll need a SIP server that knows who they are, there 
will
	  need to be paths from there to the SIP server and paths from there 
to whichever
	  on-net destinations (PSTN gateway or whatever else) they want to 
reach,
	  and so forth.

Issues:
	- It is expensive and challenging to build a good DOCSIS provisioning 
system.
	  Most of the ones that exist today only support DOCSIS 1.0.  I'm 
personally
	  unaware of any that support configuration of all of the DOCSIS 1.1 
features.
	- A CATV firm might choose not to overprovision the links between the 
cable
	  router and the Tier-1 ISP uplink.  Because it is cheaper to 
overprovision
	  those links (CATV firms already have their own dark fibre in place), 
most
	  CATV firms overprovision from the cable router on out to the global 
Internet
	  today.  In short, this is a potential issue, but probably not a big 
issue,
	  because overprovisioning is common and congestion is normally 
limited to
	  the actual DOCSIS link.
	- For a DOCSIS provisioning system to be fully running, lots of pieces 
have
	  to all be working properly and have consistent configurations.  To 
boot,
	  a DOCSIS cable modem needs to talk with a DHCP server, a TFTP server,
	  a Time-of-Day server (not NTP alas), the cable router, and so forth.

Conclusion:
	For some kinds of IP networks, it might be practical and worthwhile to 
work
	with the related firms and how to use link-specific QoS mechanisms to 
support
	IEprep services.

	In particular, DOCSIS-based IP networks have broad (though not 
universal)
	coverage and about 75% of deployed cable modems connects to one of a
	relatively small number of firms.  This means that NCS (or whoever) 
might
	be able to gain a fair amount of geographic coverage with a relatively
	smaller number of contracts than for DSL, T1, or T3 end links.

Notes:

1:	In the Netherlands, typical HFC quality is MUCH higher than in North 
America
	or Australia, so they might actually be able to use 16QAM modulation 
on the
	upstream and thereby gain that much more link capacity.  I'm not well 
informed
	about typical HFC quality in Japan or other places.

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 09:33:00 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24330
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 09:33:00 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDEZTd20186
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 09:35:29 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEZQv20183
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 09:35:26 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24325
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 09:32:26 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEY3v20154;
	Fri, 13 Dec 2002 09:34:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEXBv20132
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 09:33:11 -0500
Received: from [209.22.88.17] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id JAA24304
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 09:30:11 -0500 (EST)
Received: from mtassp3.ncr.disa.mil by [209.22.88.17]
          via smtpd (for ietf-mx.ietf.org [132.151.6.1]) with SMTP; 13 Dec 2002 14:26:02 UT
Received: by mtassp3.ncr.disa.mil with Internet Mail Service (5.5.2653.19)
	id <YHPXSKQR>; Fri, 13 Dec 2002 09:33:11 -0500
Message-ID: <7F18415E4D63CB45BB9B3A591F68D12D5B3DE9@emshqs1.ncr.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        "King, Kimberly  S."
	 <KIMBERLY.S.KING@saic.com>
Cc: "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: Summary of IEPREP work and VoIP
Date: Fri, 13 Dec 2002 09:32:59 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Ran,

-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Friday, December 13, 2002 8:45 AM
To: King, Kimberly S.
Cc: 'ieprep '
Subject: [Ieprep] Re: Summary of IEPREP work and VoIP


>or SIP proxy firewalls or IP routers that are configured to enable 
>CBQ/CoS
>or link-layer QoS mechanisms)

>>  that are enhanced to only setup the highest priority calls
>> over the access link, up to a predetermined maximum influenced by the 
>> speed
>> of the access link, possibly even supporting preemption.

>That phrasing is awkward and ambiguous since the IP-layer still does not
>have the concept of a "call".  Presumably what one meant is something
>more like:

>...that are enhanced to provide preferred forwarding over the access 
>link
>to IP datagrams that are associated with specific authorised 
>emergency-related
>flows, up to some predetermined maximim influenced by the speed of the
>access link.  This preferred forwarding might involve CBQ in routers,
>lower-drop probability for authorised emergency-related IP packets,
>or other mechanisms.  Some such mechanisms might be inside the 
>link-layer,
>rather than in the IP-layer or above.

The wordings should be flexible to give room for folks who want to implement
QoS mechanisms in other layers such as IP and Transport. There are quite a
few ISPs who implement Diff-Serv in IP. In Transport layer I saw a proposal
from one fairly known company (which I don't want to name here) that wanted
to use its proprietary ping and traceroute software to map out the state of
the Internet and route IP traffic to AVAILABLE paths in real time. This way,
packet delay and jitter can be reduced substantially. I am not saying that I
like these TWO better than link-layer QoS. 

In short, my point is we don't want to rule out anything at this EARLY
stage. :-)

An


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 09:44:02 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24509
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 09:44:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDEkV121132
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 09:46:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEkUv21129
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 09:46:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24503
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 09:43:29 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEj1v21070;
	Fri, 13 Dec 2002 09:45:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEiKv21027
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 09:44:20 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24448
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 09:41:21 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Fri, 13 Dec 2002 09:43:39 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121309440026787
 ; Fri, 13 Dec 2002 09:44:00 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPH78M8>; Fri, 13 Dec 2002 09:45:34 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A62@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: "'ieprep '" <ieprep@ietf.org>
Date: Fri, 13 Dec 2002 09:44:00 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] RE: Summary of IEPREP work and VoIP
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


> On Thursday, Dec 12, 2002, at 18:09 America/Montreal, King, 
> Kimberly S. 
> wrote:
> > Case 1:  There is a predefined relationship between SIP 
> provider and 
> > ISP
> > providing access to the ETS location and this is known in 
> advance of an
> > emergency.  (The model might be like an enhanced GETS-like 
> WorldCom SIP
> > service).  In this case, there are two main things that can be done.
> >
> > o	Call signaling where some new resource priority mechanism (this
> > mechanism will hopefully to be created in SIPPING as a result of 
> > Henning's
> > requirements document) is in SIP that can be used to 
> identify calls for
> > preferential access to ports on PSTN gateways or other end systems.
> > o	If the provider runs stateful SIP proxies, the new 
> resource priority
> > mechanism could also allow preferential treatment of proxy 
> resources.
> > o	Depending upon how this provider engineers their 
> service (suppose
> > they want to be very stateful and track how many calls are coming 
> > to/from
> > which locations), a traditional call admission control 
> could possibly 
> > be
> > supported giving preferential treatment to emergency calls 
> by allowing 
> > them
> > a greater probably of call setup than regular calls.
> > o	We take the assumption (largely substantiated for many 
> U.S. regions)
> > that network congestion, if it exists, is on the access 
> link.  Hence, 
> > to
> > increase the likelihood of good sounding calls over an 
> access link, the
> > number of calls can be controlled locally at the customer 
> location.  
> > This
> > can be achieved through the use of devices (e.g., QoS 
> appliances or SIP
> > proxy firewalls)
> 
> or SIP proxy firewalls or IP routers that are configured to enable 
> CBQ/CoS
> or link-layer QoS mechanisms)


Actually, I'm talking purely SIP signaling at this point.  Only after that
do I consider the CBQ/CoS etc. for IP layer mechanisms.  What I mean above
is that there are literally SIP proxy-firewalls and QoS devises available
that can block the call signaling based upon a variety of factors including
the current number of active calls over the access link.  I've talked with
some of these vendors about the ability to modify their products to support
a preferential treatment whereby some users have a higher priority to setup
calls during congestion (as defined by the number of calls occurring over
the access link) and pre-emption of existing calls if desired.  They
indicate this functionality could be supported if the market calls for it.
I can provide more details if you think that would be constructive.
Remember this is done locally so this scales fine and doesn't involve the
service provider.

> 
> >  that are enhanced to only setup the highest priority calls
> > over the access link, up to a predetermined maximum 
> influenced by the 
> > speed
> > of the access link, possibly even supporting preemption.
> 
> That phrasing is awkward and ambiguous since the IP-layer 
> still does not
> have the concept of a "call".  

You don't get my phrasing here because I guess it wasn't clear enough that I
am only talking signaling at this point.  I'm not worried about the IP-layer
not understanding "calls", I am using devises that do.  Here's an example
firewall http://www.ingate.com/siparators.php that includes knowledge about
calls.

>Presumably what one meant is something
> more like:
> 
> ...that are enhanced to provide preferred forwarding over the access 
> link
> to IP datagrams that are associated with specific authorised 
> emergency-related
> flows, up to some predetermined maximim influenced by the speed of the
> access link.  This preferred forwarding might involve CBQ in routers,
> lower-drop probability for authorised emergency-related IP packets,
> or other mechanisms.  Some such mechanisms might be inside the 
> link-layer,
> rather than in the IP-layer or above.

No.  I'm hoping it's clear from the above but what I mean is put something
between the customer LAN and the access link that understands SIP signaling
and the maximum number of calls supported at any given time (this is exactly
what WorldCom SIP service is doing).  Now, modify this thing to track
priority of the given calls.  Either allow a greater probably of call setup
to priority calls when the number of calls approaches the limit or pre-empt
existing lower priority calls.  If the link supports 10 quality calls, then
all ten always sound good.  However, the 10 using the link are the highest
priority ones simply by virtue of call admission control (done locally).

It's actually an easy modification.  By using such a devise, one can
essentially make the link act somewhat like the PSTN.  Who gets to talk is
determined only by signaling, you admit the most important players and
everyone admitted gets quality sounding calls (because you predetermined how
much voice can be supported over the link and possibly use CoS to help that
quality even more).  Is this clear?

> Or, if a policy and deployment is agreed upon *by the 
> operators of that 
> link*,
> one might be able to prioritise selected IP flows (of any application 
> protocol).

Yes.

> What is missing in each scenario above is discussion of codec 
> selection.

This is an important topic and it would be nice to have discussions.
However, my summary doesn't include it because we really haven't reached any
conclusions on it yet.  It's a good topic.

> 
> Also missing is discussion of the desirability of using link-layer QoS
> mechanisms (e.g. IEEE 802.1p) where they exist.

One could add that.

> 
> But overall, a helpful draft summary.

Good.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 09:47:02 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24605
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 09:47:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDEnUc21265
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 09:49:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEnUv21262
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 09:49:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24596
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 09:46:30 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEm4v21227;
	Fri, 13 Dec 2002 09:48:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDEllv21200
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 09:47:47 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24538
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 09:44:48 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.102])
	by gnat.inet.org (Postfix) with ESMTP
	id E2D4C67106; Fri, 13 Dec 2002 09:47:54 -0500 (EST)
Date: Fri, 13 Dec 2002 09:47:38 -0500
Subject: Re: [Ieprep] Re: Summary of IEPREP work and VoIP
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep '" <ieprep@ietf.org>
To: "Nguyen, An" <nguyena@ncs.gov>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <7F18415E4D63CB45BB9B3A591F68D12D5B3DE9@emshqs1.ncr.disa.mil>
Message-Id: <D3268107-0EA9-11D7-BE1C-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Dec 13, 2002, at 09:32 America/Montreal, Nguyen, An wrote:
> The wordings should be flexible to give room for folks who want to 
> implement
> QoS mechanisms in other layers such as IP and Transport. There are 
> quite a
> few ISPs who implement Diff-Serv in IP. In Transport layer I saw a 
> proposal
> from one fairly known company (which I don't want to name here) that 
> wanted
> to use its proprietary ping and traceroute software to map out the 
> state of
> the Internet and route IP traffic to AVAILABLE paths in real time. 
> This way,
> packet delay and jitter can be reduced substantially. I am not saying 
> that I
> like these TWO better than link-layer QoS.

I've been agreeing that there are things one can do intra-domain with 
DiffServ;
DiffServ is an IP-layer function (despite the IETF putting the WG into 
the
"services" portion of the "Transport & Services Area").  There are 
potentials
for intra-domain IP-layer QoS and link-layer QoS.  I don't see how one 
could
do routing stuff above the IP-layer though.

I'd like to understand what stuff can reasonably done inside the 
Transport-layer
in this regard, where the usual definition of Transport-layer would 
include
TCP, SCTP, and UDP (maybe also XTP, but that's got nil deployment).  I 
think
of the transport-layer as including congestion avoidance and control, 
but
NOT path selection.  And I'd call any form of IP routing to be part of 
the
IP-layer.  (If you don't feel comfortable listing company names on an 
IETF
list, maybe you could email me a URL for whatever the public 
documentation
is for this other widget/feature/thing ?)

Maybe we're merely not understanding each other ?  I'm pretty sure that
I don't know what stuff could be put into the transport-layer that could
possibly affect IP routing (or path-selection, which is the same thing).

> In short, my point is we don't want to rule out anything at this EARLY
> stage. :-)

I'm generally agreeable that we don't want to rule stuff out.

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 10:02:32 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA24902
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 10:02:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDF51521819
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 10:05:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDF51v21815
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 10:05:01 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA24898
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 10:02:01 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDF32v21744;
	Fri, 13 Dec 2002 10:03:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDF28v21724
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 10:02:08 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA24825
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 09:59:08 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.102])
	by gnat.inet.org (Postfix) with ESMTP
	id 9D2AD67106; Fri, 13 Dec 2002 10:02:15 -0500 (EST)
Date: Fri, 13 Dec 2002 10:01:59 -0500
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep '" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A62@mcl-its-exs01.mail.saic.com>
Message-Id: <D44F2064-0EAB-11D7-BE1C-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] Re: Summary of IEPREP work and VoIP
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Dec 13, 2002, at 09:44 America/Montreal, King, Kimberly S. 
wrote:
> Actually, I'm talking purely SIP signaling at this point.  Only after 
> that
> do I consider the CBQ/CoS etc. for IP layer mechanisms.  What I mean 
> above
> is that there are literally SIP proxy-firewalls and QoS devises 
> available
> that can block the call signaling based upon a variety of factors 
> including
> the current number of active calls over the access link.

Ahh.  Could you please edit the phrase "QoS devices" to make it clear
that you are thinking about "SIP-layer-only QoS devices" ?  Not sure 
what
the right words would be, but the current ones are confusing.

> I've talked with
> some of these vendors about the ability to modify their products to 
> support
> a preferential treatment whereby some users have a higher priority to 
> setup
> calls during congestion (as defined by the number of calls occurring 
> over
> the access link) and pre-emption of existing calls if desired.  They
> indicate this functionality could be supported if the market calls for 
> it.

Seems plausible enough, with the understanding that it is a matter of
application-layer blocking for SIP signalling, rather than blocking
IP packets of any sort.

> You don't get my phrasing here because I guess it wasn't clear enough 
> that I
> am only talking signaling at this point.  I'm not worried about the 
> IP-layer
> not understanding "calls", I am using devises that do.

Could you please try to devise alternate, more clear, wording ?

 From the URL below, it appears that the widget you are thinking of 
operates
basically as an "application-layer gateway (ALG)" or "SIP proxy server".
Maybe you could work one or both of those phrases (whichever you think
is accurate) into the new text ?

> Here's an example
> firewall http://www.ingate.com/siparators.php that includes knowledge 
> about
> calls.

The URL definitely helps understand what you mean.

> No.  I'm hoping it's clear from the above but what I mean is put 
> something
> between the customer LAN and the access link that understands SIP 
> signaling
> and the maximum number of calls supported at any given time (this is 
> exactly
> what WorldCom SIP service is doing).  Now, modify this thing to track
> priority of the given calls.  Either allow a greater probably of call 
> setup
> to priority calls when the number of calls approaches the limit or 
> pre-empt
> existing lower priority calls.  If the link supports 10 quality calls, 
> then
> all ten always sound good.  However, the 10 using the link are the 
> highest
> priority ones simply by virtue of call admission control (done 
> locally).
>
> It's actually an easy modification.  By using such a devise, one can
> essentially make the link act somewhat like the PSTN.  Who gets to 
> talk is
> determined only by signaling, you admit the most important players and
> everyone admitted gets quality sounding calls (because you 
> predetermined how
> much voice can be supported over the link and possibly use CoS to help 
> that
> quality even more).  Is this clear?

The two paragraphs above increase clarity a whole lot.  Please maybe 
take
some of that text into your next revision of the "VoIP summary" in an
effort to make things clear all around.

Now, what happens if the same access link is being used concurrently
for non-SIP/non-VoIP traffic (e.g. a real-time game that is UDP-based
and lacks congestion avoidance and control) ?

Is that not still a problem...

>> What is missing in each scenario above is discussion of codec
>> selection.
>
> This is an important topic and it would be nice to have discussions.
> However, my summary doesn't include it because we really haven't 
> reached any
> conclusions on it yet.  It's a good topic.

Maybe add a place holder ?
	"Open Issues:
		- Initial codec selection
		- Codec changes due to congestion

>> Also missing is discussion of the desirability of using link-layer QoS
>> mechanisms (e.g. IEEE 802.1p) where they exist.
>
> One could add that.

Unless one is trying to rule that out, maybe that could be added.  
Proposed
text (edit to taste) might be:

	"Link-layer QoS mechanisms should be considered for deployment as part
	of an overall system, when such mechanisms exist and can be applied
	to a particular IEprep scenario."

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 10:15:06 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25324
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 10:15:05 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDFHY722927
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 10:17:34 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFHYv22924
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 10:17:34 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25304
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 10:14:34 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFG2v22867;
	Fri, 13 Dec 2002 10:16:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFFdv22835
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 10:15:39 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25248
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 10:12:40 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Fri, 13 Dec 2002 10:15:01 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121310152213719
 ; Fri, 13 Dec 2002 10:15:22 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YW14DYXF>; Fri, 13 Dec 2002 10:15:08 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A63@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: "'ieprep '" <ieprep@ietf.org>
Date: Fri, 13 Dec 2002 10:15:20 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] RE: Summary of IEPREP work and VoIP
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


> 
> Ahh.  Could you please edit the phrase "QoS devices" to make it clear
> that you are thinking about "SIP-layer-only QoS devices" ?  Not sure 
> what
> the right words would be, but the current ones are confusing.

I agree.  The one I am most familiar with, and could do the job is by Sitara
Networks http://www.sitaranetworks.com/ .  It is sometimes referred to as a
QoS appliance but I definitely see the possibility for confusion in the name
and will indeed think of better terminology.


> 
> Seems plausible enough, with the understanding that it is a matter of
> application-layer blocking for SIP signalling, rather than blocking
> IP packets of any sort.

Absolutely, you block the SIP signaling for lower priority users so their
media streams can't setup and therefore no IP packet blocking involved.

> 
> The two paragraphs above increase clarity a whole lot.  Please maybe 
> take
> some of that text into your next revision of the "VoIP summary" in an
> effort to make things clear all around.

Yes, I'll make another version of the Summary using clearer words and
sections to further indicate form.

> 
> Now, what happens if the same access link is being used concurrently
> for non-SIP/non-VoIP traffic (e.g. a real-time game that is UDP-based
> and lacks congestion avoidance and control) ?
> 
> Is that not still a problem...

With the Sitara Network product (or rather what they told me and what I
witnessed in demo), they can handle this by putting it into a lower priority
queue.  It's kind of a DiffServ approach with application knowledge that
does the packet ordering before hitting the access router.  This is just one
possibility but it exist!


> 
> Maybe add a place holder ?
> 	"Open Issues:
> 		- Initial codec selection
> 		- Codec changes due to congestion

Yes.

> 
> >> Also missing is discussion of the desirability of using 
> link-layer QoS
> >> mechanisms (e.g. IEEE 802.1p) where they exist.
> >
> > One could add that.
> 
> Unless one is trying to rule that out, maybe that could be added.  
> Proposed
> text (edit to taste) might be:
> 
> 	"Link-layer QoS mechanisms should be considered for 
> deployment as part
> 	of an overall system, when such mechanisms exist and 
> can be applied
> 	to a particular IEprep scenario."

Sounds good.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 10:23:32 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25535
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 10:23:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDFQ1k23258
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 10:26:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFQ1v23255
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 10:26:01 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25521
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 10:23:01 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFOQv23226;
	Fri, 13 Dec 2002 10:24:26 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFNOv23162
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 10:23:24 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25469
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 10:20:24 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Fri, 13 Dec 2002 10:22:40 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121310230322357
 for <ieprep@ietf.org>; Fri, 13 Dec 2002 10:23:03 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPH7949>; Fri, 13 Dec 2002 10:24:36 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A64@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "Ieprep (E-mail)" <ieprep@ietf.org>
Date: Fri, 13 Dec 2002 10:23:02 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] National Association for Ham Radio Operators--Internet Use
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


People in my organization (ARRL Hq, the national association for ham radio
operators) have spoken over a year ago with people in your organization, but
I wanted to respond to your e-mail.  

On a daily basis, ham radio operators use our equipment for emergency
communications.  But we also use the Internet to communicate -- for
instance, it is one of several methods our local volunteers use to call out
more volunteers, and that we here at Hq use to spread the word when the FCC
declares emergency frequencies to be used in a widespread disaster.  We also
sometimes tie our equipment in with the Internet, but not so much for
emergencies.  We use the Internet to tell the world about hams' emergency
response during (and after) disasters. 

We know there is always the chance that the Internet will fail or the power
will be off, however.  So we don't rely on it.  Hams don't have an
infrastructure to accomodate our communications, so we aren't held back if
the Internet fails.

Hams are experimenting with a lot of new things involving the Internet,
including a recent test with the National Weather Service using something
called Internet Radio Linking Project (IRLP).  IRLP is when Amateur Radio
repeaters (receivers and transmitters that extend the distance that hams can
operate on the VHF and UHF frequencies) are linked via the Internet instead
of via radio. A little about some of the other Internet radio experiments is
below, and comes from a recent story on our Web site at
http://www.arrl.org/news/features/2002/04/06/1/: 

Best regards, 
        Rosalie White, K1STO 

"By way of the Internet, you can make Amateur Radio contacts. If your
computer is equipped with a microphone and a soundcard, all you need is the
right software and you can talk to hams on the other side of the planet via
their local repeater or simplex channel. Gateways provide the link between
the Internet and radio worlds and the gateways are two-way streets.  That
means that if you are within radio-shot of a gateway, you can use your
transceiver to communicate through that local gateway to talk with the hams
that are connected to the gateway via the Internet. If your radio has DTMF
capabilities, you can use its tones to command the gateway to connect with
other gateways located throughout the world. Thus, you can contact another
ham who is on the air within radio-shot of the remote gateway via this
RF-to-gateway-to-Internet-to-gateway-to-RF link. You can learn all about the
Internet-ham radio connection at the 'Amateur Radio Voice-Links via the
Internet' Web page built by Andy, GJ7JHF. He also addresses the question, Is
it really ham radio? According to GJ7JHF, if RF is involved (and it is),
then it really is ham radio!" 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 10:33:01 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25855
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 10:33:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDFZUU23884
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 10:35:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFZUv23881
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 10:35:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25850
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 10:32:30 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFY2v23811;
	Fri, 13 Dec 2002 10:34:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFXuv23796
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 10:33:56 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25803
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 10:30:56 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.109])
	by gnat.inet.org (Postfix) with ESMTP
	id EABD567106; Fri, 13 Dec 2002 10:34:04 -0500 (EST)
Date: Fri, 13 Dec 2002 10:33:48 -0500
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep '" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A63@mcl-its-exs01.mail.saic.com>
Message-Id: <464789F4-0EB0-11D7-BE1C-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] Re: Summary of IEPREP work and VoIP
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Dec 13, 2002, at 10:15 America/Montreal, King, Kimberly S. 
wrote:
>  It is sometimes referred to as a
> QoS appliance but I definitely see the possibility for confusion in 
> the name
> and will indeed think of better terminology.

Thanks.

>> Seems plausible enough, with the understanding that it is a matter of
>> application-layer blocking for SIP signalling, rather than blocking
>> IP packets of any sort.
>
> Absolutely, you block the SIP signaling for lower priority users so 
> their
> media streams can't setup and therefore no IP packet blocking involved.
>
>> The two paragraphs above increase clarity a whole lot.  Please maybe
>> take some of that text into your next revision of the "VoIP summary" 
>> in an
>> effort to make things clear all around.
>
> Yes, I'll make another version of the Summary using clearer words and
> sections to further indicate form.
>
>> Now, what happens if the same access link is being used concurrently
>> for non-SIP/non-VoIP traffic (e.g. a real-time game that is UDP-based
>> and lacks congestion avoidance and control) ?
>>
>> Is that not still a problem...
>
> With the Sitara Network product (or rather what they told me and what I
> witnessed in demo), they can handle this by putting it into a lower 
> priority
> queue.  It's kind of a DiffServ approach with application knowledge 
> that
> does the packet ordering before hitting the access router.  This is 
> just one
> possibility but it exist!

Hmm.  So one puts this box inline before the router or between a pair
of routers or such like ?  It does kind of sound like it is doing an
IP-layer function but partly like an ALG.  Middlebox of some sort
either way.

Ran
rja@extremenetworks.com


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 10:40:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA26192
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 10:40:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDFgdm24829
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 10:42:39 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFgdv24826
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 10:42:39 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA26161
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 10:39:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFf2v24743;
	Fri, 13 Dec 2002 10:41:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDFevv24722
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 10:40:57 -0500
Received: from [209.22.88.17] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id KAA26019
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 10:37:52 -0500 (EST)
Received: from mtassp3.ncr.disa.mil by [209.22.88.17]
          via smtpd (for ietf-mx.ietf.org [132.151.6.1]) with SMTP; 13 Dec 2002 15:33:43 UT
Received: by mtassp3.ncr.disa.mil with Internet Mail Service (5.5.2653.19)
	id <YHPXSNHX>; Fri, 13 Dec 2002 10:40:52 -0500
Message-ID: <7F18415E4D63CB45BB9B3A591F68D12D5B3DEA@emshqs1.ncr.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'RJ Atkinson'" <rja@extremenetworks.com>
Cc: "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: Summary of IEPREP work and VoIP
Date: Fri, 13 Dec 2002 10:40:46 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>



-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Friday, December 13, 2002 9:48 AM
To: Nguyen, An
Cc: 'ieprep '
Subject: Re: [Ieprep] Re: Summary of IEPREP work and VoIP

>I'd like to understand what stuff can reasonably done inside the 
>Transport-layer
>in this regard, where the usual definition of Transport-layer would 
>include
>TCP, SCTP, and UDP (maybe also XTP, but that's got nil deployment).  I 
>think
>of the transport-layer as including congestion avoidance and control, 
>but
>NOT path selection.  And I'd call any form of IP routing to be part of 
>the
>IP-layer.  (If you don't feel comfortable listing company names on an 
>IETF
>list, maybe you could email me a URL for whatever the public 
>documentation
>is for this other widget/feature/thing ?)

>Maybe we're merely not understanding each other ?  I'm pretty sure that
>I don't know what stuff could be put into the transport-layer that could
>possibly affect IP routing (or path-selection, which is the same thing).

When I said QoS in the Transport layer, I did not mean proprietary QoS
mechanisms put into existing Transport-layer protocols such as TCP, UDP,
SCTP and others by this company to optimize the IP traffic for its
customers. Below is what I meant:

Because of this company's coverage in the Internet, it can deploy its
optimization scheme to take a snap shot of the Internet and determine the
best paths for its customers' IP traffic to take. What it does is ping and
traceroute to statistically determine how much TCP and UDP utilizations are
at certain time intervals and make routing decisions based on these
utilization numbers. I am talking about the interdomain traffic here. 

Interesting concept ... but it still has some drawbacks ... But, no one
solution can fix everything. :-) 

You know about this company ... I will send you an email later.

An
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 12:35:22 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29254
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 12:35:22 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDHbqT32001
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 12:37:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDHbqv31998
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 12:37:52 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29250
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 12:34:50 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDHa8v31329;
	Fri, 13 Dec 2002 12:36:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDHZUv31305
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 12:35:30 -0500
Received: from imo-d04.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29208
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 12:32:28 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d04.mx.aol.com (mail_out_v34.13.) id j.29.336e96fe (3850);
	Fri, 13 Dec 2002 12:35:20 -0500 (EST)
Message-ID: <29.336e96fe.2b2b7458@aol.com>
Date: Fri, 13 Dec 2002 12:35:20 EST
Subject: Re: [Ieprep] An alternative concept of IEprep
To: rja@extremenetworks.com, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_29.336e96fe.2b2b7458_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_29.336e96fe.2b2b7458_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/12/2002 1:28:27 PM Eastern Standard Time, 
rja@extremenetworks.com writes:


>    Personally, I find the Japanese deployment of IEprep to be very
> interesting.  Trying to take their concept and generalise it a bit,
> here is an alternative (non-voice) concept for IEprep for discussion.
> 
> Ran
> rja@extremenetworks.com
> 
> ------------------------------------------------------------------------
> 
> DEPLOYMENT CONCEPT:
> 
>     While dedicated servers are not required for this concept to work well
> and be very useful, dedicated servers could help improve service 
> availability
> and reduce communications latency.
> 
>     A good strategy for the servers would be to build several  
> high-availability
> server clusters.  High-availability meaning that if a component failed,  the
> system could quickly recover and the non-failed components would  continue 
> to
> work together, though possibility with diminished storage, I/O, or CPU
> capabilities.  HA server clusters are widely available COTS products, though
> they are often not cheap.
> 
>     Clusters would be placed in ISP co-location facilities, preferably in 
> the
> same bunker-style facilities used for the ISP's core routers.  The goal 
> would
> be to have at least one server cluster inside each large ISP's core.  
> Contracts
> could be signed with the ISPs to give service restoration to those  clusters
> priority -- and if co-located in the ISP's core router POPs, there is high
> probability that the ISP could get technical staff on site quickly if 
> needed.
> Being inside an ISP's core router POP also helps ensure service 
> availability,
> because the server cluster could easily have redundant connectivity to  
> 



[MAP] This would certainly be a good idea and needed to support an IAA type 
service, but I wonder several things:

Wouldn't much of what is described be the characteristics of the ISP's core 
routers anyway?

How would a user get "priority" to reach one of the (several) clusters? In 
the GETS model, one only has to reach a nearby switch that understands GETS 
to get "priority" for the rest of the call setup. (One comment was that 80% 
of the lines in the US are so equipped.)

Mike Pierce
Artel


--part1_29.336e96fe.2b2b7458_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/12/2002 1:28:27 PM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px"> &nbsp;&nbsp;Personally, I find the Japanese deployment of IEprep to be very
<BR>interesting. &nbsp;Trying to take their concept and generalise it a bit,
<BR>here is an alternative (non-voice) concept for IEprep for discussion.
<BR>
<BR>Ran
<BR>rja@extremenetworks.com
<BR>
<BR>------------------------------------------------------------------------
<BR>
<BR>DEPLOYMENT CONCEPT:
<BR>
<BR> &nbsp;&nbsp;&nbsp;While dedicated servers are not required for this concept to work well
<BR>and be very useful, dedicated servers could help improve service availability
<BR>and reduce communications latency.
<BR>
<BR> &nbsp;&nbsp;&nbsp;A good strategy for the servers would be to build several &nbsp;high-availability
<BR>server clusters. &nbsp;High-availability meaning that if a component failed, &nbsp;the
<BR>system could quickly recover and the non-failed components would &nbsp;continue to
<BR>work together, though possibility with diminished storage, I/O, or CPU
<BR>capabilities. &nbsp;HA server clusters are widely available COTS products, though
<BR>they are often not cheap.
<BR>
<BR> &nbsp;&nbsp;&nbsp;Clusters would be placed in ISP co-location facilities, preferably in the
<BR>same bunker-style facilities used for the ISP's core routers. &nbsp;The goal would
<BR>be to have at least one server cluster inside each large ISP's core. &nbsp;Contracts
<BR>could be signed with the ISPs to give service restoration to those &nbsp;clusters
<BR>priority -- and if co-located in the ISP's core router POPs, there is high
<BR>probability that the ISP could get technical staff on site quickly if needed.
<BR>Being inside an ISP's core router POP also helps ensure service availability,
<BR>because the server cluster could easily have redundant connectivity to &nbsp;
<BR>multiple core routers.</BLOCKQUOTE>
<BR>
<BR>
<BR>
<BR>[MAP] This would certainly be a good idea and needed to support an IAA type service, but I wonder several things:
<BR>
<BR>Wouldn't much of what is described be the characteristics of the ISP's core routers anyway?
<BR>
<BR>How would a user get "priority" to reach one of the (several) clusters? In the GETS model, one only has to reach a nearby switch that understands GETS to get "priority" for the rest of the call setup. (One comment was that 80% of the lines in the US are so equipped.)
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_29.336e96fe.2b2b7458_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 12:36:09 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29283
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 12:36:09 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDHceh32026
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 12:38:40 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDHcev32023
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 12:38:40 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29272
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 12:35:38 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDHb1v31381;
	Fri, 13 Dec 2002 12:37:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDHaMv31337
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 12:36:22 -0500
Received: from imo-r07.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA29222
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 12:33:20 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-r07.mx.aol.com (mail_out_v34.13.) id j.12b.1d67a894 (3850);
	Fri, 13 Dec 2002 12:35:59 -0500 (EST)
Message-ID: <12b.1d67a894.2b2b747f@aol.com>
Date: Fri, 13 Dec 2002 12:35:59 EST
Subject: Re: [Ieprep] Re: Mike's scenario
To: rja@extremenetworks.com, fred@cisco.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_12b.1d67a894.2b2b747f_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_12b.1d67a894.2b2b747f_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/12/2002 4:34:56 PM Eastern Standard Time, 
rja@extremenetworks.com writes:


> Nope.  If (A) pays for an IP access link, then (A) gets to set the policy on 
> their
> link.  A's ISP does not get the power to set the policy on that link 
> unilaterally
> (generally speaking; the details of the ISP's contract with A obviously 
> matters here).
> The government does not get the power to set the policy on that link 
> unilaterally either.
> 
> If the government wants to have influence over A's access link(s), it seems 
> to several
> of us that the government needs to have agreements with both A and A's 
> ISP(s).  Absent
> specific agreements with A *and* A's ISP(s), then the government simply 
> does not have
> any right to influence the policy deployed on A's access link.  An agreement
> between a government and an ISP might impact how that ISP does things 
> *inside
> the ISP's own network*, but that contract(with the ISP)'s scope could not 
> legitimately
> include anything belonging to an ISP's customer (e.g. CPE access router on 
> the access
> link that is owned/operated by the customer).
> 


[MAP]
I see where a lot of the problem is. Ran is not addressing the scenario of 
most interest to me (as a DoD contractor, not an NCS contractor). The NCS 
interest is mostly for authorized emergency use of the "public" 
infrastructure, with the practical problems as stated above, but, as Fred 
says, carriers would likely be more than happy to get an additional revenue 
stream from the government to do these things.

However, I am addressng the DoD interests, which are authorized emergency 
access mostly over DoD owned (or contracted) facilities. In this case, the 
government (DoD) is both "A" and "A's ISP". In this scenario, it needs 
procedures/protocols/whatever to meet the requirements that have been 
described in my draft as well as in presentations in SIPPING and IEPREP.

I described the case of the (civilian) call from Nairobi on a limited access 
going over satellite, which can certainly become congested if too many people 
independently try to make voice calls (as stated during the presentation), 
since this is exactly one of the valid cases for a military operation, which 
often occur in unplanned places. (This includes the US military going into 
other countries to lend humanitarian aid following a disaster.) But, I've 
quickly realized that it's best to not emphasize the "military" aspect of the 
use of this requirement and that it is best to include other possible 
applications in the problem set in order for the IETF to spend its time on 
possible solutions. I've tried to describe the scenarios without resorting to 
a lot of military application language, but I guess that has led to some 
confusion.

In a message dated 12/12/2002 4:34:56 PM Eastern Standard Time, 
rja@extremenetworks.com writes:


> Instead, the repeated misunderstanding from NCS and its contractors is that 
> its
> sufficient to just have contracts with a few big ISPs.  With the Internet's
> de-centralised architecture (opposite to PSTN's centralised architecture),
> the set of agreements needed is pretty significantly larger than with the
> existing PSTN.
> 


[MAP] I really don't think the difficulty of contracting with many ISPs 
should be an issue here. I'm sure the people at NCS understand the situation. 
The only point here should be whether something can be defined which NCS can 
take to ISPs to ask for.

Mike Pierce
Artel






--part1_12b.1d67a894.2b2b747f_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/12/2002 4:34:56 PM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">Nope. &nbsp;If (A) pays for an IP access link, then (A) gets to set the policy on their
<BR>link. &nbsp;A's ISP does not get the power to set the policy on that link unilaterally
<BR>(generally speaking; the details of the ISP's contract with A obviously matters here).
<BR>The government does not get the power to set the policy on that link 
<BR>unilaterally either.
<BR>
<BR>If the government wants to have influence over A's access link(s), it seems to several
<BR>of us that the government needs to have agreements with both A and A's ISP(s). &nbsp;Absent
<BR>specific agreements with A *and* A's ISP(s), then the government simply does not have
<BR>any right to influence the policy deployed on A's access link. &nbsp;An agreement
<BR>between a government and an ISP might impact how that ISP does things *inside
<BR>the ISP's own network*, but that contract(with the ISP)'s scope could not legitimately
<BR>include anything belonging to an ISP's customer (e.g. CPE access router on the access
<BR>link that is owned/operated by the customer).
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>[MAP]
<BR>I see where a lot of the problem is. Ran is not addressing the scenario of most interest to me (as a DoD contractor, not an NCS contractor). The NCS interest is mostly for authorized emergency use of the "public" infrastructure, with the practical problems as stated above, but, as Fred says, carriers would likely be more than happy to get an additional revenue stream from the government to do these things.
<BR>
<BR>However, I am addressng the DoD interests, which are authorized emergency access mostly over DoD owned (or contracted) facilities. In this case, the government (DoD) is both "A" and "A's ISP". In this scenario, it needs procedures/protocols/whatever to meet the requirements that have been described in my draft as well as in presentations in SIPPING and IEPREP.
<BR>
<BR>I described the case of the (civilian) call from Nairobi on a limited access going over satellite, which can certainly become congested if too many people independently try to make voice calls (as stated during the presentation), since this is exactly one of the valid cases for a military operation, which often occur in unplanned places. (This includes the US military going into other countries to lend humanitarian aid following a disaster.) But, I've quickly realized that it's best to not emphasize the "military" aspect of the use of this requirement and that it is best to include other possible applications in the problem set in order for the IETF to spend its time on possible solutions. I've tried to describe the scenarios without resorting to a lot of military application language, but I guess that has led to some confusion.
<BR>
<BR>In a message dated 12/12/2002 4:34:56 PM Eastern Standard Time, rja@extremenetworks.com writes:
<BR>
<BR>
<BR><BLOCKQUOTE TYPE=CITE style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">Instead, the repeated misunderstanding from NCS and its contractors is that its
<BR>sufficient to just have contracts with a few big ISPs. &nbsp;With the Internet's
<BR>de-centralised architecture (opposite to PSTN's centralised architecture),
<BR>the set of agreements needed is pretty significantly larger than with the
<BR>existing PSTN.
<BR></FONT><FONT  COLOR="#000000" SIZE=3 FAMILY="SANSSERIF" FACE="Arial" LANG="0"></BLOCKQUOTE>
<BR></FONT><FONT  COLOR="#000000" SIZE=2 FAMILY="SANSSERIF" FACE="Arial" LANG="0">
<BR>
<BR>[MAP] I really don't think the difficulty of contracting with many ISPs should be an issue here. I'm sure the people at NCS understand the situation. The only point here should be whether something can be defined which NCS can take to ISPs to ask for.
<BR>
<BR>Mike Pierce
<BR>Artel
<BR>
<BR>
<BR>
<BR>
<BR></FONT></HTML>

--part1_12b.1d67a894.2b2b747f_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 13:18:04 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA00663
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 13:18:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDIKaC02035
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 13:20:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDIKav02032
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 13:20:36 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA00645
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 13:17:33 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDIJ1v01924;
	Fri, 13 Dec 2002 13:19:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDIHdv01844
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 13:17:39 -0500
Received: from imo-m02.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA00559
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 13:14:37 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m02.mx.aol.com (mail_out_v34.13.) id j.164.187e80ec (4312);
	Fri, 13 Dec 2002 13:17:28 -0500 (EST)
Message-ID: <164.187e80ec.2b2b7e38@aol.com>
Date: Fri, 13 Dec 2002 13:17:28 EST
Subject: Re: [Ieprep] updated requirements documents (telephony)
To: rja@extremenetworks.com
CC: K.Carlberg@cs.ucl.ac.uk, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
X-Mailer: AOL 6.0 for Windows XP US sub 51
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

In a message dated 12/13/2002 8:30:02 AM Eastern Standard Time, 
rja@extremenetworks.com writes:


On Thursday, Dec 12, 2002, at 16:47 America/Montreal, Mpierce1@aol.com 
wrote:
<snip>

[RJA] The clarity is helpful and appreciated, but please see Joel Halpern's 
previous
commentary on QoS-based routing.  There is significant negative experience 
with
routing schemes that try to include congestion as part of the routing metric.
What happens is an oscillation where traffic switches from link A to link B
because A is getting full, then link B gets full, and the traffic all switches
back.  A better approach, that is widely available today, is to use other
techniques (within a domain, ECMP is one example or MPLS-TE might be 
another).


[MAP] I understand the dificulty. That is why the existing text is in the 
"Issues" section of the draft, not in the "Requirements". The draft says 
"this utility may be difficult to achieve", so I read that to say "it might 
be possible". I didn't propose changing that idea.

[MAP]
> I suggest this section on Alternate Paths be split into two separate 
> sub-parts to address the two "layers" within IP and relate each to 
> what the PSTN did (or couldn't do) as follows:
>
> 1) Alternate paths
>
> Experience with GETS over the PSTN has shown the utility of alternate 
> paths to a destination to help facilitate emergency-related 
> communications. This is done at the application layer, independent of 
> any lower layer rerouting (SONET rings, etc) that naturally applies to 
> all traffic using the link, whether emergency or normal. Application 
> of emergency telephony services to the Internet (and IP in general) 
> means that there are two separate issues to be addressed.

[RJA] What do you mean by "application-layer" and "alternate paths" ?


[MAP] See comment below.


[RJA] In the case of the Internet, path selection is an IP-layer function 
performed
by the routing system, not an application-layer function.


[MAP} Which is mainly why I made the (somewhat editorial) suggestion to split 
the text into two sub-sections, to address the two layers separately. I think 
a lot of our disagreement is that I make statements pertaining to the 
"application layer" and some immediately think about the IP-layer. (like most 
of the discussion about Accounting).

[MAP]
> a) IP Packet Transport layer: Unlike the PSTN, which creates a fixed 
> path (circuit) during call setup phase, the Internet uses dynamic 
> routing for IP packets. This dynamic routing capability automatically 
> causes IP packets to travel the best current path, which may change 
> during the life of the "call" due to failures of infrastructure 
> components. However, at the time of "call" set up (at the application 
> layer), the route choice for the packet transport layer has a single 
> choice (the best current one). An issue to be addressed is the utility 
> and possibility of providing additional routes (even if less 
> efficient) for packet flows for higher priority "calls".

[RJA] I definitely don't follow what you are trying to get at above.  Please 
try to
explain in more detail what you are trying to do here.



[MAP] I presume your question is related to the final sentence, since the 
rest were either copied direct from the draft or are simple statements of 
fact. I believe the final sentence, which relates to the possiblity of 
providing methods to have additional routes for the packet flows for 
emergency traffic, is straigthforward (to descibe, not necessarily to 
actually do). If it can't do something similar, then this would be one area 
in which the Internet would be inferior to the PSTN.

It has been identified that there is no capability in the Internet to route 
around congestion. Emergency service users believe they have a need for this 
capability ("requirement", in their vocabulary).

[MAP]
> b) Call application layer:  The Internet infrastructure itself does 
> not have the concept of a "call" or the concept of "call setup", 
> although IP telephony applications using it would have 
> application-layer awareness of calls or the call setup concept. 
> Alternate path routing at the application layer may be useful in 
> choosing one of possibly several egress points to the PSTN from the 
> Internet.

[RJA] Please explain what "alternate path routing at the application layer" 
is or
might be, with reference to the usual Internet Architecture.


[MAP] The above sentence about "alternate path routing at the application 
layer" was copied verbatim from the draft (which has your name on it). I was 
simply putting it under a heading that said "Call application layer". What 
did it mean in the original draft?

Mike Pierce
Artel
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 13:20:03 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA00737
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 13:20:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDIMZT02168
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 13:22:35 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDIMYv02165
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 13:22:34 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA00728
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 13:19:32 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDIL2v02091;
	Fri, 13 Dec 2002 13:21:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDIK2v02008
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 13:20:02 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA00631
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 13:16:59 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.123])
	by gnat.inet.org (Postfix) with ESMTP
	id 7CDA867105; Fri, 13 Dec 2002 13:20:09 -0500 (EST)
Date: Fri, 13 Dec 2002 13:19:52 -0500
Subject: Re: [Ieprep] An alternative concept of IEprep
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep@ietf.org
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <29.336e96fe.2b2b7458@aol.com>
Message-Id: <7914D006-0EC7-11D7-9A70-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Dec 13, 2002, at 12:35 America/Montreal, Mpierce1@aol.com 
wrote:
> In[MAP] This would certainly be a good idea and needed to support an 
> IAA type
> service, but I wonder several things:
>
> Wouldn't much of what is described be the characteristics of the ISP's 
> core routers anyway?

	Hmm.  Sorry I was confusing.  Not exactly sure which bits I messed up 
on,
but I'll give it a try.  When I gave details about HA, I was talking 
about a server
in the sense of like an IBM AIX box (or a Sun Solaris server or 
whatever brand)
-- to host whichever applications (e.g. web, email, IRC, or whatever).  
Apologies
if I got that confused with an ISP's core router, which as you say, 
might have HA
properties for its own reasons.

	If this is not clear or I've answer the wrong question, let me know 
what
is unclear and I'll try again.

> How would a user get "priority" to reach one of the (several) 
> clusters? In the GETS model, one only has to reach a nearby switch 
> that understands GETS to get "priority" for the rest of the call 
> setup. (One comment was that 80% of the lines in the US are so 
> equipped.)

	That scenario description was not so much trying to "solve" the 
"priority" issue,
as to come up with candidate applications that are useful and also work 
well even
during congested or low-bandwidth or other unhappy operating 
conditions.  That is,
these are things that might still work in operating environments where 
some/all
voice codecs might not work well in.

	The same (router CoS/filtering/CBQ, switch CoS/filtering/CBQ) 
mechanisms
can be used for access links.  The one good thing here about IP-layer 
or link-layer
mechanisms is that they can work for any sort of traffic.  If there are 
a small
number of such server clusters, then one might be able to get dedicated 
access
link preference for all traffic to/from those clusters.  (The downside 
of deploying
it in that manner is that it would also give preference to any traffic 
from a
bad guy trying to reach those servers, so security issues remain.)

	In the core, the only thing I was proposing would be that ISP A and 
ISP B
would each have contracts with NCS (or whoever) that gave restoration 
of service
between that ISP's core and the co-located server cluster priority and 
that the
core links inter-connecting A's core and B's core also get priority 
restoration.
Within a given core, the concept was relying on the usual 
over-provisioning strategy.


(MORE GENERICALLY, NOT TIED TO THIS CONCEPT, ...)

	An advantage of the PSTN is that maybe 10K to 30K POTS lines terminate
in a single device (Class 5 switch) and control is centralised in that 
device.
I can't think of any deployed router that has 10K tail circuits 
attached to it
(or even close to that).  And control is de-centralised, so each end of 
the circuit
has some control.

	The typical ISP T1/E1 deployment topology (with ISP router on one end 
and
a customer router on the other end) looks somewhat like a Class-5 
switch connected
via a T1 or ISDN PRI to an over-subscribed corporate PBX than like a 
POTS line
to a house.  What is the GETS strategy when the phone is behind a 
corporate PBX
and that PBX has more offered load in outbound call attempts than it 
has capacity
to connect to the Class-5 phone switch ?  Does NCS have a special 710 
contract
with most corporate PBX owners ?  Is it just best effort in that case ? 
  I'm curious.
And what about PBXs that haven't been configured to know about 710 at 
all ?

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 13:55:06 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA01730
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 13:55:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDIvZG03959
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 13:57:35 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDIvZv03956
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 13:57:35 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA01725
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 13:54:35 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDIu4v03915;
	Fri, 13 Dec 2002 13:56:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDItfv03896
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 13:55:41 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA01706
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 13:52:40 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Fri, 13 Dec 2002 13:54:58 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121313552027861
 ; Fri, 13 Dec 2002 13:55:20 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YW141S7F>; Fri, 13 Dec 2002 13:55:06 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A6D@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>, Mpierce1@aol.com
Cc: ieprep@ietf.org
Date: Fri, 13 Dec 2002 13:55:17 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] GETS calls placed from corporate PBXs
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

>What is the GETS strategy when the phone is behind a 
> corporate PBX
> and that PBX has more offered load in outbound call attempts than it 
> has capacity
> to connect to the Class-5 phone switch ?  

I think (and Janet or others please correct me if I'm wrong here) that the
GETS user would get a busy signal, same as any other user if the link to the
PSTN is filled.  (This is different behavior than a PBX that is enabled to
support MLPP in DoD.)

>Does NCS have a special 710 
> contract
> with most corporate PBX owners ?  

This would be news to me.  I don't think so.

>Is it just best effort in 
> that case ? 
>   I'm curious.

Yes.

> And what about PBXs that haven't been configured to know about 710 at 
> all ?

GETS operational support folks, if they discover problems (hotels are a
common examples where PBXs might not understand 710) start a process to get
the PBX configured correctly.  Sometimes users discover and report problems
too.  The point is, there is a way to fix that problem.  

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 15:34:35 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04680
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 15:34:35 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDKb6K10096
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 15:37:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDKb6v10085
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 15:37:06 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04675
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 15:34:03 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDKZRv09797;
	Fri, 13 Dec 2002 15:35:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDKYEv09739
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 15:34:14 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04591
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 15:31:11 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.138])
	by gnat.inet.org (Postfix) with ESMTP
	id EA86167106; Fri, 13 Dec 2002 15:34:19 -0500 (EST)
Date: Fri, 13 Dec 2002 15:32:58 -0500
Subject: Re: [Ieprep] Re: Mike's scenario
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <12b.1d67a894.2b2b747f@aol.com>
Message-Id: <1173BF04-0EDA-11D7-9A70-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Dec 13, 2002, at 12:35 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP]
> However, I am addressng the DoD interests, which are authorized 
> emergency access mostly over DoD owned (or contracted) facilities. In 
> this case, the government (DoD) is both "A" and "A's ISP".

My bad.  Terribly sorry, thanks for the correction.

> In this scenario, it needs procedures/protocols/whatever to meet the 
> requirements that have been described in my draft as well as in 
> presentations in SIPPING and IEPREP.

And this is explicitly a VoIP scenario or a any-kind-of-application 
over IP scenario ?

> I described the case of the (civilian) call from Nairobi on a limited 
> access going over satellite, which can certainly become congested if 
> too many people independently try to make voice calls (as stated 
> during the presentation), since this is exactly one of the valid cases 
> for a military operation, which often occur in unplanned places. (This 
> includes the US military going into other countries to lend 
> humanitarian aid following a disaster.) But, I've quickly realized 
> that it's best to not emphasize the "military" aspect of the use of 
> this requirement and that it is best to include other possible 
> applications in the problem set in order for the IETF to spend its 
> time on possible solutions. I've tried to describe the scenarios 
> without resorting to a lot of military application language, but I 
> guess that has led to some confusion.

Yes, but no real harm done.

There is one really really fundamental difference between a DISN-only 
scenario
or a DoD-only scenario and an NCS scenario:

	A DoD-only or DISN-only scenario has a single administrative domain.
	The civilian IEprep scenario has arbitrary numbers of administrative 
domains.

A direct result is that those 2 scenarios probably need totally 
different
system engineering approaches.  They might be able to reuse some 
mechanisms
here and there, but the overall deployment concept is significantly 
different.

For a single-domain approach, where some single policy can be applied 
all
over, then end-to-end DiffServ is an approach that the US DoD used in 
the
past in the DDN. Modulo the operational issues with "precedence creep" 
that
were described by Rex a while back, it should just work fine.  The IETF
redefined the IP ToS bits since then, but most products are generic 
enough
to implement at least the Precedence Bits from RFC-791 without any 
change
to anything (as in shipping code today from many vendors).  I can 
imagine
some desire to get the original precedence scheme documented publicly,
but if that's to be undertaken as an IETF standards activity, then it 
really
is a DiffServ WG matter.

Similarly, I can totally believe that DoD might have accounting needs 
for
its own networks.  But it is a pretty far stretch to go from those needs
to say that all civilian IEprep scenarios will need accounting.  And
I'd still like to understand more nearly what kinds of things this part
of the DoD community [1] is looking to perform accounting upon.  Details
would help a great deal.

So it would help a lot if the non-civilian participants on this list 
would
please clearly differentiate "civilian IEprep" scenarios from the 
"military"
scenarios.

Ran
rja@extremenetworks.com

[1] US DoD is big enough that it probably doesn't have any single view 
on any topic.
Folks in various places espouse various policies, but DoD is big enough 
that almost
no policy really gets universally implemented.
	

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 15:48:54 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04879
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 15:48:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDKpPm10834
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 15:51:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDKpPv10831
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 15:51:25 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04835
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 15:48:22 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDKo2v10794;
	Fri, 13 Dec 2002 15:50:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDKnEv10759
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 15:49:14 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04805
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 15:46:11 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.138])
	by gnat.inet.org (Postfix) with ESMTP
	id 12A6A67105; Fri, 13 Dec 2002 15:49:21 -0500 (EST)
Date: Fri, 13 Dec 2002 15:49:03 -0500
Subject: Re: [Ieprep] updated requirements documents (telephony)
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: K.Carlberg@cs.ucl.ac.uk, ieprep@ietf.org
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <164.187e80ec.2b2b7e38@aol.com>
Message-Id: <5042D0E8-0EDC-11D7-9A70-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Dec 13, 2002, at 13:17 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP} Which is mainly why I made the (somewhat editorial) suggestion 
> to split
> the text into two sub-sections, to address the two layers separately. 
> I think
> a lot of our disagreement is that I make statements pertaining to the
> "application layer" and some immediately think about the IP-layer. 
> (like most
> of the discussion about Accounting).

I'm absolutely an IP-layer kind of person.  So unless something has a 
clearly
marked layer, my head will pretty much always start with the IP layer.

And it would help the Accounting thread if you'd call it "SIP 
Accounting"
if the scope is limited to SIP or "Application-layer accounting" if the
scope is limited to stuff in the application-layer.  This was not clear
either from the presentation or from previous discussions.  I'd still
like to understand what specific things one is trying to perform
accounting upon.

> [MAP] I presume your question is related to the final sentence, since 
> the
> rest were either copied direct from the draft or are simple statements 
> of
> fact. I believe the final sentence, which relates to the possiblity of
> providing methods to have additional routes for the packet flows for
> emergency traffic, is straigthforward (to descibe, not necessarily to
> actually do). If it can't do something similar, then this would be one 
> area
> in which the Internet would be inferior to the PSTN.

Lets avoid flamage.  I'll silently s/inferior/different/ for everyone's 
sanity.

> It has been identified that there is no capability in the Internet to 
> route
> around congestion. Emergency service users believe they have a need 
> for this
> capability ("requirement", in their vocabulary).

Actually, from your earlier note today, maybe that means that DoD IP 
network
standards folks believe they have a need for this.  If I understood 
your note
earlier correctly, you are concerned with the military scenario, rather 
than the
civilian IEprep scenario.  At least some folks working at/with NCS do 
not
seem to believe that NCS needs a capability for IEprep to route around 
congestion
as long as there is appropriate system engineering so that congestion 
can't
occur in a given place.

And routing around "access links", which is where the congestion really 
might
live, generally means buying a parallel link for openers.  Because 
unless
there is some other link, no other path exists.  And a colloquial term 
for
"access link" is "tail circuit".   If a second path exists, then that 
path
will be known via the routing system.  In a whole lot of cases (most 
cases
of the deployed Internet), there is no second circuit.  For example, if
the T1 into my office goes down or gets congestion, I'm in trouble 
because
there is no other path here that could be used.

> [MAP] The above sentence about "alternate path routing at the 
> application
> layer" was copied verbatim from the draft (which has your name on it). 
> I was
> simply putting it under a heading that said "Call application layer". 
> What
> did it mean in the original draft?

Out of context, it wasn't clear to me what was meant.  Those particular 
words
(like most of them in either draft) were Ken's, so I'll let Ken explain
(lest I botch the job and muddy the waters more thoroughly than usual 
:-).

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 15:55:01 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA05000
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 15:55:01 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDKvWu11110
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 15:57:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDKvWv11107
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 15:57:32 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04993
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 15:54:29 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDKu3v11056;
	Fri, 13 Dec 2002 15:56:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDKrVv10979
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 15:53:31 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA04940
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 15:50:29 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.138])
	by gnat.inet.org (Postfix) with ESMTP
	id E5E6267105; Fri, 13 Dec 2002 15:53:39 -0500 (EST)
Date: Fri, 13 Dec 2002 15:53:22 -0500
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A6D@mcl-its-exs01.mail.saic.com>
Message-Id: <EA79142D-0EDC-11D7-9A70-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] Re: GETS calls placed from corporate PBXs
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Dec 13, 2002, at 13:55 America/Montreal, King, Kimberly S. 
wrote:
> I think (and Janet or others please correct me if I'm wrong here) that 
> the
> GETS user would get a busy signal, same as any other user if the link 
> to the
> PSTN is filled.  (This is different behavior than a PBX that is 
> enabled to
> support MLPP in DoD.)

Thanks.  I *really* think that we need to differentiate the DoD 
Autovon/MLPP
scenarios from the civilian IEprep scenarios.  There are major 
functional
differences in the deployed systems (GETS vs Autovon).  There are major
differences in needs (GETS reportedly has no pre-emption; Autovon 
absolutely does).
And the solution spaces look different (civilian IEprep is inherently an
inter-domain problem whilst DoD Autovon or IP-layer MLPP is a single 
domain
problem).

>> Does NCS have a special 710
>> contract
>> with most corporate PBX owners ?
>
> This would be news to me.  I don't think so.

So the deployed GETS service is best-effort on the access link from the
corporate handset to the corporate PBX and best-effort from the 
corporate
PBX to the PSTN RBOC Class-5 switch.

>> And what about PBXs that haven't been configured to know about 710 at
>> all ?
>
> GETS operational support folks, if they discover problems (hotels are a
> common examples where PBXs might not understand 710) start a process 
> to get
> the PBX configured correctly.  Sometimes users discover and report 
> problems
> too.  The point is, there is a way to fix that problem.

I'd suspect that lots of corporate PBXs and lots of universities also 
don't
have 710 configured.

Thanks,

Ran
rja@extremenetworks.com


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 16:14:12 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05448
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 16:14:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDLGik12280
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 16:16:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDLGiv12277
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 16:16:44 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05444
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 16:13:41 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDLF2v12221;
	Fri, 13 Dec 2002 16:15:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDLEwv12192
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 16:14:58 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05421
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 16:11:55 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAHK6N>; Fri, 13 Dec 2002 16:12:21 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149F9@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        "King, Kimberly  S."
	 <KIMBERLY.S.KING@saic.com>
Cc: ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: GETS calls placed from corporate PBXs
Date: Fri, 13 Dec 2002 16:12:12 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

If 710 doesn't work from a particular phone, there are some 800 numbers
which will take you straight to a GETS capable IXC.  So as long as 800
works, you are OK.

-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Friday, December 13, 2002 3:53 PM
To: King, Kimberly S.
Cc: ieprep
Subject: [Ieprep] Re: GETS calls placed from corporate PBXs



On Friday, Dec 13, 2002, at 13:55 America/Montreal, King, Kimberly S. 
wrote:
> I think (and Janet or others please correct me if I'm wrong here) that 
> the
> GETS user would get a busy signal, same as any other user if the link 
> to the
> PSTN is filled.  (This is different behavior than a PBX that is 
> enabled to
> support MLPP in DoD.)

Thanks.  I *really* think that we need to differentiate the DoD 
Autovon/MLPP
scenarios from the civilian IEprep scenarios.  There are major 
functional
differences in the deployed systems (GETS vs Autovon).  There are major
differences in needs (GETS reportedly has no pre-emption; Autovon 
absolutely does).
And the solution spaces look different (civilian IEprep is inherently an
inter-domain problem whilst DoD Autovon or IP-layer MLPP is a single 
domain
problem).

>> Does NCS have a special 710
>> contract
>> with most corporate PBX owners ?
>
> This would be news to me.  I don't think so.

So the deployed GETS service is best-effort on the access link from the
corporate handset to the corporate PBX and best-effort from the 
corporate
PBX to the PSTN RBOC Class-5 switch.

>> And what about PBXs that haven't been configured to know about 710 at
>> all ?
>
> GETS operational support folks, if they discover problems (hotels are a
> common examples where PBXs might not understand 710) start a process 
> to get
> the PBX configured correctly.  Sometimes users discover and report 
> problems
> too.  The point is, there is a way to fix that problem.

I'd suspect that lots of corporate PBXs and lots of universities also 
don't
have 710 configured.

Thanks,

Ran
rja@extremenetworks.com


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 16:20:55 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05688
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 16:20:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDLNRf12630
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 16:23:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDLNRv12627
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 16:23:27 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05682
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 16:20:24 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDLM1v12589;
	Fri, 13 Dec 2002 16:22:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDLLtv12570
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 16:21:55 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA05592
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 16:18:51 -0500 (EST)
Received: from extremenetworks.com (unknown [10.0.8.138])
	by gnat.inet.org (Postfix) with ESMTP
	id 4C44767105; Fri, 13 Dec 2002 16:22:03 -0500 (EST)
Date: Fri, 13 Dec 2002 16:21:45 -0500
Subject: Re: [Ieprep] Re: GETS calls placed from corporate PBXs
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>, ieprep <ieprep@ietf.org>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149F9@chntex02.is.dyncorp.com>
Message-Id: <E1BB06DE-0EE0-11D7-9A70-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Dec 13, 2002, at 16:12 America/Montreal, Gunn, Janet wrote:

> If 710 doesn't work from a particular phone, there are some 800 numbers
> which will take you straight to a GETS capable IXC.  So as long as 800
> works, you are OK.

OK, that's useful.  But in such a case, is one operating best-effort
on the inbound links at least as far as the first RBOC switch ?
Is it best effort from the first RBOC switch inward to the destination ?

And is there any pre-emption in GETS ?  That is will a GETS call cause
a non-GETS call that was already placed and is currently in use to get
dropped.  My earlier understanding was "no", but I want to be sure I
understood correctly.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 13 18:55:19 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA10536
	for <ieprep-archive@odin.ietf.org>; Fri, 13 Dec 2002 18:55:19 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBDNvqi22340
	for ieprep-archive@odin.ietf.org; Fri, 13 Dec 2002 18:57:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDNvqv22337
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 13 Dec 2002 18:57:52 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA10506
	for <ieprep-web-archive@ietf.org>; Fri, 13 Dec 2002 18:54:47 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDNuAv22300;
	Fri, 13 Dec 2002 18:56:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBDNtrv22286
	for <ieprep@optimus.ietf.org>; Fri, 13 Dec 2002 18:55:53 -0500
Received: from kc-msxproto2.kc.umkc.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA10411
	for <ieprep@ietf.org>; Fri, 13 Dec 2002 18:52:48 -0500 (EST)
Received: from KC-MAIL2.kc.umkc.edu ([134.193.143.162] RDNS failed) by kc-msxproto2.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.5329);
	 Fri, 13 Dec 2002 17:55:45 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Date: Fri, 13 Dec 2002 17:55:44 -0600
Message-ID: <A29143A40AEAE3459FF829D1DD43316101D9B09D@KC-MAIL2.kc.umkc.edu>
Thread-Topic: [Ieprep] Re: Summary of IEPREP work and VoIP
Thread-Index: AcKitLBDBkF+IB7VTOGQTcXm/MVypgAPV/+w
From: "Ayyasamy, Senthilkumar  (UMKC-Student)" <saq66@umkc.edu>
To: "Nguyen, An" <nguyena@ncs.gov>, "RJ Atkinson" <rja@extremenetworks.com>,
        "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: "ieprep " <ieprep@ietf.org>
X-OriginalArrivalTime: 13 Dec 2002 23:55:45.0215 (UTC) FILETIME=[26DEA4F0:01C2A303]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBDNtrv22287
Subject: [Ieprep] e2e diffserv and ieprep
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


> There are quite a few ISPs who implement Diff-Serv in IP. 

1. Global crossing has deployed diffserv.
The whole Doctoral dissertation of Xipeng Xiao is about 
experience in deploying Diffserv across global crossing.

For people who have time/energy  can look into his PhD
work.
http://www.cse.msu.edu/~xiaoxipe/papers/thesis/thesis.pdf
( It will really be interesting to read his dissertation.
I could understand most of his thesis with less background.)


Else, read three of his papers published in IEEE communication/
networks magazine.)

1."A Practical Approach for Providing QoS in the Internet Backbone"
  with Thomas Telkamp
  IEEE Communications Magazine, Dec. 2002. 
2."Traffic Engineering with MPLS in the Internet" 
  IEEE Network Magazine, Mar. 2000.
3."Internet QoS: A Big Picture"
  IEEE Network Magazine, March 1999.

For people who dont have access to ieeexplorer, you can grab a copy 
from http://www.cse.msu.edu/~xiaoxipe/.


2. Operational view point:

For a more operation point of view, look for thomas telkamp's
presentation along with Dave/Randy/Lixia in oregon nanog.
http://www.nanog.org/mtg-0210/complexity.html

If you want to hear more on what thomas is saying:
http://www.nanog.org/mtg-0210/telkamp.html


3. If you have started liking QoS, look for Ash
submission to tewg.
http://www.research.att.com/~jrex/jerry/

Although, their is a lot of literature regarding QoS routing,
i think reading this draft sums up everything.

---------------------------------------------------
If your don't like QoS concept but still want to use 
operational heuristics/theoretical techniques to 
  - optimize your routing strategies (traffic aware routing etc.)
  - Do traffic engineering without MPLS
  - efficiently plan your capacity.
  - How to give high availability for demanding traffic.
  - know how much to provision your backbone
  - find the rogue (high bw flows, non-responsive flows, DoS aggregate
    traffic.)
see the works from
   http://ipmon.sprintlabs.com/
   http://www.research.att.com/~jrex/

esp. sprintlabs has works claiming that their backbone has 
less (burstiness, packet reordering). Also, about how 
they help high  availability for VoIP traffic without QoS.
If your a fan of applying theoretical principles like
teletraffic models,ARIMA modeling, Queuing theory,
and network calculus to practical network problems, sprintlabs
work has a lot to offer.

One caveat: Both sprintlabs/ATT have a very good measurement 
system for getting up to date details about Backbone. They
have passive measurement monitoring systems which samples
packets and gives them details about their backbone. Based on
those details, they can do all the above applications.
 But, lets take an ISP which doesn't have such a measurement
system(?) and do brute force over provisioning without 
effective planning. I don't know but in some cases, it might 
backfire(?). In transportation networks, their is a concept 
called Braess paradox. Braess' paradox is said 
to occur in a network if the addition of an extra link leads
to a worse performance. *Theoretically*, Frank Kelly proved its 
existence in loss networks. 
http://citeseer.nj.nec.com/bean95braes.html

But, sure ...it will not exist in practical network scenarios.
But, my point is, their might be some cases where unplanned 
provisioning can backfire. Might be operators help me understand
whether they occur or not...


Some related interesting Internet drafts:
http://www.ietf.org/internet-drafts/draft-liljenstolpe-tewg-cwbcp-02.txt
http://www.ietf.org/internet-drafts/draft-bhattacharyya-monitoring-sprint-00.txt
ftp://ftp.sprintlabs.com/diot/mpls-white-paper.txt


> In short, my point is we don't want to rule out anything at this EARLY
> stage.

Exactly.. the direction/tone of the discussion forced me to think that ieprep is 
ruling out many technologies standardized by IETF. For example, if end to
end diffserv is not standardized by IETF, it doesn't mean that ieprep cant 
work anything involving DIffserv. Operators cant still do end to end diffserv 
by using RFC 2998 or do static provisioning across the domains. 

The only point that stuck me for no enough input for continuing e2e Diffserv 
in IETF is, 
  - lack of wide spread diffserv deployment/Business models 
    for inter-domain. (but this should not be a criteria as IETF assumes
    all inter-domain contracting issues in place.)
  - Their is no enough control plane work. All that is standardized by 
    Diffserv WG is data plane work. Still, it needs a signaling mechanism to
    do control plane work (so, we have to wait for NSIS output
    http://www.ietf.org/internet-drafts/draft-salsano-bgrpp-arch-00.txt) 
    So, once we have inter-domain signaling in place, e2e Diffserv work can be 
    continued.

Other than this, we have seen many initiatives regarding e2e Diffserv/deployment.
For example, i2 Bandwidth broker architecture, BuDs RG in IRTF( It is now 
defunct..I don't know about their deliverables but it is about diffserv deployment)
and Diffserv deployment BOF( http://www.cs.ucl.ac.uk/research/decides/) 
IEPG meeting in Atlanta IETF had a presentation about Diffserv deployment
although in a research network (http://www.potaroo.net/iepg/november2002/geant.pdf)

I was forced to write this as the recent IAB minutes hints about ieprep WG and
its relation to e2e diffserv.

I still stand by my previous comments  
https://www1.ietf.org/mail-archive/working-groups/ieprep/current/msg00828.html
about many problems with QoS deployment.
But, the whole point is that, ieprep cannot simply reject a technology standardized
by IETF since most of the operators currently dont use them. We have to unravel
all the stones before coming to the conclusion that this particular extension works 
best for ieprep.









_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 14 07:04:12 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01239
	for <ieprep-archive@odin.ietf.org>; Sat, 14 Dec 2002 07:04:12 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBEC6eh31106
	for ieprep-archive@odin.ietf.org; Sat, 14 Dec 2002 07:06:40 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBEC6ev31103
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 14 Dec 2002 07:06:40 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01234
	for <ieprep-web-archive@ietf.org>; Sat, 14 Dec 2002 07:03:41 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBEC5Dv31087;
	Sat, 14 Dec 2002 07:05:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBEC4qv31060
	for <ieprep@optimus.ietf.org>; Sat, 14 Dec 2002 07:04:52 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01150
	for <ieprep@ietf.org>; Sat, 14 Dec 2002 07:01:53 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Sat, 14 Dec 2002 07:04:14 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121407043600351
 ; Sat, 14 Dec 2002 07:04:36 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YW14F79X>; Sat, 14 Dec 2002 07:04:23 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A70@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson '" <rja@extremenetworks.com>,
        "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>
Cc: "'ieprep '" <ieprep@ietf.org>
Date: Sat, 14 Dec 2002 07:04:36 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] Autovon/MLPP scenarios
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Ran said:
>Thanks.  I *really* think that we need to differentiate the DoD 
>Autovon/MLPP scenarios from the civilian IEprep scenarios.  
>There are major functional differences in the deployed systems 
>(GETS vs Autovon).  There are major differences in needs (GETS 
>reportedly has no pre-emption; Autovon absolutely does).
>And the solution spaces look different (civilian IEprep is inherently 
>an inter-domain problem whilst DoD Autovon or IP-layer MLPP is a 
>single domain problem).

While basically single domain, it isn't exactly as simple "the government
(DoD) is both "A" and "A"'s ISP'.  The reality is A (post, camps and
stations) have very different policies than A's ISP (DISA).  In this case (a
problem for DoD but not ieprep), is that DoD may have to agree with itself
to set the policy on those access links!

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 14 07:13:53 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01358
	for <ieprep-archive@odin.ietf.org>; Sat, 14 Dec 2002 07:13:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBECGMj31881
	for ieprep-archive@odin.ietf.org; Sat, 14 Dec 2002 07:16:22 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBECGMv31878
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 14 Dec 2002 07:16:22 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01350
	for <ieprep-web-archive@ietf.org>; Sat, 14 Dec 2002 07:13:22 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBECF4v31865;
	Sat, 14 Dec 2002 07:15:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBECEbv31841
	for <ieprep@optimus.ietf.org>; Sat, 14 Dec 2002 07:14:37 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01341
	for <ieprep@ietf.org>; Sat, 14 Dec 2002 07:11:37 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Sat, 14 Dec 2002 07:14:03 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121407142602531
 ; Sat, 14 Dec 2002 07:14:26 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YW14F8DX>; Sat, 14 Dec 2002 07:14:12 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A71@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson '" <rja@extremenetworks.com>,
        "'Gunn, Janet '" <Janet.Gunn@dyncorp.com>
Cc: "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>,
        "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: GETS calls placed from corporate PBXs
Date: Sat, 14 Dec 2002 07:14:25 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Ran asked:
>But in such a case, is one operating best-effort
>on the inbound links at least as far as the first RBOC switch ?

Yes, if using the 800 numbers to access GETS, there is no preferential
treatment at the LEC.  

>Is it best effort from the first RBOC switch inward to the destination ?

No, best effort from the first RBOC switch but once at one of the IXCs (the
800 numbers are tied to GETS supporting IXCs), there is preferential
treatment for the remainder of the call.

>And is there any pre-emption in GETS ?  That is will a GETS call cause
>a non-GETS call that was already placed and is currently in use to get
>dropped.  My earlier understanding was "no", but I want to be sure I
>understood correctly.

You are correct.  No pre-emption is done and I have heard it said many times
that it isn't needed either.  This is because the PSTN has a huge amount of
resources.  Simply by waiting a bit (some extra queuing for a line to become
available), one can usually succeed because capacity will almost always
become available during a short time interval. 

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 14 07:21:51 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01450
	for <ieprep-archive@odin.ietf.org>; Sat, 14 Dec 2002 07:21:51 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBECOKr32041
	for ieprep-archive@odin.ietf.org; Sat, 14 Dec 2002 07:24:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBECOKv32038
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 14 Dec 2002 07:24:20 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01441
	for <ieprep-web-archive@ietf.org>; Sat, 14 Dec 2002 07:21:20 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBECN1v32013;
	Sat, 14 Dec 2002 07:23:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBECMev31991
	for <ieprep@optimus.ietf.org>; Sat, 14 Dec 2002 07:22:40 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA01433
	for <ieprep@ietf.org>; Sat, 14 Dec 2002 07:19:37 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBECGIo13006;
	Sat, 14 Dec 2002 04:16:18 -0800 (PST)
Date: Sat, 14 Dec 2002 04:16:18 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
cc: "'RJ Atkinson '" <rja@extremenetworks.com>,
        "'Gunn, Janet '" <Janet.Gunn@dyncorp.com>,
        "'ieprep '" <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: GETS calls placed from corporate PBXs
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A71@mcl-its-exs01.mail.saic.com>
Message-ID: <Pine.GSO.4.44.0212140413000.12564-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Sat, 14 Dec 2002, King, Kimberly  S. wrote:
    > No pre-emption is done and I have heard it said many times
    > that it isn't needed either.  This is because the PSTN has a huge
    > amount of resources.  Simply by waiting a bit (some extra queuing
    > for a line to become available), one can usually succeed because
    > capacity will almost always become available during a short time
    > interval.

Heh.  So here we see where the two systems are alike, at a sort of quantum
level.  In the packet switched network, there's a single selected path
between any two points, and it's either in use, or not in use, and the
state changes on a microsecond basis.

Leading, of course, to the same conclusion.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 14 09:40:40 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA02919
	for <ieprep-archive@odin.ietf.org>; Sat, 14 Dec 2002 09:40:40 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBEEh9G04911
	for ieprep-archive@odin.ietf.org; Sat, 14 Dec 2002 09:43:09 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBEEh9v04908
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 14 Dec 2002 09:43:09 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA02912
	for <ieprep-web-archive@ietf.org>; Sat, 14 Dec 2002 09:40:09 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBEEfNv04864;
	Sat, 14 Dec 2002 09:41:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBEEeQv04832
	for <ieprep@optimus.ietf.org>; Sat, 14 Dec 2002 09:40:26 -0500
Received: from bells.cs.ucl.ac.uk (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id JAA02792
	for <ieprep@ietf.org>; Sat, 14 Dec 2002 09:37:26 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.25215-0@bells.cs.ucl.ac.uk>; Sat, 14 Dec 2002 14:40:20 +0000
To: ieprep@ietf.org
Subject: Re: [Ieprep] updated requirements documents (telephony)
Date: Sat, 14 Dec 2002 14:40:19 +0000
Message-ID: <22344.1039876819@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


> > [MAP] The above sentence about "alternate path routing at the 
> > application
> > layer" was copied verbatim from the draft (which has your name on it). 
> > I was
> > simply putting it under a heading that said "Call application layer". 
> > What
> > did it mean in the original draft?
>
> Out of context, it wasn't clear to me what was meant.  Those particular
> words (like most of them in either draft) were Ken's, so I'll let Ken
> explain (lest I botch the job and muddy the waters more thoroughly than
> usual :-).

I'm sorry, but its hard to dig out what the actual question is.  if
there is still an open question, could it be rephrased?  otherwise, I'll
assume that the thread is closed and stick to my previous comments.

regards,

-ken

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 14 11:42:21 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04709
	for <ieprep-archive@odin.ietf.org>; Sat, 14 Dec 2002 11:42:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBEGiqx10210
	for ieprep-archive@odin.ietf.org; Sat, 14 Dec 2002 11:44:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBEGiqv10207
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 14 Dec 2002 11:44:52 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04705
	for <ieprep-web-archive@ietf.org>; Sat, 14 Dec 2002 11:41:50 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBEGhOv10155;
	Sat, 14 Dec 2002 11:43:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBEGglv10113
	for <ieprep@optimus.ietf.org>; Sat, 14 Dec 2002 11:42:47 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA04665
	for <ieprep@ietf.org>; Sat, 14 Dec 2002 11:39:45 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Sat, 14 Dec 2002 11:42:11 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121411423301586
 for <ieprep@ietf.org>; Sat, 14 Dec 2002 11:42:33 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YKPH8LTV>; Sat, 14 Dec 2002 11:44:08 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A77@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
Date: Sat, 14 Dec 2002 11:42:33 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] Understanding Congestion
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

A good first step in modifying a system is a firm grasp of the problem to be
solved.  If the problem is congestion, then here is a start
http://www.research.att.com/~smb/papers/tprc.pdf

Kimberly

"The PTN is better able to control congestion than the Internet is. When a
phone switch or telephone transmission facility reaches saturation, new
callers receive "reorder" (i.e., "fast" busy) signals and no further calls
are accepted. This forestalls increased load and congestion. PTN operations
staff can even block call attempts to a given destination at sources,
thereby saving network resources from being wasted on calls that are
unlikely to be completed. For example, when an earthquake occurs near San
Francisco, the operations staff might decide to block almost all incoming
calls to the affected area codes from throughout the entire PTN.

Congestion management in the Internet is problematic, in part, because no
capabilities exist for managing traffic associated with specific users,
connections, sources, or destinations, and it would be difficult to
implement such capabilities. All that an Internet router can do is to
discard packets when its buffers become full. To implement fair allocation
of resources and bandwidth, routers would have to store information about
users and connections, something they are not built to do. Retaining such
information would require large amounts of storage.

Managing this storage would be difficult, because Internet routers do not
normally process call tear-down messages. Furthermore, the concept of a
"user"-that is, an entity that originates or receives traffic-is not part of
the network or transport layers of the Internet protocols. Chokingback load
offered by specific hosts (in analogy with PTN "reorder" signals) is also
not an option for preventing Internet congestion, since an Internet protocol
(IP)-capable host can concurrently have connections open to many
destinations. Stopping all flows from the host is clearly inappropriate.
Highly dynamic traffic flows between ISPs are particularly problematic.
Here, very high speed (e.g., OC-12) circuits are used to carry traffic
between millions of destinations over short intervals and the traffic mix
can completely change over a few seconds.

Although congestion in the Internet is nominally an IP-layer phenomena
routers have too many packets for a given link measures for dealing
successfully with congestion have been deployed in the transmission control
protocol (TCP) layer.  Some newer algorithms work at the IP level, but there
is an inadequate knowledge base, especially for defining and enforcing
flexible and varied procedures for congestion control. One suggestion
involves retaining information about flows from which packets have been
repeatedly dropped. Such flows are deemed uncooperative and, as such, are
subjected to additional penalties; cooperating flows respond to indications
of congestion by slowing down their transmissions.

Some work has been done, but very little is known about usage patterns, flow
characteristics, and other relevant parameters of current Internet traffic,
much less how these patterns may evolve in the future. Having such
information is likely to enable better congestion control methods. However,
usage patterns are dictated by the application designs, and as new
applications arise and become popular, traffic characteristics change.
Today, the use of the Web has radically changed packet sizes from a time
when file transfer and e-mail were the principal applications. Even within
the Web environment, when a very popular Web site arises, news of its
location spreads quickly and traffic flows shift noticeably.

There are two further difficulties associated with managing congestion in
networks.

First, there appears to be a tension between implementing congestion
management and enforcing network security. A congestion control mechanism
may need to inspect and even modify traffic being managed, but strong
network security mechanisms will prohibit reading and modifying traffic en
route. For example, congestion control in the Internet might be improved if
IP and TCP headers were inspected and modified but the use of IPsec will
prevent such actions.

A second difficulty arises when a network comprises multiple, independent
but interconnected providers. In the Internet, no single party is either
capable of or responsible for most end-to-end connections, and local
optimizations performed by individual providers may lead to poor overall
utilization of network resources or sub-optimal global behavior. In the PTN,
which was designed for a world with comparatively few telephone companies
but in which switches can be trusted, competitive pressures are now forcing
telephone companies to permit widespread interconnections between switches
that may not be trustworthy. This opens telephone networks to both malicious
and non-malicious failures."


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 14 18:16:13 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA13281
	for <ieprep-archive@odin.ietf.org>; Sat, 14 Dec 2002 18:16:13 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBENIlj27813
	for ieprep-archive@odin.ietf.org; Sat, 14 Dec 2002 18:18:47 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBENIlv27810
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 14 Dec 2002 18:18:47 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA13274
	for <ieprep-web-archive@ietf.org>; Sat, 14 Dec 2002 18:15:41 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBENH7v27761;
	Sat, 14 Dec 2002 18:17:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBEJ7dv16294
	for <ieprep@optimus.ietf.org>; Sat, 14 Dec 2002 14:07:39 -0500
Received: from joy.songbird.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA07535
	for <ieprep@ietf.org>; Sat, 14 Dec 2002 14:04:36 -0500 (EST)
Received: from bbprime.brandenburg.com (208.184.79.252.songbird.com [208.184.79.252] (may be forged))
	by joy.songbird.com (8.9.3/8.9.3) with ESMTP id LAA24344;
	Sat, 14 Dec 2002 11:07:17 -0800
Date: Sat, 14 Dec 2002 07:57:42 -0800
From: Dave Crocker <dhc@dcrocker.net>
X-Mailer: The Bat! (v1.62 Christmas Edition) Personal
Organization: Brandenburg InternetWorking
X-Priority: 3 (Normal)
Message-ID: <8571788666.20021214075742@brandenburg.com>
To: Marshall Eubanks <tme@multicasttech.com>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] RE: PSTN collapse
In-Reply-To: <A1E990FA-0C4F-11D7-9B57-003065BA697E@multicasttech.com>
References: <A1E990FA-0C4F-11D7-9B57-003065BA697E@multicasttech.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Marshall,

Tuesday, December 10, 2002, 6:56:58 AM, you wrote:
Marshall> When I lived in California we went through two significant (not major) 
Marshall> earthquakes.
Marshall> Both times we could not complete an outbound call for some hours - but 
Marshall> once
Marshall> we got through, the connection was fine.

Folks,

Marshall> That sounds like a PSTN collapse to me.

I once did a presentation to some telephone company folks and cited the
telephone situation after the 1989 quake, referring to it a the quake's
breaking the phone systems.

They got very upset with me.

In an emergency, the phones are automatically shut down for regular use, to
ensure that there is capacity for the emergency calls.

Such an actions is very much NOT "collapse".

d/
-- 
 Dave <mailto:dhc@dcrocker.net>
 Brandenburg InternetWorking <http://www.brandenburg.com>
 t +1.408.246.8253; f +1.408.850.1850

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 14 18:16:13 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA13283
	for <ieprep-archive@odin.ietf.org>; Sat, 14 Dec 2002 18:16:13 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBENIlV27828
	for ieprep-archive@odin.ietf.org; Sat, 14 Dec 2002 18:18:47 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBENIlv27821
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 14 Dec 2002 18:18:47 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA13273
	for <ieprep-web-archive@ietf.org>; Sat, 14 Dec 2002 18:15:42 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBENHAv27777;
	Sat, 14 Dec 2002 18:17:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBEJ7jv16298
	for <ieprep@optimus.ietf.org>; Sat, 14 Dec 2002 14:07:45 -0500
Received: from joy.songbird.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA07538
	for <ieprep@ietf.org>; Sat, 14 Dec 2002 14:04:42 -0500 (EST)
Received: from bbprime.brandenburg.com (208.184.79.252.songbird.com [208.184.79.252] (may be forged))
	by joy.songbird.com (8.9.3/8.9.3) with ESMTP id LAA24347;
	Sat, 14 Dec 2002 11:07:18 -0800
Date: Sat, 14 Dec 2002 08:08:18 -0800
From: Dave Crocker <dhc@dcrocker.net>
X-Mailer: The Bat! (v1.62 Christmas Edition) Personal
Organization: Brandenburg InternetWorking
X-Priority: 3 (Normal)
Message-ID: <5272424651.20021214080818@brandenburg.com>
To: Fred Baker <fred@cisco.com>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] RE: PSTN collapse
In-Reply-To: <5.2.0.9.2.20021209152513.03467b20@mira-sjcm-4.cisco.com>
References: <5.2.0.9.2.20021209152513.03467b20@mira-sjcm-4.cisco.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Fred,

Monday, December 9, 2002, 3:34:30 PM, you wrote:
Fred> At 04:57 PM 12/9/2002 -0600, Ayyasamy, Senthilkumar  (UMKC-Student) wrote:
>>I am aware of only RFC 896, which talks about congestive collapse
>>and how it leads to "Internet meltdown".

Fred> You should also take a look at RFC 970, which describes the mechanism for 
Fred> congestive collapse. This did in fact occur in late 1987 and most of 1988. 
Fred> The issue was that most links in the 56 KBPS NSFNET were running at 
Fred> capacity. Your queuing theory courses should have told you that 100% 
Fred> utilization implies infinite delay,

After a number of years of doing Arpanet measurements, Leonard Kleinrock
includes a slide in presentations about the net, showing an interesting
curve.  Y was delay and X was throughput.  The curve was almost horizontal
for quite awhile, then had a rather sharp knee as then was highly vertical.

Kleinrock noted that this summarized the performance of Arpanet packet
switching. He's an excellent teacher, able to make queuing theory actually
sound interesting. But he also knows when to keep things simple

He summarized the curve as: throughput is very very good, until it gets very
very bad.

In effect, he said, all of the queuing mechanisms in a switch are useful
only in the very short range of the knee. When the system is under sustained
load that is at capacity, queuing does not help.

Why bother with all that mechanisms, he then asked?

Transients.  Queuing mechanisms are for *transient* high loads.

d/
-- 
 Dave <mailto:dhc@dcrocker.net>
 Brandenburg InternetWorking <http://www.brandenburg.com>
 t +1.408.246.8253; f +1.408.850.1850

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 14 21:34:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA16250
	for <ieprep-archive@odin.ietf.org>; Sat, 14 Dec 2002 21:34:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBF2b5L03084
	for ieprep-archive@odin.ietf.org; Sat, 14 Dec 2002 21:37:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBF2b5v03073
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 14 Dec 2002 21:37:05 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA16181
	for <ieprep-web-archive@ietf.org>; Sat, 14 Dec 2002 21:34:00 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBF2ZEv02832;
	Sat, 14 Dec 2002 21:35:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBF2YAv02797
	for <ieprep@optimus.ietf.org>; Sat, 14 Dec 2002 21:34:10 -0500
Received: from pfwhqs1.ncr.disa.mil (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id VAA15930
	for <ieprep@ietf.org>; Sat, 14 Dec 2002 21:31:00 -0500 (EST)
Received: from mtahqs3.ncr.disa.mil by pfwhqs1.ncr.disa.mil
          via smtpd (for ietf-mx.ietf.org [132.151.6.1]) with SMTP; 15 Dec 2002 02:30:18 UT
Received: by mtahqs3.ncr.disa.mil with Internet Mail Service (5.5.2653.19)
	id <YXSQV9AW>; Sat, 14 Dec 2002 21:35:38 -0500
Message-ID: <7F18415E4D63CB45BB9B3A591F68D12D5B3DF4@emshqs1.ncr.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>,
        "''RJ Atkinson ' '"
	 <rja@extremenetworks.com>,
        "''Gunn, Janet ' '" <Janet.Gunn@dyncorp.com>
Cc: "''ieprep ' '" <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: GETS calls placed from corporate PBXs
Date: Sat, 14 Dec 2002 21:33:53 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Ran,

-----Original Message-----
>From: King, Kimberly  S.
>To: 'RJ Atkinson '; 'Gunn, Janet '
>Cc: 'King, Kimberly  S. '; 'ieprep '
>Sent: 12/14/02 7:14 AM
>Subject: RE: [Ieprep] Re: GETS calls placed from corporate PBXs

>Ran asked:
>>But in such a case, is one operating best-effort
>>on the inbound links at least as far as the first RBOC switch ?

>Yes, if using the 800 numbers to access GETS, there is no preferential
>treatment at the LEC.  

>>Is it best effort from the first RBOC switch inward to the destination
?

>No, best effort from the first RBOC switch but once at one of the IXCs
>(the
>800 numbers are tied to GETS supporting IXCs), there is preferential
>treatment for the remainder of the call.

Two of several GETS features are "trunking queuing" and exemption from
management control" that give GETS high probability call completion during
congestion.

1. Trunk Queuing allows a GETS user to queue his/her calls and wait for the
next available; and

2. With exemption, GETS calls don't get blocked. 

With these two designed to support GETS, I think GETS gets better service
than just "best effort".

An
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 09:42:39 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA06396
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 09:42:38 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGEjAQ25132
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 09:45:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGEjAv25129
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 09:45:10 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA06388
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 09:42:07 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGEhFv25075;
	Mon, 16 Dec 2002 09:43:16 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGEgav25049
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 09:42:36 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA06352
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 09:39:33 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAHMWG>; Mon, 16 Dec 2002 09:40:00 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A4149FE@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'Dave Crocker'" <dhc@dcrocker.net>, Fred Baker <fred@cisco.com>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] RE: PSTN collapse
Date: Mon, 16 Dec 2002 09:39:48 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

That is a standard queueing theory curve- nothing specific to the Internet.

And the interesting thing is that, by changing the scale, you can make the
"knee of the curve" appear (to the human eye) to be in different places.
Therefore you should always be suspicious of quantitative arguments based on
the "location" the knee of the curve.

But the qualitative point is the same.

Queueing with infinite buffers, or where "arrivals" are dropped
indiscriminately when the buffer is full, is only useful for handling
transients.

However, "discriminatory" queueing, where only "preferred" arrivals are
allowed to queue, or  where "preferred" arrivals can "bump" "non-preferred"
arrivals when the buffer is full, make a difference for the preferred
arrivals (as long as they are only a small percentage of the total) even in
sustained congestion.

-----Original Message-----
From: Dave Crocker [mailto:dhc@dcrocker.net]
Sent: Saturday, December 14, 2002 11:08 AM
To: Fred Baker
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] RE: PSTN collapse


Fred,

Monday, December 9, 2002, 3:34:30 PM, you wrote:
Fred> At 04:57 PM 12/9/2002 -0600, Ayyasamy, Senthilkumar  (UMKC-Student)
wrote:
>>I am aware of only RFC 896, which talks about congestive collapse
>>and how it leads to "Internet meltdown".

Fred> You should also take a look at RFC 970, which describes the mechanism
for 
Fred> congestive collapse. This did in fact occur in late 1987 and most of
1988. 
Fred> The issue was that most links in the 56 KBPS NSFNET were running at 
Fred> capacity. Your queuing theory courses should have told you that 100% 
Fred> utilization implies infinite delay,

After a number of years of doing Arpanet measurements, Leonard Kleinrock
includes a slide in presentations about the net, showing an interesting
curve.  Y was delay and X was throughput.  The curve was almost horizontal
for quite awhile, then had a rather sharp knee as then was highly vertical.

Kleinrock noted that this summarized the performance of Arpanet packet
switching. He's an excellent teacher, able to make queuing theory actually
sound interesting. But he also knows when to keep things simple

He summarized the curve as: throughput is very very good, until it gets very
very bad.

In effect, he said, all of the queuing mechanisms in a switch are useful
only in the very short range of the knee. When the system is under sustained
load that is at capacity, queuing does not help.

Why bother with all that mechanisms, he then asked?

Transients.  Queuing mechanisms are for *transient* high loads.

d/
-- 
 Dave <mailto:dhc@dcrocker.net>
 Brandenburg InternetWorking <http://www.brandenburg.com>
 t +1.408.246.8253; f +1.408.850.1850

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 10:01:06 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA07227
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 10:01:06 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGF3cX25918
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 10:03:38 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGF3bv25915
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 10:03:37 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA07213
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 10:00:34 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGF22v25845;
	Mon, 16 Dec 2002 10:02:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGF1Qv25828
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 10:01:26 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07139
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 09:58:23 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 16 Dec 2002 10:00:14 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121610003516983
 ; Mon, 16 Dec 2002 10:00:35 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YW14H6JT>; Mon, 16 Dec 2002 10:00:21 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A84@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Gunn, Janet'" <Janet.Gunn@DynCorp.com>,
        "'Dave Crocker'" <dhc@dcrocker.net>, Fred Baker <fred@cisco.com>
Cc: ieprep@ietf.org
Subject: RE: [Ieprep] RE: PSTN collapse
Date: Mon, 16 Dec 2002 10:00:33 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

All,

As a mathematician, I *love* queuing theory and if interested, Kleinrock's
Queuing Systems is a great place to start.  

But...we might be getting in the weeds here.  I'm still in search of a clear
concise problem definition.

Kimberly

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 12:24:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA11449
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 12:24:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGHQgf02450
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 12:26:42 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGHQgv02447
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 12:26:42 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA11418
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 12:23:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGHPHv02402;
	Mon, 16 Dec 2002 12:25:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGHO2v02363
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 12:24:02 -0500
Received: from joy.songbird.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA11391
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 12:20:59 -0500 (EST)
Received: from bbprime.brandenburg.com (208.184.79.252.songbird.com [208.184.79.252] (may be forged))
	by joy.songbird.com (8.9.3/8.9.3) with ESMTP id JAA32586
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 09:23:41 -0800
Date: Mon, 16 Dec 2002 09:23:44 -0800
From: Dave Crocker <dhc2@dcrocker.net>
X-Mailer: The Bat! (v1.62 Christmas Edition) Personal
Organization: Brandenburg InternetWorking
X-Priority: 3 (Normal)
Message-ID: <18321262523.20021216092344@brandenburg.com>
To: ieprep@ietf.org
Subject: Re: [Ieprep] RE: PSTN collapse
In-Reply-To: <CBED705A7FD2D311865500508B1089410A4149FE@chntex02.is.dyncorp.com>
References: <CBED705A7FD2D311865500508B1089410A4149FE@chntex02.is.dyncorp.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Folks,

Monday, December 16, 2002, 6:39:48 AM, you wrote:
Janet> However, "discriminatory" queueing, where only "preferred" arrivals are
Janet> allowed to queue, or  where "preferred" arrivals can "bump" "non-preferred"
Janet> arrivals when the buffer is full, make a difference for the preferred
Janet> arrivals (as long as they are only a small percentage of the total) even in
Janet> sustained congestion.

Unfortunately, this category of mechanism is at key line of attack for QOS,
which is notably lacking from the Internet. So discriminatory queuing might
sound great, but has nothing to do with the current, global Internet.

And the reason I am responding, after Kimberly's nudge to drop this thread,
is to encourage people to note that the IEPREP effort is not going to be
useful unless and until discussion pays attention to technology realities
and operations realities. The latter are the more challenging.

After two years, it is clear that a focus on making infrastructure changes,
to support IEPREP, is not going to be productive.

Discussion needs to look for discrete, mechanisms that can "contribute" to
IEPREP utility through incremental adoption, rather than looking for major
changes that solve large parts of the problem space.

d/
-- 
 Dave <mailto:dhc2@dcrocker.net>
 Brandenburg InternetWorking <http://www.brandenburg.com>
 t +1.408.246.8253; f +1.408.850.1850

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 13:55:52 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14158
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 13:55:52 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGIwPP07330
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 13:58:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGIwPv07327
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 13:58:25 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14143
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 13:55:21 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGIuHv07237;
	Mon, 16 Dec 2002 13:56:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGItQv07191
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 13:55:26 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14046
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 13:52:21 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id DDD9367103; Mon, 16 Dec 2002 13:56:08 -0500 (EST)
Date: Mon, 16 Dec 2002 13:55:20 -0500
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep '" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A70@mcl-its-exs01.mail.saic.com>
Message-Id: <ECEDCBF7-1127-11D7-91A2-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] Re: Autovon/MLPP scenarios
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Saturday, Dec 14, 2002, at 07:04 America/Montreal, King, Kimberly S. 
wrote:
> While basically single domain, it isn't exactly as simple "the 
> government
> (DoD) is both "A" and "A"'s ISP'.  The reality is A (post, camps and
> stations) have very different policies than A's ISP (DISA).  In this 
> case (a
> problem for DoD but not ieprep), is that DoD may have to agree with 
> itself
> to set the policy on those access links!

Organisational dynamics can be quite complex, but are not within IETF's
purview.  On paper at least, someone at a high-enough-level within DoD
could mandate a single policy for all DoD IP networks.  Some DISA folks
would say that DISA already has that authority.  Others might point to
JCS as the source of such authority.  At least its pretty clear that if
the Secretary of Defence issued an instruction, that would be binding.

I don't think the IETF ought to spend any cycles worrying about whether
DoD chooses to have a single policy or not.  From an external view,
which is all the IETF can really have, the DoD AutoVon over IP concept
would be a single-domain concept, not inter-domain.

By contrast, the civilian IEprep scenario is no kidding really really
inter-domain, without any body in a position to mandate a common policy
for ISP A and ISP B.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 13:59:55 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14297
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 13:59:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGJ2Sp07496
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 14:02:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJ2Sv07493
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 14:02:28 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14226
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 13:59:23 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJ12v07457;
	Mon, 16 Dec 2002 14:01:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJ05v07402
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 14:00:05 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14193
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 13:57:01 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 16 Dec 2002 13:59:25 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121613594915743
 for <ieprep@ietf.org>; Mon, 16 Dec 2002 13:59:49 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YW142Z70>; Mon, 16 Dec 2002 13:59:35 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77A9A@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "Ieprep (E-mail)" <ieprep@ietf.org>
Date: Mon, 16 Dec 2002 13:59:44 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] Core competency
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

ISP backbones stand up in grueling 30-day performance test.

 http://www.nwfusion.com/research/2002/1216isptest.html?net

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 14:00:32 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA14376
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 14:00:32 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGJ36S07545
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 14:03:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJ36v07542
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 14:03:06 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA14343
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 14:00:01 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJ13v07473;
	Mon, 16 Dec 2002 14:01:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJ0ov07427
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 14:00:50 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14206
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 13:57:46 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP id AD1EB67103
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 14:01:33 -0500 (EST)
Date: Mon, 16 Dec 2002 14:00:44 -0500
Subject: Re: [Ieprep] Understanding Congestion
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
From: RJ Atkinson <rja@extremenetworks.com>
To: ieprep <ieprep@ietf.org>
Content-Transfer-Encoding: 7bit
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A77@mcl-its-exs01.mail.saic.com>
Message-Id: <AE0F1253-1128-11D7-91A2-00039357A82A@extremenetworks.com>
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


A key item to understand is that the PSTN does not really have the
concept of congestion avoidance (e.g. slow down the bits/second rate
of a session to help the network avoid being congested), whereas that
is central to the Internet.

This is not a better/worse thing, just as call rejection capability
isn't a better/worse thing.  The architectures are fundamentally
different.  One needs to work with the architecture for the system
in hand, not some other architecture.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 14:02:55 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA14509
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 14:02:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGJ5Sn07638
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 14:05:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJ5Sv07635
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 14:05:28 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA14468
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 14:02:23 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJ43v07591;
	Mon, 16 Dec 2002 14:04:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJ33v07538
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 14:03:03 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA14326
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 13:59:59 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 4843467103; Mon, 16 Dec 2002 14:03:46 -0500 (EST)
Date: Mon, 16 Dec 2002 14:02:57 -0500
Subject: Re: [Ieprep] RE: PSTN collapse
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: Marshall Eubanks <tme@multicasttech.com>, ieprep@ietf.org
To: Dave Crocker <dhc@dcrocker.net>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <8571788666.20021214075742@brandenburg.com>
Message-Id: <FCFF982A-1128-11D7-91A2-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Saturday, Dec 14, 2002, at 10:57 America/Montreal, Dave Crocker 
wrote:
> I once did a presentation to some telephone company folks and cited the
> telephone situation after the 1989 quake, referring to it a the quake's
> breaking the phone systems.
>
> They got very upset with me.
>
> In an emergency, the phones are automatically shut down for regular 
> use, to
> ensure that there is capacity for the emergency calls.
>
> Such an actions is very much NOT "collapse".

Such shutdowns are normally performed on inbound calls from outside
the disaster area, in order to leave more capacity for outbound
calls originating within the disaster area.  So loss of service to
phones in the disaster area that were still connected by good copper 
wires
would constitute a reasonable count for measuring congestion collapse.

Of course, I don't know what was in your presentation, so can't say
whether I'd agree with the phone company assertions or not.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 14:07:56 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA14654
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 14:07:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGJAUH08407
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 14:10:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJATv08404
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 14:10:29 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA14618
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 14:07:25 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJ92v08340;
	Mon, 16 Dec 2002 14:09:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGJ8iv08311
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 14:08:44 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA14564
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 14:05:39 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id C28B367103; Mon, 16 Dec 2002 14:09:26 -0500 (EST)
Date: Mon, 16 Dec 2002 14:08:37 -0500
Subject: Re: [Ieprep] Re: GETS calls placed from corporate PBXs
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>,
        "''Gunn, Janet ' '" <Janet.Gunn@dyncorp.com>,
        "''ieprep ' '" <ieprep@ietf.org>
To: "Nguyen, An" <nguyena@ncs.gov>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <7F18415E4D63CB45BB9B3A591F68D12D5B3DF4@emshqs1.ncr.disa.mil>
Message-Id: <C7BB9E86-1129-11D7-91A2-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Saturday, Dec 14, 2002, at 21:33 America/Montreal, Nguyen, An wrote:
> Two of several GETS features are "trunking queuing" and exemption from
> management control" that give GETS high probability call completion 
> during
> congestion.
>
> 1. Trunk Queuing allows a GETS user to queue his/her calls and wait 
> for the
> next available; and
>
> 2. With exemption, GETS calls don't get blocked.
>
> With these two designed to support GETS, I think GETS gets better 
> service
> than just "best effort".

I believe that is true *within the administrative boundary of 
RBOCs/CLECs
that have contracts with NCS for GETS* provided the 710 area code is 
being
used.

These features are NOT enabled for GETS within my company's PBXs.
They probably are NOT enabled in many companies' PBXs.  They are
reportedly NOT enabled for the outbound call portion in situations
where the end-user is accessing GETS via the 800 (presumably using 800
because 710 is not supported at the end phone handset).

My question was specific to the situation where the end user is on a 
phone
handset that is connected to a corporate PBX where no GETS contract is
in place and 710 is not a supported area code, that is, my question
was not "in general", but in a specific scenario.

The relevance of this, by the way, is that accessing the Internet in an
unplanned manner from a random tail circuit is comparable to accessing
the PSTN from behind a random corporate PBX.  In both cases, it is hard
to pre-plan for any link congestion that might be present.  In the 
Internet
case, congestion is uncommon inside the core even during a crisis, 
whereas
there are credible reports that the PSTN does suffer congestion inside 
its
core during a crisis.

But thanks anyway,

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 15:21:59 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA16553
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 15:21:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGKOX612139
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 15:24:33 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGKOXv12136
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 15:24:33 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA16543
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 15:21:27 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGKN4v12109;
	Mon, 16 Dec 2002 15:23:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGKMLv12075
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 15:22:21 -0500
Received: from sj-msg-core-3.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA16487
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 15:19:15 -0500 (EST)
Received: from imop.cisco.com (IDENT:mirapoint@imop.cisco.com [171.69.11.44])
	by sj-msg-core-3.cisco.com (8.12.2/8.12.2) with ESMTP id gBGKLLjS002197;
	Mon, 16 Dec 2002 12:21:21 -0800 (PST)
Received: from cisco.com (ssh-sjc-1.cisco.com [171.68.225.134])
	by imop.cisco.com (Mirapoint Messaging Server MOS 3.2.1-GA)
	with ESMTP id ABL37832;
	Mon, 16 Dec 2002 12:18:38 -0800 (PST)
Date: Mon, 16 Dec 2002 12:22:15 -0800
Subject: Re: [Ieprep] RE: PSTN collapse
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep@ietf.org
To: Dave Crocker <dhc2@dcrocker.net>
From: Rohan Mahy <rohan@cisco.com>
In-Reply-To: <18321262523.20021216092344@brandenburg.com>
Message-Id: <112DB9D2-1134-11D7-8C59-0003938AF740@cisco.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, December 16, 2002, at 09:23 AM, Dave Crocker wrote:
> Discussion needs to look for discrete, mechanisms that can 
> "contribute" to
> IEPREP utility through incremental adoption, rather than looking for 
> major
> changes that solve large parts of the problem space.

I absolutely agree.

thanks,
-rohan

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 16:22:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA18720
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 16:22:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGLP4D15280
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 16:25:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGLP4v15277
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 16:25:04 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA18714
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 16:22:00 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGLNBv15189;
	Mon, 16 Dec 2002 16:23:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGLMfv15159
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 16:22:41 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA18660
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 16:19:37 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAH35D>; Mon, 16 Dec 2002 16:20:03 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A414A01@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        "Nguyen, An"
	 <nguyena@ncs.gov>
Cc: "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>,
        "''ieprep ' '"
	 <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: GETS calls placed from corporate PBXs
Date: Mon, 16 Dec 2002 16:19:52 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Just to make clear.

I am not aware of any plan to REPLICATE GETS implementation in the "big I"
Internet.

Some of the "strengths" of GETS are not applicable in the Internet, e.g.
there are no "Network Controls" (such as call gapping) to be exempt from.

But conversely, just because you can't do something in GETS doesn't mean you
shouldn't TRY to do it on the Internet.

Case in point. 
- In GETS, you don't get any special service from the originating handset to
the LEC.  Not because it isn't desirable, but because it isn't practical.
- But in WPS, you DO get special service directly form the originating
handset (queueing for a radio channel).
- If Something analogous is not PRACTICAL in the Internet, so be it.  But
don't use the fact that it is not PRACTICAL in GETS to infer that it is not
DESIRABLE in the Internet.



-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Monday, December 16, 2002 2:09 PM
To: Nguyen, An
Cc: 'King, Kimberly S. '; Gunn, Janet; ''ieprep ' '
Subject: Re: [Ieprep] Re: GETS calls placed from corporate PBXs



On Saturday, Dec 14, 2002, at 21:33 America/Montreal, Nguyen, An wrote:
> Two of several GETS features are "trunking queuing" and exemption from
> management control" that give GETS high probability call completion 
> during
> congestion.
>
> 1. Trunk Queuing allows a GETS user to queue his/her calls and wait 
> for the
> next available; and
>
> 2. With exemption, GETS calls don't get blocked.
>
> With these two designed to support GETS, I think GETS gets better 
> service
> than just "best effort".

I believe that is true *within the administrative boundary of 
RBOCs/CLECs
that have contracts with NCS for GETS* provided the 710 area code is 
being
used.

These features are NOT enabled for GETS within my company's PBXs.
They probably are NOT enabled in many companies' PBXs.  They are
reportedly NOT enabled for the outbound call portion in situations
where the end-user is accessing GETS via the 800 (presumably using 800
because 710 is not supported at the end phone handset).

My question was specific to the situation where the end user is on a 
phone
handset that is connected to a corporate PBX where no GETS contract is
in place and 710 is not a supported area code, that is, my question
was not "in general", but in a specific scenario.

The relevance of this, by the way, is that accessing the Internet in an
unplanned manner from a random tail circuit is comparable to accessing
the PSTN from behind a random corporate PBX.  In both cases, it is hard
to pre-plan for any link congestion that might be present.  In the 
Internet
case, congestion is uncommon inside the core even during a crisis, 
whereas
there are credible reports that the PSTN does suffer congestion inside 
its
core during a crisis.

But thanks anyway,

Ran
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 16:40:02 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19192
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 16:40:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGLgad16615
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 16:42:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGLgav16612
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 16:42:36 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19186
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 16:39:31 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGLf2v16579;
	Mon, 16 Dec 2002 16:41:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGLe5v16550
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 16:40:05 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19156
	for <Ieprep@ietf.org>; Mon, 16 Dec 2002 16:37:01 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for Ieprep@ietf.org; Mon, 16 Dec 2002 16:39:21 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121616394327169
 ; Mon, 16 Dec 2002 16:39:43 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YW14J38N>; Mon, 16 Dec 2002 16:39:30 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64231E67D@mcl-its-exs01.mail.saic.com>
From: "Garrity, Jack" <JOHN.J.GARRITY@saic.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>
Cc: "'King, Kimberly  S.'" <KIMBERLY.S.KING@saic.com>,
        "'Gunn, Janet'" <Janet.Gunn@dyncorp.com>,
        "'Ieprep@ietf.org'" <Ieprep@ietf.org>
Subject: RE: [Ieprep] Re: GETS calls placed from corporate PBXs
Date: Mon, 16 Dec 2002 16:39:36 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Ran,

I hope this isn't disruptive, since you've gotten at least one other
response to your email at the bottom.  Here are my responses to three points
you raised.



[RJA] "...is one operating best-effort on the inbound
      links at least as far as the first RBOC switch?

Everyone competes for dial tone on an equal basis, unless one purchases
Essential Line Service or its equivalent.  Every switching system I know of
has the capability albeit with their own brand name.  I don't know how many
GETS-user organizations have purchased the service, nor how many other
organizations.  My suspicion: next to none.

[RJA] "Is it best effort from the first RBOC switch inward
      to the destination?"

I take it inward means onward.  In most cases, the first RBOC switch invokes
Alternate Carrier Routing and other enhanced GETS features that it has been
paid to provide.  "Most cases" implies greater than 85 percent of the
nation's access lines.  Since many of those access lines serve PBXs, I
imagine the measure implies a larger percentage of potential users.

[RJA] "And is there any pre-emption in GETS?"

Absolutely not.  The received wisdom is: that would violate the
Communications Act of 1934.  As an aside, read the Comm Act and survey the
challenges to that section; there is no real prohibition to preemption.  My
personal view is that the telephone industry has brainwashed government
users for several generations.  If you're worried about preemption, change
the law.

~~~ Jack ~~~
Work: 703/676-2358 | Fax: 703/848-0299
Home: 703/278-9250 | Cell: 703/508-0065
E-mail: garrityj@saic.com


-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Friday, December 13, 2002 4:22 PM
To: Gunn, Janet
Cc: King, Kimberly S.; ieprep
Subject: Re: [Ieprep] Re: GETS calls placed from corporate PBXs


On Friday, Dec 13, 2002, at 16:12 America/Montreal, Gunn, Janet wrote:

> If 710 doesn't work from a particular phone, there are some 800 numbers
> which will take you straight to a GETS capable IXC.  So as long as 800
> works, you are OK.

OK, that's useful.  But in such a case, is one operating best-effort
on the inbound links at least as far as the first RBOC switch ?
Is it best effort from the first RBOC switch inward to the destination ?

And is there any pre-emption in GETS ?  That is will a GETS call cause
a non-GETS call that was already placed and is currently in use to get
dropped.  My earlier understanding was "no", but I want to be sure I
understood correctly.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 16:57:53 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19710
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 16:57:53 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGM0RX17728
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 17:00:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGM0Rv17725
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 17:00:27 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19693
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 16:57:22 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGLx3v17408;
	Mon, 16 Dec 2002 16:59:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGLwAv17374
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 16:58:10 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA19548
	for <Ieprep@ietf.org>; Mon, 16 Dec 2002 16:55:06 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id DEAE167103; Mon, 16 Dec 2002 16:58:52 -0500 (EST)
Date: Mon, 16 Dec 2002 16:58:02 -0500
Subject: Re: [Ieprep] Re: GETS calls placed from corporate PBXs
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'Ieprep@ietf.org'" <Ieprep@ietf.org>
To: "Garrity, Jack" <JOHN.J.GARRITY@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64231E67D@mcl-its-exs01.mail.saic.com>
Message-Id: <72BA5ACE-1141-11D7-8143-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


Thanks very much.  The clarity is very helpful.


On Monday, Dec 16, 2002, at 16:39 America/Montreal, Garrity, Jack wrote:
> [RJA] "Is it best effort from the first RBOC switch inward
>       to the destination?"
>
> I take it inward means onward.  In most cases, the first RBOC switch 
> invokes
> Alternate Carrier Routing and other enhanced GETS features that it has 
> been
> paid to provide.  "Most cases" implies greater than 85 percent of the
> nation's access lines.  Since many of those access lines serve PBXs, I
> imagine the measure implies a larger percentage of potential users.

OK.  So there should be preferred access from the RBOC switch on the
T1 (or whatever) that connects to the destination's PBX.

(And I suppose what happens from the destination gateway PBX depends
on how that company/organisation has configured their PBX.)

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 17:05:21 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA19855
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 17:05:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGM7tn18792
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 17:07:55 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGM7tv18789
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 17:07:55 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA19844
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 17:04:50 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGM6Cv18083;
	Mon, 16 Dec 2002 17:06:15 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGM5Hv18051
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 17:05:17 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA19792
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 17:02:12 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id EFBE067103; Mon, 16 Dec 2002 17:05:59 -0500 (EST)
Date: Mon, 16 Dec 2002 17:05:09 -0500
Subject: Re: [Ieprep] Re: GETS calls placed from corporate PBXs
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "''ieprep ' '" <ieprep@ietf.org>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <CBED705A7FD2D311865500508B1089410A414A01@chntex02.is.dyncorp.com>
Message-Id: <714CC34E-1142-11D7-8143-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 16, 2002, at 16:19 America/Montreal, Gunn, Janet wrote:
> I am not aware of any plan to REPLICATE GETS implementation in the 
> "big I"
> Internet.

Very good to know.  There has been confusion on this point.

> Some of the "strengths" of GETS are not applicable in the Internet, 
> e.g.
> there are no "Network Controls" (such as call gapping) to be exempt 
> from.
>
> But conversely, just because you can't do something in GETS doesn't 
> mean you
> shouldn't TRY to do it on the Internet.
>
> Case in point.
> - In GETS, you don't get any special service from the originating 
> handset to
> the LEC.  Not because it isn't desirable, but because it isn't 
> practical.
> - But in WPS, you DO get special service directly form the originating
> handset (queueing for a radio channel).
> - If Something analogous is not PRACTICAL in the Internet, so be it.  
> But
> don't use the fact that it is not PRACTICAL in GETS to infer that it 
> is not
> DESIRABLE in the Internet.

Fair enough, but maybe the above caveat about "practical" also means 
that at least
some things (e.g. end-to-end QoS mechanisms getting deployed widely) 
that some folks
desire are not likely to happen either here in IEprep or in the 
deployed Internet.

Of course, there are also various things that might be practical (albeit
tedious to get all the end-user contracts in place and configure all 
the boxes),
many of which (e.g. edge router CoS) have already been disussed here.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 18:36:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA21684
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 18:36:30 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBGNd5m23707
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 18:39:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGNd5v23704
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 18:39:05 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA21679
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 18:35:59 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGNbOv23650;
	Mon, 16 Dec 2002 18:37:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBGNaBv23018
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 18:36:11 -0500
Received: from joy.songbird.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA21611
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 18:33:05 -0500 (EST)
Received: from bbprime.brandenburg.com (208.184.79.252.songbird.com [208.184.79.252] (may be forged))
	by joy.songbird.com (8.9.3/8.9.3) with ESMTP id PAA17769;
	Mon, 16 Dec 2002 15:35:47 -0800
Date: Mon, 16 Dec 2002 15:35:59 -0800
From: Dave Crocker <dhc2@dcrocker.net>
X-Mailer: The Bat! (v1.62 Christmas Edition) Personal
Organization: Brandenburg InternetWorking
X-Priority: 3 (Normal)
Message-ID: <1943598000.20021216153559@brandenburg.com>
To: RJ Atkinson <rja@extremenetworks.com>
CC: ieprep@ietf.org
Subject: Re: [Ieprep] RE: PSTN collapse
In-Reply-To: <FCFF982A-1128-11D7-91A2-00039357A82A@extremenetworks.com>
References: <FCFF982A-1128-11D7-91A2-00039357A82A@extremenetworks.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

RJ,

Monday, December 16, 2002, 11:02:57 AM, you wrote:
RJ> Such shutdowns are normally performed on inbound calls from outside
...
RJ> Of course, I don't know what was in your presentation, so can't say
RJ> whether I'd agree with the phone company assertions or not.

The anecdote was about outbound calls, but the nature of GETS makes it
pretty clear that a sufficient emergency will close the system in both
directions, except for 'authorized' users.

FYI, the anecdote was about Van Jacobsen being at Stanford when the quake
hit, and immediately wanting to contact his girlfriend, back in Berkeley.
He tried to call, but of course that did not work.

Out of frustration, he sent her some email, though I am told he only did
that for the sake of having something to do.  However, he got a response
from her within a relatively short time.

He was astonished that things were working that well, so he did a traceroute
to find the path.

Sure enough, the Internet path from Stanford to Berekely went down to L.A.,
across the U.S. and then back across to Berekely.

Or, at least, that is the way I heard the story.

d/
-- 
 Dave <mailto:dhc2@dcrocker.net>
 Brandenburg InternetWorking <http://www.brandenburg.com>
 t +1.408.246.8253; f +1.408.850.1850

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 19:28:20 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA22471
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 19:28:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBH0UtR25447
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 19:30:55 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH0Utv25444
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 19:30:55 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA22465
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 19:27:49 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH0T3v25399;
	Mon, 16 Dec 2002 19:29:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH0Ssv25384
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 19:28:54 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA22453
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 19:25:47 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBH0MNo00388;
	Mon, 16 Dec 2002 16:22:23 -0800 (PST)
Date: Mon, 16 Dec 2002 16:22:23 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
cc: "'RJ Atkinson'" <rja@extremenetworks.com>, "Nguyen, An" <nguyena@ncs.gov>,
        "'King, Kimberly  S. '" <KIMBERLY.S.KING@saic.com>,
        "''ieprep ' '" <ieprep@ietf.org>
Subject: RE: [Ieprep] Re: GETS calls placed from corporate PBXs
In-Reply-To: <CBED705A7FD2D311865500508B1089410A414A01@chntex02.is.dyncorp.com>
Message-ID: <Pine.GSO.4.44.0212161613350.317-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Mon, 16 Dec 2002, Gunn, Janet wrote:
    > Just because you can't do something in GETS doesn't mean you
    > shouldn't TRY to do it on the Internet.

I'm in complete agreement.  The strengths of VoIP and the Internet
generally are vastly different than those of the PSTN, and I think we can
expect to see a few really useful functions which would be difficult to
perform in another environment.  For example:

1) Handset-to-handset cryptographic authentication.

2) Handset-to-handset evaluation of loss and per-handset-pair tuning of
   forward error correction.

3) Persistent connections, which stay up through radical re-routing of the
   actual path of the traffic, whether with user-perceivable drop-outs or
   not.  This is similar to the robustness which the cellular network
   displays during cell-to-cell handoffs, but can potentially withstand
   much longer periods of temporarily disruption.

4) Multiply-hubbed partial-mesh call-bridges, so that a conference call
   can withstand the loss of one or more audio-stream multiplexing
   facilities or loss of connectivity between some participant pairs,
   without actually dropping anyone off the call.  This is a big one,
   wich can only be done in a packet-switched environment.

I'm sure many of you can imagine many other similar advantages.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 19:43:20 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA22669
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 19:43:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBH0jue26439
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 19:45:56 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH0juv26436
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 19:45:56 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA22663
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 19:42:49 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH0i4v26383;
	Mon, 16 Dec 2002 19:44:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH0hEv26359
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 19:43:14 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA22606
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 19:40:07 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBH0apo00562;
	Mon, 16 Dec 2002 16:36:51 -0800 (PST)
Date: Mon, 16 Dec 2002 16:36:51 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: Re: [Ieprep] Understanding Congestion
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77A77@mcl-its-exs01.mail.saic.com>
Message-ID: <Pine.GSO.4.44.0212161624450.317-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      Kimberly King quotes Steve Bellovin:
    > Congestion management in the Internet is problematic, in part, because no
    > capabilities exist for managing traffic associated with specific users,
    > connections, sources, or destinations, and it would be difficult to
    > implement such capabilities.

While the former was true at the time Steve wrote this in 1998, it is no
longer so.  The latter, that it was difficult to implement, was
demonstrated to be true.  :-)

The fix was the combination of uRPF, real-time backscatter analysis, and
the use of iBGP to distribute blackhole sources/destinations within the
backbone.

The limitation is the same as that of any kind of tagging: you can't trust
such information entering from outside the AS, since it's a DoS attack
rather than valid.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 16 20:20:07 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA23146
	for <ieprep-archive@odin.ietf.org>; Mon, 16 Dec 2002 20:20:07 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBH1Mi828029
	for ieprep-archive@odin.ietf.org; Mon, 16 Dec 2002 20:22:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH1Miv28026
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 16 Dec 2002 20:22:44 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA23143
	for <ieprep-web-archive@ietf.org>; Mon, 16 Dec 2002 20:19:36 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH1L4v27949;
	Mon, 16 Dec 2002 20:21:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH1Kpv27911
	for <ieprep@optimus.ietf.org>; Mon, 16 Dec 2002 20:20:51 -0500
Received: from kc-msxproto4.kc.umkc.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA23117
	for <ieprep@ietf.org>; Mon, 16 Dec 2002 20:17:44 -0500 (EST)
Received: from KC-MAIL2.kc.umkc.edu ([134.193.143.162] RDNS failed) by kc-msxproto4.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.5329);
	 Mon, 16 Dec 2002 19:20:43 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: RE: [Ieprep] Understanding Congestion
Date: Mon, 16 Dec 2002 19:20:42 -0600
Message-ID: <A29143A40AEAE3459FF829D1DD43316101D9B0A4@KC-MAIL2.kc.umkc.edu>
Thread-Topic: [Ieprep] Understanding Congestion
Thread-Index: AcKlZWRLIstb2wbpSFOx0rSYAn0cOgAAzfQw
From: "Ayyasamy, Senthilkumar  (UMKC-Student)" <saq66@umkc.edu>
To: "Bill Woodcock" <woody@pch.net>,
        "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: <ieprep@ietf.org>
X-OriginalArrivalTime: 17 Dec 2002 01:20:43.0091 (UTC) FILETIME=[84AE0A30:01C2A56A]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBH1Kpv27912
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


> > Congestion management in the Internet is problematic, in part, because no
> > capabilities exist for managing traffic associated with  specific users,
> > connections, sources, or destinations, 

In addition to Bill's list, a simple ACL is the fundamental thing for 
doing the above stuff.

> 
> The fix was the combination of uRPF, real-time backscatter 
> analysis,

and sink holes

>  and the use of iBGP to distribute blackhole sources/destinations 
> within the  backbone.

Black hole filtering will be useful layer 3 work and not layer 4. 
But that is not a major limitation.

> The limitation is the same as that of any kind of tagging: 
> you can't trust such information entering from outside the 
> AS, since it's a DoS attack  rather than valid.

yaa.. Black-hole filtering can come in handy for DoS attackers.
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 02:36:34 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA08231
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 02:36:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBH7dED24443
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 02:39:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH7dEv24440
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 02:39:14 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA08227
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 02:36:03 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH7bRv24319;
	Tue, 17 Dec 2002 02:37:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBH7ajv23724
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 02:36:45 -0500
Received: from kcmso2.proxy.att.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id CAA08187
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 02:33:34 -0500 (EST)
Received: from attrh3i.attrh.att.com ([135.71.62.12])
	by kcmso2.proxy.att.com (AT&T IPNS/MSO-4.0) with ESMTP id gBH704Dv018100
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 01:36:34 -0600 (CST)
Received: from OCCLUST03EVS1.ugd.att.com (135.71.164.11) by attrh3i.attrh.att.com (6.5.032)
        id 3DF6BD4E0031A9C5; Tue, 17 Dec 2002 02:36:29 -0500
X-MimeOLE: Produced By Microsoft Exchange V6.0.5762.3
content-class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="us-ascii"
Subject: RE: [Ieprep] Understanding Congestion
Date: Tue, 17 Dec 2002 02:36:28 -0500
Message-ID: <456FD0E7B215B24ABBAD711614E4A7A201842549@OCCLUST03EVS1.ugd.att.com>
Thread-Topic: [Ieprep] Understanding Congestion
Thread-Index: AcKlNZqQOVWCLRpySf+yPRNqqWs8LwAZpaYA
From: "Dvorak, Charles A (Chuck), ALASO" <cdvorak@att.com>
To: "RJ Atkinson" <rja@extremenetworks.com>, "ieprep" <ieprep@ietf.org>
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBH7ajv23725
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

FYI, systems for dealing with dynamic load (when fixed bandwidth is
getting congested) have been around since the 60s--TASI (time assignment
speech interpolation, I think) which was analog and then DCME (digital
circuit multiplication eqpt) in the early 1980s, followed by PCME
(packet circuit multiplication eqpt) in the late 1980s. Several US and
ITU standards were produced on these techniques, which were widely used
in international network segments.

Each of these systems produced certain speech impairments under severe
load, but even so they were capable of transporting up to 8 or 10 times
the traffic that the bandwidth would have otherwise supported. (Data
traffic was not compressed as much as voice, if at all).

In the 1990s carriers deployed real-time network routing, which was more
powerful than the non-RT dynamic routing techniques deployed in the
1980s. Both methods route calls to paths that are not congested.

The PSTN did *better* than having a "concept of congestion avoidance."
It actually *used it* decades ago, and probably everyone reading this
message has sometime in their life spoken over a phone call using one or
more of these techniques.

> -----Original Message-----
> From: RJ Atkinson [mailto:rja@extremenetworks.com]
> Sent: Monday, December 16, 2002 2:01 PM
> To: ieprep
> Subject: Re: [Ieprep] Understanding Congestion
> 
> A key item to understand is that the PSTN does not really have the
> concept of congestion avoidance (e.g. slow down the bits/second rate
> of a session to help the network avoid being congested), whereas that
> is central to the Internet.
> 
> This is not a better/worse thing, just as call rejection capability
> isn't a better/worse thing.  The architectures are fundamentally
> different.  One needs to work with the architecture for the system
> in hand, not some other architecture.
> 
> Ran
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep
> 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 09:28:34 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA14495
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 09:28:34 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHEV5a13463
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 09:31:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHEV5v13460
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 09:31:05 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA14490
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 09:28:03 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHET5v13383;
	Tue, 17 Dec 2002 09:29:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHESpv13352
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 09:28:51 -0500
Received: from imo-m03.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA14426
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 09:25:48 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-m03.mx.aol.com (mail_out_v34.13.) id i.174.13b0670d (3850);
	Tue, 17 Dec 2002 09:28:44 -0500 (EST)
Message-ID: <174.13b0670d.2b308e9c@aol.com>
Date: Tue, 17 Dec 2002 09:28:44 EST
Subject: Re: [Ieprep] updated requirements documents (telephony)
To: K.Carlberg@cs.ucl.ac.uk, ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_174.13b0670d.2b308e9c_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_174.13b0670d.2b308e9c_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/14/2002 9:42:08 AM Eastern Standard Time, 
K.Carlberg@cs.ucl.ac.uk writes:


[KC] I'm sorry, but its hard to dig out what the actual question is.  if
there is still an open question, could it be rephrased?  otherwise, I'll
assume that the thread is closed and stick to my previous comments

[MAP] The original comment was mostly an editorial one as follows:

---------------------------------------------------------
I suggest this section on Alternate Paths be split into two separate 
sub-parts to address the two "layers" within IP and relate each to what the 
PSTN did (or couldn't do) as follows: 

1) Alternate paths 

Experience with GETS over the PSTN has shown the utility of alternate paths 
to a destination to help facilitate emergency-related communications. This is 
done at the application layer, independent of any lower layer rerouting 
(SONET rings, etc) that naturally applies to all traffic using the link, 
whether emergency or normal. Application of emergency telephony services to 
the Internet (and IP in general) means that there are two separate issues to 
be addressed. 

a) IP Packet Transport layer: Unlike the PSTN, which creates a fixed path 
(circuit) during call setup phase, the Internet uses dynamic routing for IP 
packets. This dynamic routing capability automatically causes IP packets to 
travel the best current path, which may change during the life of the "call" 
due to failures of infrastructure components. However, at the time of "call" 
set up (at the application layer), the route choice for the packet transport 
layer has a single choice (the best current one). An issue to be addressed is 
the utility and possibility of providing additional routes (even if less 
efficient) for packet flows for higher priority "calls". 

b) Call application layer:  The Internet infrastructure itself does not have 
the concept of a "call" or the concept of "call setup", although IP telephony 
applications using it would have application-layer awareness of calls or the 
call setup concept. Alternate path routing at the application layer may be 
useful in choosing one of possibly several egress points to the PSTN from the 
Internet. 

--------------------------------------------
This was mostly to help avoid the problems caused by the confusion between 
the two layers when each person focuses on a different part when reading it. 
Essentially, the only idea I added was the statement about "An issue to be 
addressed is the utility and possibility of providing additional routes" (for 
the IP transport layer), an idea already hinted at by the statement in the 
draft that is "may be difficult to achieve".

Mike Pierce
Artel

--part1_174.13b0670d.2b308e9c_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/14/2002 9:42:08 AM Eastern Standard Time, K.Carlberg@cs.ucl.ac.uk writes:
<BR>
<BR>
<BR>[KC] I'm sorry, but its hard to dig out what the actual question is. &nbsp;if
<BR>there is still an open question, could it be rephrased? &nbsp;otherwise, I'll
<BR>assume that the thread is closed and stick to my previous comments
<BR>
<BR>[MAP] The original comment was mostly an editorial one as follows:
<BR>
<BR>---------------------------------------------------------
<BR>I suggest this section on Alternate Paths be split into two separate sub-parts to address the two "layers" within IP and relate each to what the PSTN did (or couldn't do) as follows: 
<BR>
<BR>1) Alternate paths 
<BR>
<BR>Experience with GETS over the PSTN has shown the utility of alternate paths to a destination to help facilitate emergency-related communications. This is done at the application layer, independent of any lower layer rerouting (SONET rings, etc) that naturally applies to all traffic using the link, whether emergency or normal. Application of emergency telephony services to the Internet (and IP in general) means that there are two separate issues to be addressed. 
<BR>
<BR>a) IP Packet Transport layer: Unlike the PSTN, which creates a fixed path (circuit) during call setup phase, the Internet uses dynamic routing for IP packets. This dynamic routing capability automatically causes IP packets to travel the best current path, which may change during the life of the "call" due to failures of infrastructure components. However, at the time of "call" set up (at the application layer), the route choice for the packet transport layer has a single choice (the best current one). An issue to be addressed is the utility and possibility of providing additional routes (even if less efficient) for packet flows for higher priority "calls". 
<BR>
<BR>b) Call application layer: &nbsp;The Internet infrastructure itself does not have the concept of a "call" or the concept of "call setup", although IP telephony applications using it would have application-layer awareness of calls or the call setup concept. Alternate path routing at the application layer may be useful in choosing one of possibly several egress points to the PSTN from the Internet. 
<BR>
<BR>--------------------------------------------
<BR>This was mostly to help avoid the problems caused by the confusion between the two layers when each person focuses on a different part when reading it. Essentially, the only idea I added was the statement about "An issue to be addressed is the utility and possibility of providing additional routes" (for the IP transport layer), an idea already hinted at by the statement in the draft that is "may be difficult to achieve".
<BR>
<BR>Mike Pierce
<BR>Artel</FONT></HTML>

--part1_174.13b0670d.2b308e9c_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 09:41:19 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA14742
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 09:41:19 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHEhoO14532
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 09:43:50 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHEhov14529
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 09:43:50 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA14725
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 09:40:48 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHEg2v14447;
	Tue, 17 Dec 2002 09:42:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHEftv14433
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 09:41:55 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA14699
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 09:38:47 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 8515667103; Tue, 17 Dec 2002 09:42:39 -0500 (EST)
Date: Tue, 17 Dec 2002 09:41:41 -0500
Subject: Re: [Ieprep] Understanding Congestion
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "ieprep" <ieprep@ietf.org>
To: "Dvorak, Charles A (Chuck), ALASO" <cdvorak@att.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <456FD0E7B215B24ABBAD711614E4A7A201842549@OCCLUST03EVS1.ugd.att.com>
Message-Id: <A80C5622-11CD-11D7-80DC-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 17, 2002, at 02:36 America/Montreal, Dvorak, Charles A 
(Chuck), ALASO wrote:
> The PSTN did *better* than having a "concept of congestion avoidance."
> It actually *used it* decades ago, and probably everyone reading this
> message has sometime in their life spoken over a phone call using one 
> or
> more of these techniques.

A POTS handset does not have a codec or any concept of congestion 
avoidance.
The PSTN has never had end-to-end congestion avoidance as one result.  
The Internet
has since [VJ88].  As discussion here has repeatedly noted, the 
potential congestion
point is on the access link, not in the core, so the existence of 
PSTN-unique
technology for congestion avoidance in the PSTN core is not 
tremendously relevant
to IEprep.  Fortunately for us, the Internet does not have significant 
congestion
in the core, even during natural disasters or man-made disasters.

The problem spaces are DIFFERENT.  The solution spaces will necessarily
also be DIFFERENT.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 10:32:41 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA15712
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 10:32:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHFZDb16968
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 10:35:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHFZDv16965
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 10:35:13 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA15707
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 10:32:10 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHFXHv16856;
	Tue, 17 Dec 2002 10:33:17 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHFWkv16822
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 10:32:46 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA15640
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 10:29:42 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Tue, 17 Dec 2002 10:32:04 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121710322820095
 ; Tue, 17 Dec 2002 10:32:28 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <ZBBRFLY9>; Tue, 17 Dec 2002 10:34:03 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77AA4@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Bill Woodcock'" <woody@pch.net>,
        "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
Date: Tue, 17 Dec 2002 10:32:27 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

> 
>       Kimberly King quotes Steve Bellovin:
>     > Congestion management in the Internet is problematic, 
> in part, because no
>     > capabilities exist for managing traffic associated with 
> specific users,
>     > connections, sources, or destinations, and it would be 
> difficult to
>     > implement such capabilities.
> 
> While the former was true at the time Steve wrote this in 
> 1998, it is no
> longer so.  The latter, that it was difficult to implement, was
> demonstrated to be true.  :-)
> 
> The fix was the combination of uRPF, real-time backscatter 
> analysis, and
> the use of iBGP to distribute blackhole sources/destinations 
> within the
> backbone.
> 
> The limitation is the same as that of any kind of tagging: 
> you can't trust
> such information entering from outside the AS, since it's a DoS attack
> rather than valid.

I understand your point here but just to be clear, 
prohibiting denial of service attacks that
are based on forged source addresses and doing it in a
scaleable accurate way against address aggregates
is a different problem than congestion management of
individual flows or user sessions.

The case that has been worked on is illegitimate bulk
traffic at layer 3... legitimate traffic from customers
at the session layer seems to be a completely different
problem to me...

Kimberly



Disclaimer:  My statements represent my own thinking and do not necessarily 
reflect the position of anyone else in the world including my employer. 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 10:36:11 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA15798
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 10:36:11 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHFciK17696
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 10:38:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHFciv17693
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 10:38:44 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA15785
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 10:35:40 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHFb1v17056;
	Tue, 17 Dec 2002 10:37:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHFa4v16987
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 10:36:04 -0500
Received: from chntex03.is.dyncorp.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA15730
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 10:33:00 -0500 (EST)
Received: by chntex03.is.dyncorp.com with Internet Mail Service (5.5.2653.19)
	id <XDXAHQ4P>; Tue, 17 Dec 2002 10:33:27 -0500
Message-ID: <CBED705A7FD2D311865500508B1089410A414A0A@chntex02.is.dyncorp.com>
From: "Gunn, Janet" <Janet.Gunn@DynCorp.com>
To: "'Dave Crocker'" <dhc2@dcrocker.net>,
        RJ Atkinson
	 <rja@extremenetworks.com>
Cc: ieprep@ietf.org
Subject: Network controls RE: [Ieprep] RE: PSTN collapse
Date: Tue, 17 Dec 2002 10:33:18 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


The PSTN  has a wide variety of network controls that can be activated,
depending on the exact nature of the situation.

These include 
Blocking all calls OUT of the area  identified
Blocking all calls INTO the destination identified
Regulating the rate at which calls are submitted to a destination.

And a bunch of others.

The choice of which to activate varies on a case by case basis, based not
just on the traffic itself, but on the external cause.  For instance, in a
natural emergency, you may want people in the area to be able to make "I'm
OK" outbound calls (while blocking incoming calls), but in the case of
something like the Oklahoma City bombing, you might block both incoming and
outgoing calls, to make resources available for communication within the
area.

One of the things that GETS does is to provide exemption form these
controls.

But it is NOT an indication of the PSTN "collapsing"- it is an indication of
the PSTN adapting to prevent collapse. (If you didn't do it, you could end
up with lots of resources being wasted trying to set up calls which could
not complete, blocking the calls which COULD complete.)

-----Original Message-----
From: Dave Crocker [mailto:dhc2@dcrocker.net]
Sent: Monday, December 16, 2002 6:36 PM
To: RJ Atkinson
Cc: ieprep@ietf.org
Subject: Re: [Ieprep] RE: PSTN collapse


RJ,

Monday, December 16, 2002, 11:02:57 AM, you wrote:
RJ> Such shutdowns are normally performed on inbound calls from outside
...
RJ> Of course, I don't know what was in your presentation, so can't say
RJ> whether I'd agree with the phone company assertions or not.

The anecdote was about outbound calls, but the nature of GETS makes it
pretty clear that a sufficient emergency will close the system in both
directions, except for 'authorized' users.

FYI, the anecdote was about Van Jacobsen being at Stanford when the quake
hit, and immediately wanting to contact his girlfriend, back in Berkeley.
He tried to call, but of course that did not work.

Out of frustration, he sent her some email, though I am told he only did
that for the sake of having something to do.  However, he got a response
from her within a relatively short time.

He was astonished that things were working that well, so he did a traceroute
to find the path.

Sure enough, the Internet path from Stanford to Berekely went down to L.A.,
across the U.S. and then back across to Berekely.

Or, at least, that is the way I heard the story.

d/
-- 
 Dave <mailto:dhc2@dcrocker.net>
 Brandenburg InternetWorking <http://www.brandenburg.com>
 t +1.408.246.8253; f +1.408.850.1850

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 11:49:36 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA17237
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 11:49:36 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHGq8M21521
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 11:52:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHGq7v21518
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 11:52:07 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA17233
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 11:49:05 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHGo5v21453;
	Tue, 17 Dec 2002 11:50:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHGnqv21425
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 11:49:52 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA17192
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 11:46:49 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Tue, 17 Dec 2002 11:49:07 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121711493220475
 for <ieprep@ietf.org>; Tue, 17 Dec 2002 11:49:32 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <ZBBRF3N4>; Tue, 17 Dec 2002 11:51:07 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77AA8@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "Ieprep (E-mail)" <ieprep@ietf.org>
Date: Tue, 17 Dec 2002 11:49:29 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] Summary of ieprep work and VoIP--Take 2
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

I have created a summary below organizing information into various
contextual 
situations and options that I hope will provide some clarity.
______________________


We assume an ETS location is connected via an access link to the Internet. 
We also take the assumption (largely substantiated for many U.S. regions)
that network congestion at layer 3, if it exists, is on the access link.  

We assume the access link carries both VoIP signaling and VoIP media
streams. 
If other non-VoIP traffic is also carried, then there is a tradeoff to make 
between the performance of voice and non-voice applications.  For 
example, enough VoIP on a congested link (e.g., 10-40% packet drop rates) 
can be sufficient to make the link essentially unusable for TCP traffic.  

Case A: Bundled Internet access and SIP service provider. 
In this case, there is a predefined relationship between SIP provider and
ISP
providing access to the ETS location (e.g., WorldCom SIP service) and this 
is known in advance of an emergency.  Both preferential signaling and CoS 
type techniques are potentially applicable.  

Case A Signaling Technique 1 
Assume there is a SIP resource priority mechanism available to
identify emergency calls.  (Henning's Requirements document on this
subject will be examined by the SIPPING working group and we hope
the outcome will be such a mechanism in SIP.)  
This mechanism could identify emergency calls thereby facilitating
preferential access to ports on PSTN gateways or other end systems.

Case A Signaling Technique 2
If the provider runs stateful SIP proxies, the new resource priority
mechanism could also allow preferential treatment of proxy resources.  

Case A Signaling Technique 3 
One highly stateful solution would be for the provider to track how 
many calls are coming to/from which locations' access link, and apply a 
traditional call admission control.  If the provider has multiple SIP
servers, then 
this will further require communication between those servers to track 
coordinate information.  Again, either higher probability of call setup 
or pre-emption could be supported.  The WG recognizes that one of the major
aspects of VoIP is the decoupling of signaling and the switching matrix 
and hence solutions like this will have to approximate congestion based upon
VoIP
call counts over predetermined access links and will have some interesting 
scaling properties.  

Case A. Signaling Technique 4
The number of calls traversing the access link can be controlled by placing
a middle box (e.g., stateful SIP-proxy firewall, SIP-aware QoS middlebox) on
the 
customer premise at the access link.  This middle box can be
provisioned to allow a predetermined number of concurrent VoIP calls over
the
access link and deny further call attempts when that maximum number is met,
analogous to a traditional PBX.  However, this middle box could be
enhanced to understand call priority and allow preferential treatment to 
emergency calls by either granting them a greater probability of call setup
during
times when the link is close to the predetermined maximum or by enabling 
preemption.  

Case A. CoS on the Access link Technique 1
The customer may apply CoS at the CPE (e.g., router filters, switch filters,
QoS middleboxes).  Also, Link-layer QoS mechanisms should be considered for 
deployment as part of an overall system, when such mechanisms exist.

Case A. CoS on the Access link Technique 2
The service provider may also apply CoS at the service provider end of
the access link(e.g., Sprint CoS service).  If a policy and deployment is
agreed 
upon by the operators of that link, it is also possible to prioritize
selected IP 
flows of nearly any application protocol.  Note that some protocols do not
use 
fixed ports, so one can't use normal filters to handle those application 
protocols such as the data streams for VoIP traffic.


Case B.  The ISP provides a CoS service but any SIP services come from a
provider whose service is distributed in other places in the Internet.
Basically, here the ISP provides transport with some CoS but isn't coupled
with a SIP service.

Signaling Technique 1,2 and 4 are potentially applicable from Case A above.
Signaling Technique 3 is no longer appropriate since
the SIP provider has no information about how the
user is obtaining access, it doesn't really make sense to do "call admission
control".  (For example, 10 users at the same location could rely on 10
different SIP service providers - none of the service provider can determine
what is flowing over the access link they happen to all be using.)  

CoS on the Access link Technique 1 and 2 also apply.



Case C.  The ISP provides a best effort service and any SIP services come
from a provider whose service is distributed in many places in the Internet.

Signaling Technique 1,2 and 4 are applicable from Case A above.

CoS on the Access link Technique 1 also applies.

Open Issues:
- Initial codec selection
-Codec changes due to congestion
-Additional mechanisms


Kimberly



Disclaimer:  My statements represent my own thinking and do not necessarily 
reflect the position of anyone else in the world including my employer. 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 12:33:59 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA18504
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 12:33:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHHaVM24346
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 12:36:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHHaVv24343
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 12:36:31 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA18486
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 12:33:28 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHHZ7v24242;
	Tue, 17 Dec 2002 12:35:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHHYjv24205
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 12:34:45 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA18409
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 12:31:42 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 96E7767103; Tue, 17 Dec 2002 12:35:39 -0500 (EST)
Date: Tue, 17 Dec 2002 12:34:40 -0500
Subject: Re: [Ieprep] Summary of ieprep work and VoIP--Take 2
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "Ieprep (E-mail)" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77AA8@mcl-its-exs01.mail.saic.com>
Message-Id: <D26F7E8A-11E5-11D7-80DC-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 17, 2002, at 11:49 America/Montreal, King, Kimberly S. 
wrote:
> For example, enough VoIP on a congested link (e.g., 10-40% packet drop 
> rates)
> can be sufficient to make the link essentially unusable for TCP 
> traffic.

I think that overstates the case.  There is evidence (e.g. USN 
deployments
of IP to ships at sea) that TCP/IPv4 works fine over 2.4 Kbps WAN radio 
links.
Unless the non-congestion-avoiding traffic consumes *all* of the link 
bandwidth,
TCP/SCTP will still work.  The response times of interactive 
congestion-avoiding
traffic traffic (whether TCP, SCTP, or other) will be adversely 
impacted by the
reduction in offered load caused by the congestion avoidance 
mechanisms, but that
will not make the link "essentially unusuable" for such traffic.

In particular, I myself have used a shared 2.4 Kbps RF link for 
interactive
(Telnet/TCP/IPv4) traffic halfway around the world (~Asia to US East 
Coast)
to read email.  It works OK.  It would be more pleasant with more 
bandwidth,
but it is not the case that TCP didn't work adequately on a shared very
low bandwidth very high latency link.

We should also be careful to note the stated IAB concern from Sally 
Floyd that
congestion-avoidance needs to be added to VoIP application protocols 
(e.g. RTP)
if VJ-compatible congestion avoidance is not already present in those 
protocols.
It is my understanding that work is in-progress on this topic in 
various IETF
Working Groups, so it is likely that VoIP application protocols will be 
implementing
and deploying VJ-compatible congestion avoidance in future.  This, by 
the way,
implies that those application protocols will be doing some form of 
codec
changing based in part on the degree of congestion present end-to-end 
(if any
congestion is present).

> Case A. CoS on the Access link Technique 2
> The service provider may also apply CoS at the service provider end of
> the access link(e.g., Sprint CoS service).  If a policy and deployment 
> is
> agreed
> upon by the operators of that link, it is also possible to prioritize
> selected IP
> flows of nearly any application protocol.  Note that some protocols do 
> not
> use
> fixed ports, so one can't use normal filters to handle those 
> application
> protocols such as the data streams for VoIP traffic.

Editing for clarity, I might edit the last sentence to read more like 
this:

	Note that some application-layer protocols (e.g. VoIP sessions) do not 
use
	fixed transport-layer ports, so one might not be able to use 
port-centric
	filters to cause those application protocols to be handled differently 
than
	other IP traffic.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 12:58:04 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA18988
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 12:58:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHI0ai25952
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 13:00:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHI0av25949
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 13:00:36 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA18984
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 12:57:33 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHHx4v25864;
	Tue, 17 Dec 2002 12:59:04 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHHwpv25848
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 12:58:51 -0500
Received: from kc-msxproto2.kc.umkc.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA18952
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 12:55:47 -0500 (EST)
Received: from KC-MAIL2.kc.umkc.edu ([134.193.143.162] RDNS failed) by kc-msxproto2.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.5329);
	 Tue, 17 Dec 2002 11:58:46 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: RE: [Ieprep] Summary of ieprep work and VoIP--Take 2
Date: Tue, 17 Dec 2002 11:58:45 -0600
Message-ID: <A29143A40AEAE3459FF829D1DD43316101D9B0A7@KC-MAIL2.kc.umkc.edu>
Thread-Topic: [Ieprep] Summary of ieprep work and VoIP--Take 2
Thread-Index: AcKl8qXgX8vwZqf9RmGz4GdMXBcCGQAABmgQ
From: "Ayyasamy, Senthilkumar  (UMKC-Student)" <saq66@umkc.edu>
To: "RJ Atkinson" <rja@extremenetworks.com>,
        "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: "Ieprep (E-mail)" <ieprep@ietf.org>, <floyd@icir.org>
X-OriginalArrivalTime: 17 Dec 2002 17:58:46.0686 (UTC) FILETIME=[F2157BE0:01C2A5F5]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBHHwpv25849
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


> 
> We should also be careful to note the stated IAB concern from Sally 
> Floyd that congestion-avoidance needs to be added to VoIP application
> protocols  (e.g. RTP) if VJ-compatible congestion avoidance is not 
> already present in those  protocols. 
 
Clipped from dccp discussion:
http://www1.ietf.org/mail-archive/working-groups/dccp/current/msg00075.html

   > * The original remarks in the IEPREP meeting (diatribe that 
   > could be skipped): 
   > 
   > In the IEPREP meeting, the example that occasioned my remark was
   > that of a best-effort, VoIP connection to Nairobi, with good quality,
   > with reports on the viewgraph of typical packet drop rates of 5-40%.
   > It turned out that the VoIP connection was sending 64Kbps, on a
   > 128Kbps link with many users sharing the link.  The voice quality
   > was good.  My question was about the experience of the TCP connections
   > sharing the link at the time.  My own (strong) view is that a
   > best-effort VoIP connection that sent at 64Kbps in the face of a
   > persistent packet drop rate of 40% would be unacceptable, and should
   > explicitly *NOT* be sanctioned by IETF standards.  Even if it is
   > the case that there is never any congestion in Northern hemisphere
   > Internet backbones and transoceanic links.  And even if there in
   > fact is a "SHOULD" somewhere in one of the RTP drafts about congestion
   > control.
   > 
   > - Sally
   > (writing as in individual)
   > http://www.icir.org/floyd/
   > 




> It is my understanding that work is in-progress on this topic in 
> various IETF Working Groups, so it is likely that VoIP application 
> protocols will be  implementing and deploying VJ-compatible congestion 
> avoidance in future. 

To be precise..It is still in research stage. As of now, We have TFRC for
applications using fixed packet size(ietf-tsvwg-tfrc-05). As per colin,
Voice codecs generally sent packets at constant duration and hence current
TRFC proposal will not be applicable. thats the reason why colin was
skeptical of using dccp for VoIP in ieprep list.  But, as per sally floyd, 
their is a possible submission on TRFR-PS for RTP applications. More details 
from sally floyd's reply in the link mentioned above.
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 13:00:54 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA19110
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 13:00:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHI3RU26110
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 13:03:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHI3Rv26107
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 13:03:27 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA19103
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 13:00:23 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHI21v26037;
	Tue, 17 Dec 2002 13:02:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHI1Sv26009
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 13:01:28 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA19003
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 12:58:25 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Tue, 17 Dec 2002 13:00:44 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121713010826797
 ; Tue, 17 Dec 2002 13:01:08 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <YW14LXH5>; Tue, 17 Dec 2002 13:00:54 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77AAD@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: "Ieprep (E-mail)" <ieprep@ietf.org>, "'floyd@icir.org'" <floyd@icir.org>
Subject: RE: [Ieprep] Summary of ieprep work and VoIP--Take 2
Date: Tue, 17 Dec 2002 13:01:00 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

> On Tuesday, Dec 17, 2002, at 11:49 America/Montreal, King, 
> Kimberly S. 
> wrote:
> > For example, enough VoIP on a congested link (e.g., 10-40% 
> packet drop 
> > rates)
> > can be sufficient to make the link essentially unusable for TCP 
> > traffic.
> 
> I think that overstates the case. 

I'll be clearer.  From an offline conversation I had with Sally Floyd, 
she doesn't think TCP connections are going to be doing very well 
with 10-40% packet drop rates over a congested link.  She expressed
concern that without end-to-end congestion control for VoIP that
somewhere there is going to be a congested link and that VoIP
traffic will contribute significantly to that congestion resulting in 
a link completely unusable for TCP traffic.

I don't think I misunderstood and I hope I haven't
overstated but I've cc'd her just in case.

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 13:48:55 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA20098
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 13:48:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHIpS928958
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 13:51:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHIpSv28955
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 13:51:28 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA20078
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 13:48:24 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHIo5v28867;
	Tue, 17 Dec 2002 13:50:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHInAv28835
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 13:49:10 -0500
Received: from kc-msxproto2.kc.umkc.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA20034
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 13:46:05 -0500 (EST)
Received: from KC-MAIL2.kc.umkc.edu ([134.193.143.162] RDNS failed) by kc-msxproto2.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.5329);
	 Tue, 17 Dec 2002 12:49:04 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: RE: [Ieprep] Understanding Congestion
Date: Tue, 17 Dec 2002 12:49:03 -0600
Message-ID: <A29143A40AEAE3459FF829D1DD43316101D9B0A6@KC-MAIL2.kc.umkc.edu>
Thread-Topic: [Ieprep] Understanding Congestion
Thread-Index: AcKl4aLT8v9W7M6UTlWi2xTgUnIYBQAEF+Mg
From: "Ayyasamy, Senthilkumar  (UMKC-Student)" <saq66@umkc.edu>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>,
        "Bill Woodcock" <woody@pch.net>
Cc: <ieprep@ietf.org>
X-OriginalArrivalTime: 17 Dec 2002 18:49:04.0638 (UTC) FILETIME=[F8EC69E0:01C2A5FC]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBHInAv28836
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


> >Kimberly King quotes Steve Bellovin:
> >> Congestion management in the Internet is problematic, 
> >> in part, because no capabilities exist for managing traffic
> >>  associated with specific users,
> >> connections, sources, or destinations, and it would be 
> >> difficult to implement such capabilities.
> > 
> > While the former was true at the time Steve wrote this in 
> > 1998, it is no longer so.  The latter, that it was 
> > difficult to implement, was demonstrated to be true.  :-)
> > 
> > The fix was the combination of uRPF, real-time backscatter 
> > analysis, and
> > the use of iBGP to distribute blackhole sources/destinations 
> > within the
> > backbone.
> > 
> > The limitation is the same as that of any kind of tagging: 
> > you can't trust
> > such information entering from outside the AS, since it's a 
> DoS attack
> > rather than valid.

> prohibiting denial of service attacks that
> are based on forged source addresses and doing it in a
> scaleable accurate way against address aggregates
> is a different problem 

congestion management of high bandwidth flows and aggregate 
based traffic is a challenging problem to be tackled.
RED-PD was suggested as a solution for aggregate based traffic.
You have to see Steve's work on RED-PD along with ratul
and sally floyd. 


 

> than congestion management of
> individual flows or user sessions. 
> The case that has been worked on is illegitimate bulk
> traffic at layer 3... legitimate traffic from customers
> at the session layer seems to be a completely different
> problem to me...

A possible way for congestion management of a legitimate  
traffic from customer can by maintaining is using a
combination of techniques mentioned by Bill.
Doesnt RFC 2914 answer such issues? Section 3.2 talks about 
fairness. Also, ecm by use of Congestion managers (RFC 3124) can 
also be helpful  if your so pariticular about specific user or 
application traffic.

But, I dont think this is a major problem. Their are many other
problems like how effectively we do congestion control for 
high bandwidth flows and the associated fairness issues.






_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 13:50:56 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA20139
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 13:50:55 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHIrTt29046
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 13:53:29 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHIrSv29043
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 13:53:28 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA20131
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 13:50:24 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHIq7v29006;
	Tue, 17 Dec 2002 13:52:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHIpdv28978
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 13:51:39 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA20095
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 13:48:35 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 7E78B67103; Tue, 17 Dec 2002 13:52:33 -0500 (EST)
Date: Tue, 17 Dec 2002 13:51:33 -0500
Subject: Re: [Ieprep] Summary of ieprep work and VoIP--Take 2
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "Ieprep (E-mail)" <ieprep@ietf.org>, "'floyd@icir.org'" <floyd@icir.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77AAD@mcl-its-exs01.mail.saic.com>
Message-Id: <8FE8861D-11F0-11D7-80DC-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 17, 2002, at 13:01 America/Montreal, King, Kimberly S. 
wrote:
> I'll be clearer.  From an offline conversation I had with Sally Floyd,
> she doesn't think TCP connections are going to be doing very well
> with 10-40% packet drop rates over a congested link.  She expressed
> concern that without end-to-end congestion control for VoIP that
> somewhere there is going to be a congested link and that VoIP
> traffic will contribute significantly to that congestion resulting in
> a link completely unusable for TCP traffic.

We need to give Sally a chance to clarify, I think.

Things we seem to agree on:
- Letting non-congestion-avoiding traffic crowd out congestion-avoiding
   traffic is a very bad thing from an architectural or system 
engineering
   perspective.
- VoIP or any other application that doesn't include 
congestion-avoidance
   needs to be modified to implement and use congestion-avoidance.

We seem to disagree that the parameters you outline will result in "a 
link
completely unusable for TCP traffic".  In particular, my operational
experience says that 10+ % drop rates are annoying, but that TCP-based
applications (e.g. use of Telnet to read email over a 2.4 Kbps shared 
WAN
radio link) are still usable.  Now I'd prefer to have more bandwidth
in all cases, but there is a different between "completely unusable" and
"not as high performance as one might prefer".

Regards,

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 14:17:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA20834
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 14:17:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHJJi430540
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 14:19:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHJJiv30537
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 14:19:44 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA20821
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 14:16:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHJIEv30405;
	Tue, 17 Dec 2002 14:18:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHJCgv30264
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 14:12:42 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA20588
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 14:09:36 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Tue, 17 Dec 2002 14:11:49 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121714120112884
 ; Tue, 17 Dec 2002 14:12:01 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <ZBBRFSCL>; Tue, 17 Dec 2002 14:13:36 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77AAE@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>,
        "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: "Ieprep (E-mail)" <ieprep@ietf.org>, "'floyd@icir.org'" <floyd@icir.org>
Subject: RE: [Ieprep] Summary of ieprep work and VoIP--Take 2
Date: Tue, 17 Dec 2002 14:11:53 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

> We seem to disagree that the parameters you outline will result in "a 
> link
> completely unusable for TCP traffic".  In particular, my operational
> experience says that 10+ % drop rates are annoying, but that TCP-based
> applications (e.g. use of Telnet to read email over a 2.4 Kbps shared 
> WAN
> radio link) are still usable.  Now I'd prefer to have more bandwidth
> in all cases, but there is a different between "completely 
> unusable" and
> "not as high performance as one might prefer".

If the "you" you are addressing is me then read my words 
again.  

"If other non-VoIP traffic is also carried, then there is a 
tradeoff to make between the performance of voice and non-voice 
applications.  For example, enough VoIP on a congested link 
(e.g., 10-40% packet drop rates) can be sufficient to make the
link essentially unusable for TCP traffic. "

To say "completely unusable" means it will not work. 
"Essentially unusable" affords a more subjective 
view of whether it works or not. In any event, we aren't 
discussing a potential RFC here -- I've written
a summary email message. :-)

Kimberly

 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 14:20:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA20875
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 14:20:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHJMht30702
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 14:22:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHJMhv30699
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 14:22:43 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA20870
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 14:19:38 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHJL2v30656;
	Tue, 17 Dec 2002 14:21:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHJKgv30620
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 14:20:42 -0500
Received: from imo-d10.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA20854
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 14:17:37 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d10.mx.aol.com (mail_out_v34.13.) id i.198.126ebefd (4446);
	Tue, 17 Dec 2002 14:20:29 -0500 (EST)
Message-ID: <198.126ebefd.2b30d2fd@aol.com>
Date: Tue, 17 Dec 2002 14:20:29 EST
Subject: Re: [Ieprep] updated requirements documents (general)
To: K.Carlberg@cs.ucl.ac.uk, rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: multipart/alternative; boundary="part1_198.126ebefd.2b30d2fd_boundary"
X-Mailer: AOL 6.0 for Windows XP US sub 51
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


--part1_198.126ebefd.2b30d2fd_boundary
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

In a message dated 12/13/2002 8:13:24 AM Eastern Standard Time, 
K.Carlberg@cs.ucl.ac.uk writes:

[KC] As for your presentation in Atlanta, I have to apologize that while i
was present, I was engaged in another conversation.  It was my
impression that your presentation was specifically scoped at MLPP and
not ETS as a whole, which may explain in part why there were no
objections.  And finally, working group consensus is predominantly
reached on the list.  I didn't hear a call for humms when you spoke :-)

[MAP] No, my presentation was saying that the requirement for accounting 
(auditing) at the application layer (not the packet transport layer) is a 
common requirement for all "priority" services. That is, any capability that 
gives the user the ability to get a better service, at the possible 
disadvantage of others, requires auditing to detect misuse. This inlcludes 
ETS, MLPP, and a couple that are out-of-scope of IEPREP such as 911/112.

I believe the objections that have been raised in the past were mostly 
related to beliefs that I was advocating counting packets, a notion which I 
hope I dispelled once and for all in my presentation. Also, I believe it has 
been clearly stated that IEPREP listing something as a "requirement" does not 
in any way mandate that any specific vendor or carrier must provide it. Given 
this, I would like to hear objections to identifying "Accounting" as a 
requirement that must be included in any specification for providing 
"priority services", that is, objections to the specific text I've proposed 
below.


In a message dated 12/13/2002 8:26:03 AM Eastern Standard Time,  
rja@extremenetworks.com writes:


> [MAP] 1) Accounting: Following my presentation in Atlanta at which no 
> objection was raised,

[RJA] Disagree.  There were several objections from operators during the 
meeting.  I was
in the room. 

[MAP] Kimberly's notes of the IEPREP meeting, distributed Nov 22, don't make 
any mention of any objection to what I was presenting and I didn't hear any. 
(The notes contain a reference to "Peter Lothberg chimed in on the 
unreasonableness of packet level accounting", in spite of the fact that I 
began my presentation with a clear statement that "I am not proposing packet 
level accounting.")


 [RJA] Moreover, the list discussion takes precedence over in-person meetings
under IETF process rules (unlike with some other standards groups).  And 
there are
numerous unresolved objections here on the list -- mainly because you haven't 
made
it clear what exactly you want to account for.

[MAP] See my first comment above.


[RJA] It would help further discussion a great deal if you could be more 
clear and very
specific in what objects you want.  Depending on what they are, folks here 
might
well agree with the specifics.

[MAP] In my presentation I listed some accounting objects that "are most 
common". I don't believe it is appropriate to discuss (or agree on) specific 
items at this point. I see that as a later function of SIPPING (if really 
part of SIP, which I presume it is) and is clearly application dependent. I 
certainly think that this list provides enough specificity to allow anyone to 
raise concrete objections to Accounting that are relevant to the issue at 
hand. I'm sorry, but I don't believe I have heard any objections which were 
addressing the subject at hand. There have been several about packet level 
accounting (which I never proposed) and some claim that no one would need to 
do accounting since "everyone" would charge flat-rate anyway (I'm not 
addressing the needs for billing), and I think also some claims that the 
network wouldn't care about misuse if has unlimited capacity (hardly worth 
commenting on).


To repeat that page of my presentation for those who missed it:

--------------------------------------------------------
What may be useful?

For each call setup or attempt, the following are most common:
- Calling and called party identification
- Time and duration of call attempt
- Priority level or special treatment requested and received
- Password or access code used *
- Disposition of call attempt

* It is recognized that security policies would normally require 
that the actual password or access code should not be stored in 
the accounting record, but there must be some way to determine, 
when examining the records, which code was (mis)used. For example, 
when a simple 4-digit pin is entered, which is sent through the 
audio path "in the clear" anyway, then storing the actual code 
received is not a further security problem.
---------------------------------------------------------

[MAP] To repeat my proposal:

> this should be moved to the Requirements section. I suggest the 
> following text, which I believe correctly reflects the need and avoids 
> any bad implications:
>
> 9) Accounting (auditing)
>
> Accounting represents a method of tracking actual usage of a service 
> for purposes of capacity and trend analysis, cost allocation, 
> auditing, and billing [RFC2975]. We assume that the usage of any 
> service better than best effort will be tracked and may subsequently 
> be billed to the user. Accounting for billing purposes is not 
> addressed as a general requirement for ETS. However, solutions used to 
> realize ETS should include the capability of an accounting mechanism 
> to support auditing as defined in [RFC2975].

[RJA] Disagree.  I'd be happy to add text saying that a solution "must not 
preclude
an operator from performing accounting", but not text making it a requirement.

[MAP] I never suggested any text saying that it is a requirement for an 
operator to perform accounting. The proposed text says "solutions ... should 
include the capability ... to support auditing." Anyway, the text in 2.1 of 
the draft pretty well dispells any idea that anything in this document places 
actual requirements on a service provider.


Mike Pierce
Artel


--part1_198.126ebefd.2b30d2fd_boundary
Content-Type: text/html; charset="US-ASCII"
Content-Transfer-Encoding: 7bit

<HTML><FONT FACE=arial,helvetica><FONT  SIZE=2>In a message dated 12/13/2002 8:13:24 AM Eastern Standard Time, K.Carlberg@cs.ucl.ac.uk writes:
<BR>
<BR>[KC] As for your presentation in Atlanta, I have to apologize that while i
<BR>was present, I was engaged in another conversation. &nbsp;It was my
<BR>impression that your presentation was specifically scoped at MLPP and
<BR>not ETS as a whole, which may explain in part why there were no
<BR>objections. &nbsp;And finally, working group consensus is predominantly
<BR>reached on the list. &nbsp;I didn't hear a call for humms when you spoke :-)
<BR>
<BR>[MAP] No, my presentation was saying that the requirement for accounting (auditing) at the application layer (not the packet transport layer) is a common requirement for all "priority" services. That is, any capability that gives the user the ability to get a better service, at the possible disadvantage of others, requires auditing to detect misuse. This inlcludes ETS, MLPP, and a couple that are out-of-scope of IEPREP such as 911/112.
<BR>
<BR>I believe the objections that have been raised in the past were mostly related to beliefs that I was advocating counting packets, a notion which I hope I dispelled once and for all in my presentation. Also, I believe it has been clearly stated that IEPREP listing something as a "requirement" does not in any way mandate that any specific vendor or carrier must provide it. Given this, I would like to hear objections to identifying "Accounting" as a requirement that must be included in any specification for providing "priority services", that is, objections to the specific text I've proposed below.
<BR>
<BR>
<BR>In a message dated 12/13/2002 8:26:03 AM Eastern Standard Time, &nbsp;rja@extremenetworks.com writes:
<BR>
<BR>
<BR>&gt; [MAP] 1) Accounting: Following my presentation in Atlanta at which no 
<BR>&gt; objection was raised,
<BR>
<BR>[RJA] Disagree. &nbsp;There were several objections from operators during the meeting. &nbsp;I was
<BR>in the room. 
<BR>
<BR>[MAP] Kimberly's notes of the IEPREP meeting, distributed Nov 22, don't make any mention of any objection to what I was presenting and I didn't hear any. (The notes contain a reference to "Peter Lothberg chimed in on the unreasonableness of packet level accounting", in spite of the fact that I began my presentation with a clear statement that "I am not proposing packet level accounting.")
<BR>
<BR>
<BR> [RJA] Moreover, the list discussion takes precedence over in-person meetings
<BR>under IETF process rules (unlike with some other standards groups). &nbsp;And there are
<BR>numerous unresolved objections here on the list -- mainly because you haven't made
<BR>it clear what exactly you want to account for.
<BR>
<BR>[MAP] See my first comment above.
<BR>
<BR>
<BR>[RJA] It would help further discussion a great deal if you could be more clear and very
<BR>specific in what objects you want. &nbsp;Depending on what they are, folks here might
<BR>well agree with the specifics.
<BR>
<BR>[MAP] In my presentation I listed some accounting objects that "are most common". I don't believe it is appropriate to discuss (or agree on) specific items at this point. I see that as a later function of SIPPING (if really part of SIP, which I presume it is) and is clearly application dependent. I certainly think that this list provides enough specificity to allow anyone to raise concrete objections to Accounting that are relevant to the issue at hand. I'm sorry, but I don't believe I have heard any objections which were addressing the subject at hand. There have been several about packet level accounting (which I never proposed) and some claim that no one would need to do accounting since "everyone" would charge flat-rate anyway (I'm not addressing the needs for billing), and I think also some claims that the network wouldn't care about misuse if has unlimited capacity (hardly worth commenting on).
<BR>
<BR>
<BR>To repeat that page of my presentation for those who missed it:
<BR>
<BR>--------------------------------------------------------
<BR>What may be useful?
<BR>
<BR>For each call setup or attempt, the following are most common:
<BR>- Calling and called party identification
<BR>- Time and duration of call attempt
<BR>- Priority level or special treatment requested and received
<BR>- Password or access code used *
<BR>- Disposition of call attempt
<BR>
<BR>* It is recognized that security policies would normally require 
<BR>that the actual password or access code should not be stored in 
<BR>the accounting record, but there must be some way to determine, 
<BR>when examining the records, which code was (mis)used. For example, 
<BR>when a simple 4-digit pin is entered, which is sent through the 
<BR>audio path "in the clear" anyway, then storing the actual code 
<BR>received is not a further security problem.
<BR>---------------------------------------------------------
<BR>
<BR>[MAP] To repeat my proposal:
<BR>
<BR>&gt; this should be moved to the Requirements section. I suggest the 
<BR>&gt; following text, which I believe correctly reflects the need and avoids 
<BR>&gt; any bad implications:
<BR>&gt;
<BR>&gt; 9) Accounting (auditing)
<BR>&gt;
<BR>&gt; Accounting represents a method of tracking actual usage of a service 
<BR>&gt; for purposes of capacity and trend analysis, cost allocation, 
<BR>&gt; auditing, and billing [RFC2975]. We assume that the usage of any 
<BR>&gt; service better than best effort will be tracked and may subsequently 
<BR>&gt; be billed to the user. Accounting for billing purposes is not 
<BR>&gt; addressed as a general requirement for ETS. However, solutions used to 
<BR>&gt; realize ETS should include the capability of an accounting mechanism 
<BR>&gt; to support auditing as defined in [RFC2975].
<BR>
<BR>[RJA] Disagree. &nbsp;I'd be happy to add text saying that a solution "must not preclude
<BR>an operator from performing accounting", but not text making it a requirement.
<BR>
<BR>[MAP] I never suggested any text saying that it is a requirement for an operator to perform accounting. The proposed text says "solutions ... should include the capability ... to support auditing." Anyway, the text in 2.1 of the draft pretty well dispells any idea that anything in this document places actual requirements on a service provider.
<BR>
<BR>
<BR>Mike Pierce
<BR>Artel
<BR></FONT></HTML>

--part1_198.126ebefd.2b30d2fd_boundary--
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 15:45:19 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA23194
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 15:45:19 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHKlsq03102
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 15:47:54 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHKlsv03099
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 15:47:54 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA23188
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 15:44:48 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHKk7v03033;
	Tue, 17 Dec 2002 15:46:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHKjjv03011
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 15:45:45 -0500
Received: from bells.cs.ucl.ac.uk (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id PAA23167
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 15:42:39 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.23058-0@bells.cs.ucl.ac.uk>; Tue, 17 Dec 2002 20:45:22 +0000
To: Mpierce1@aol.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] updated requirements documents (general)
In-reply-to: Your message of "Tue, 17 Dec 2002 14:20:29 EST." <198.126ebefd.2b30d2fd@aol.com>
Date: Tue, 17 Dec 2002 20:45:20 +0000
Message-ID: <25185.1040157920@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


> [MAP] No, my presentation was saying that the requirement for accounting
> (auditing) at the application layer (not the packet transport layer) is
> a common requirement for all "priority" services. That is, any

to be precise, neither the General Requirements nor the Telephony
requirements document use the term '"priority" services'.  as you will
recall the term "priority" has caused a flurry of arguements on this
list, the previous list, and at meetings.  so if you are speaking
in terms of how the Pierce-draft(s) describe "priority", then the
above subject line you have used is incorrect and we are speaking 
past each other.

> capability that gives the user the ability to get a better service, at
> the possible disadvantage of others, requires auditing to detect
> misuse. This inlcludes ETS, MLPP, and a couple that are out-of-scope of
> IEPREP such as 911/112.
>
> I believe the objections that have been raised in the past were mostly
> related to beliefs that I was advocating counting packets, a notion
> which I hope I dispelled once and for all in my presentation. Also, I
> believe it has been clearly stated that IEPREP listing something as a
> "requirement" does not in any way mandate that any specific vendor or
> carrier must provide it. Given this, I would like to hear objections to
> identifying "Accounting" as a requirement that must be included in any
> specification for providing "priority services", that is, objections to
> the specific text I've proposed below.

One of Ran's objections stated that IAM doesn't not have a requirement
for accounting.  This in line with a conversation I had with one of
the IAM designers a while back.  This by itself means that there is a
type of application in which an accounting requirement is simply
out of scope.

the General requirements draft is meant to be minimal.  If there is a
need to go beyond it, one produces a draft that addresses a more 
specific case.  an example of this was done with the telephony
requiremenst draft.  in this case a statement about accounting was
placed in that draft's requirements section.

if you wish to continue this thread, please provide a response to the 
following:

1. address Ran's objection above.  by adding a requirement that
articulates a statement about accounting, how do you deal with an
application that has no concept about accounting?

2. provide a reason why the current structure of the two carlberg-ets-*
requirements (general followed by more specific) does not work.

3) why doesn't a statement about an accouting requirement in the
telephony requiremenst draft satisfy your perceived needs as they
relate to MLPP?  Is there another case that you are contemplating?
if so, what is it?

-ken
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 16:06:43 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA23760
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 16:06:43 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHL9IC04634
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 16:09:18 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHL9Iv04631
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 16:09:18 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA23738
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 16:06:12 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHL7Pv04433;
	Tue, 17 Dec 2002 16:07:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHL2Nv03637
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 16:02:23 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA23520
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 15:59:17 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 0F65B67103; Tue, 17 Dec 2002 16:03:17 -0500 (EST)
Date: Tue, 17 Dec 2002 16:02:16 -0500
Subject: Re: [Ieprep] Summary of ieprep work and VoIP--Take 2
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "Ieprep (E-mail)" <ieprep@ietf.org>, "'floyd@icir.org'" <floyd@icir.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77AAE@mcl-its-exs01.mail.saic.com>
Message-Id: <D28A447C-1202-11D7-80DC-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Tuesday, Dec 17, 2002, at 14:11 America/Montreal, King, Kimberly S. 
wrote:
> "If other non-VoIP traffic is also carried, then there is a
> tradeoff to make between the performance of voice and non-voice
> applications.  For example, enough VoIP on a congested link
> (e.g., 10-40% packet drop rates) can be sufficient to make the
> link essentially unusable for TCP traffic. "
>
> To say "completely unusable" means it will not work.

The phrase "completely unusable" was used in your second note.

> "Essentially unusable" affords a more subjective
> view of whether it works or not.

I was able to read email just fine on a congested 2.4 Kbps
radio link, which seems significantly at variance from
"essentially unusable".  So I don't think that claim is quite
right.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 17 16:29:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA24371
	for <ieprep-archive@odin.ietf.org>; Tue, 17 Dec 2002 16:29:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBHLVkm05734
	for ieprep-archive@odin.ietf.org; Tue, 17 Dec 2002 16:31:46 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHLVkv05731
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 17 Dec 2002 16:31:46 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA24367
	for <ieprep-web-archive@ietf.org>; Tue, 17 Dec 2002 16:28:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHLU3v05690;
	Tue, 17 Dec 2002 16:30:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHLTNv05642
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 16:29:23 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA24311
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 16:26:16 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBHLKwo07829;
	Tue, 17 Dec 2002 13:20:58 -0800 (PST)
Date: Tue, 17 Dec 2002 13:20:58 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77AA4@mcl-its-exs01.mail.saic.com>
Message-ID: <Pine.GSO.4.44.0212171319290.7471-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Tue, 17 Dec 2002, King, Kimberly  S. wrote:
    > prohibiting denial of service attacks that
    > are based on forged source addresses and doing it in a
    > scaleable accurate way against address aggregates
    > is a different problem than congestion management of
    > individual flows or user sessions.

Right...  The combination of uRPF, backscatter, and iBGP is used to sink
traffic to a destination /32, regardless of source address.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 18 09:35:20 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA09556
	for <ieprep-archive@odin.ietf.org>; Wed, 18 Dec 2002 09:35:19 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBIEbqJ08732
	for ieprep-archive@odin.ietf.org; Wed, 18 Dec 2002 09:37:52 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIEbqv08729
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 18 Dec 2002 09:37:52 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA09511
	for <ieprep-web-archive@ietf.org>; Wed, 18 Dec 2002 09:34:48 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIEaEv08053;
	Wed, 18 Dec 2002 09:36:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIEZgv08037
	for <ieprep@optimus.ietf.org>; Wed, 18 Dec 2002 09:35:42 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA09383
	for <ieprep@ietf.org>; Wed, 18 Dec 2002 09:32:38 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Wed, 18 Dec 2002 09:35:01 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002121809352500971
 ; Wed, 18 Dec 2002 09:35:25 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <ZBBRF74S>; Wed, 18 Dec 2002 09:37:01 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77AB3@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Bill Woodcock'" <woody@pch.net>,
        "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
Date: Wed, 18 Dec 2002 09:35:24 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Bill said: 
> Right...  The combination of uRPF, backscatter, and iBGP is 
> used to sink
> traffic to a destination /32, regardless of source address.

Firewalls...how many user sessions are behind a single /32? 
Potentially hundreds right?  Some could be emergency 
while others are not.  

Your reply suggests that you refute my claim that this 
addresses a different problem that congestion management of 
individual flows or user sessions over the global Internet.  
You have provided a solution.  What ieprep problem does 
this solve?  


Kimberly






Disclaimer:  My statements represent my own thinking and do not necessarily 
reflect the position of anyone else in the world including my employer. 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 18 10:34:22 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10747
	for <ieprep-archive@odin.ietf.org>; Wed, 18 Dec 2002 10:34:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBIFatx11327
	for ieprep-archive@odin.ietf.org; Wed, 18 Dec 2002 10:36:55 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIFatv11324
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 18 Dec 2002 10:36:55 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10735
	for <ieprep-web-archive@ietf.org>; Wed, 18 Dec 2002 10:33:50 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIFZ8v11273;
	Wed, 18 Dec 2002 10:35:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIFYmv11246
	for <ieprep@optimus.ietf.org>; Wed, 18 Dec 2002 10:34:48 -0500
Received: from mail1.cisco.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10697
	for <ieprep@ietf.org>; Wed, 18 Dec 2002 10:31:43 -0500 (EST)
Received: from BGREENEXP1 (dhcp-10-24-19-68.cisco.com [10.24.19.68]) by mail1.cisco.com (8.8.6 (PHNE_14041)/CISCO.SERVER.1.2) with ESMTP id HAA16526; Wed, 18 Dec 2002 07:34:42 -0800 (PST)
Reply-To: <bgreene@cisco.com>
From: "Barry Raveendran Greene" <bgreene@cisco.com>
To: "'Nguyen, An'" <nguyena@ncs.gov>, "'Bill Woodcock'" <woody@pch.net>,
        "'King, Kimberly  S.'" <KIMBERLY.S.KING@saic.com>
Cc: <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
Date: Wed, 18 Dec 2002 07:34:18 -0800
Organization: Cisco Systems
Message-ID: <00e801c2a6aa$ee6542e0$4413180a@cisco.com>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="US-ASCII"
Content-Transfer-Encoding: 7bit
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook, Build 10.0.4024
In-Reply-To: <7F18415E4D63CB45BB9B3A591F68D12D02EF4C6A@emshqs1.ncr.disa.mil>
X-MimeOLE: Produced By Microsoft MimeOLE V6.00.2800.1106
Importance: Normal
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


Looks like people need to come up to speed on the realities of uRPF,
Remote Triggered Black Hole Filtering, Sink Holes, and a bunch of other
techniques we're using today.

An - your understanding of uRPF is the FUD understanding - not the
reality. I'm working on an new "uRPF Deployment Experience" whitepaper.
But until then, check out the ISP Security tutorial at:

	ftp://ftp-eng.cisco.com/cons/isp/security/

Barry

> -----Original Message-----
> From: ieprep-admin@ietf.org [mailto:ieprep-admin@ietf.org] On Behalf
Of
> Nguyen, An
> Sent: Wednesday, December 18, 2002 7:23 AM
> To: 'Bill Woodcock'; King, Kimberly S.
> Cc: 'ieprep@ietf.org'
> Subject: RE: [Ieprep] Understanding Congestion
> 
> Bill,
> 
> >-----Original Message-----
> >From: Bill Woodcock [mailto:woody@pch.net]
> >Sent: Tuesday, December 17, 2002 4:21 PM
> >To: King, Kimberly S.
> >Cc: 'ieprep@ietf.org'
> >Subject: RE: [Ieprep] Understanding Congestion
> >
> >
> >      On Tue, 17 Dec 2002, King, Kimberly  S. wrote:
> >    > prohibiting denial of service attacks that
> >    > are based on forged source addresses and doing it in a
> >    > scaleable accurate way against address aggregates
> >    > is a different problem than congestion management of
> >    > individual flows or user sessions.
> >
> >Right...  The combination of uRPF, backscatter, and iBGP is used to
sink
> >traffic to a destination /32, regardless of source address.
> >
> 
> My understanding of Unicast Reverse Path Forwarding is that uRPF
provides
> a
> way to do a trace back to make sure the source address is not spoofed
> according to ACL's and route map's. This tool is only good for DoS,
not
> Distributed DoS in my opinion. It has some drawbacks, though:  1. It
does
> not work well on interfaces internal to the network because of its
routing
> asymmetry; and 2. It does not work well on multihomed environments.
> 
> By the way, sinking too much traffic would cause a DoS as well because
you
> spend so much CPU resource on discarding suspicious packets instead of
> forwarding packets.
> 
> I have a question for you: if we have a DDoS of a magnitude , for the
lack
> of better word, of (NIMDA + CODE RED)^n, where n is a large number,
would
> the currently configured IP backbone be ready to withstand it?
> 
> An
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 18 10:34:41 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10764
	for <ieprep-archive@odin.ietf.org>; Wed, 18 Dec 2002 10:34:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBIFbEb11698
	for ieprep-archive@odin.ietf.org; Wed, 18 Dec 2002 10:37:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIFbEv11687
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 18 Dec 2002 10:37:14 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA10742
	for <ieprep-web-archive@ietf.org>; Wed, 18 Dec 2002 10:34:10 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIFO8v10923;
	Wed, 18 Dec 2002 10:24:08 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIFNWv10871
	for <ieprep@optimus.ietf.org>; Wed, 18 Dec 2002 10:23:32 -0500
Received: from [209.22.88.17] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id KAA10435
	for <ieprep@ietf.org>; Wed, 18 Dec 2002 10:20:27 -0500 (EST)
Received: from mtassp3.ncr.disa.mil by [209.22.88.17]
          via smtpd (for ietf-mx.ietf.org [132.151.6.1]) with SMTP; 18 Dec 2002 15:16:08 UT
Received: by mtassp3.ncr.disa.mil with Internet Mail Service (5.5.2653.19)
	id <Z1KYL94M>; Wed, 18 Dec 2002 10:23:30 -0500
Message-ID: <7F18415E4D63CB45BB9B3A591F68D12D02EF4C6A@emshqs1.ncr.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'Bill Woodcock'" <woody@pch.net>,
        "King, Kimberly  S."
	 <KIMBERLY.S.KING@saic.com>
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
Date: Wed, 18 Dec 2002 10:23:24 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Bill,

>-----Original Message-----
>From: Bill Woodcock [mailto:woody@pch.net]
>Sent: Tuesday, December 17, 2002 4:21 PM
>To: King, Kimberly S.
>Cc: 'ieprep@ietf.org'
>Subject: RE: [Ieprep] Understanding Congestion
>
>
>      On Tue, 17 Dec 2002, King, Kimberly  S. wrote:
>    > prohibiting denial of service attacks that
>    > are based on forged source addresses and doing it in a
>    > scaleable accurate way against address aggregates
>    > is a different problem than congestion management of
>    > individual flows or user sessions.
>
>Right...  The combination of uRPF, backscatter, and iBGP is used to sink
>traffic to a destination /32, regardless of source address.
>

My understanding of Unicast Reverse Path Forwarding is that uRPF provides a
way to do a trace back to make sure the source address is not spoofed
according to ACL's and route map's. This tool is only good for DoS, not
Distributed DoS in my opinion. It has some drawbacks, though:  1. It does
not work well on interfaces internal to the network because of its routing
asymmetry; and 2. It does not work well on multihomed environments.

By the way, sinking too much traffic would cause a DoS as well because you
spend so much CPU resource on discarding suspicious packets instead of
forwarding packets.  

I have a question for you: if we have a DDoS of a magnitude , for the lack
of better word, of (NIMDA + CODE RED)^n, where n is a large number, would
the currently configured IP backbone be ready to withstand it?

An

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 18 11:58:08 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA12684
	for <ieprep-archive@odin.ietf.org>; Wed, 18 Dec 2002 11:58:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBIH0fd16542
	for ieprep-archive@odin.ietf.org; Wed, 18 Dec 2002 12:00:41 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIH0fv16539
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 18 Dec 2002 12:00:41 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA12675
	for <ieprep-web-archive@ietf.org>; Wed, 18 Dec 2002 11:57:37 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIGxAv16400;
	Wed, 18 Dec 2002 11:59:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIGwjv16306
	for <ieprep@optimus.ietf.org>; Wed, 18 Dec 2002 11:58:45 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA12629
	for <ieprep@ietf.org>; Wed, 18 Dec 2002 11:55:41 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBIGoTo14145;
	Wed, 18 Dec 2002 08:50:29 -0800 (PST)
Date: Wed, 18 Dec 2002 08:50:29 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "Nguyen, An" <nguyena@ncs.gov>
cc: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>,
        "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
In-Reply-To: <7F18415E4D63CB45BB9B3A591F68D12D02EF4C6A@emshqs1.ncr.disa.mil>
Message-ID: <Pine.GSO.4.44.0212180849460.14120-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Wed, 18 Dec 2002, Nguyen, An wrote:
    > My understanding of Unicast Reverse Path Forwarding is that uRPF provides a
    > way to do a trace back to make sure the source address is not spoofed
    > according to ACL's and route map's.

No, sorry, please either RTFM or pay attention to people's postings.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 18 12:05:10 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA12840
	for <ieprep-archive@odin.ietf.org>; Wed, 18 Dec 2002 12:05:10 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBIH7hl17779
	for ieprep-archive@odin.ietf.org; Wed, 18 Dec 2002 12:07:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIH7gv17776
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 18 Dec 2002 12:07:42 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA12821
	for <ieprep-web-archive@ietf.org>; Wed, 18 Dec 2002 12:04:39 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIH62v17098;
	Wed, 18 Dec 2002 12:06:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIH5Tv17061
	for <ieprep@optimus.ietf.org>; Wed, 18 Dec 2002 12:05:29 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA12792
	for <ieprep@ietf.org>; Wed, 18 Dec 2002 12:02:24 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBIGrWo14156;
	Wed, 18 Dec 2002 08:53:32 -0800 (PST)
Date: Wed, 18 Dec 2002 08:53:32 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77AB3@mcl-its-exs01.mail.saic.com>
Message-ID: <Pine.GSO.4.44.0212180851300.14120-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Wed, 18 Dec 2002, King, Kimberly  S. wrote:
    > Firewalls...how many user sessions are behind a single /32?

That's a NAT function rather than a firewall function, of course.  If
someone is foolish enough to use a NAT, there's little that anyone can do
to help them.

    > Your reply suggests that you refute my claim that this
    > addresses a different problem that congestion management of
    > individual flows or user sessions over the global Internet.

Mmm, you're correct that it does not in fact allow per _user_ or per
_flow_ traffic dumping, but neither of those are properties which are
visible at layer 3.  These are the techniques which are used in the
Internet, at layer 3.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 18 13:53:16 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA15779
	for <ieprep-archive@odin.ietf.org>; Wed, 18 Dec 2002 13:53:16 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBIItoM23888
	for ieprep-archive@odin.ietf.org; Wed, 18 Dec 2002 13:55:50 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIItov23885
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 18 Dec 2002 13:55:50 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA15762
	for <ieprep-web-archive@ietf.org>; Wed, 18 Dec 2002 13:52:45 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIIs7v23648;
	Wed, 18 Dec 2002 13:54:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIIrJv23512
	for <ieprep@optimus.ietf.org>; Wed, 18 Dec 2002 13:53:19 -0500
Received: from [209.22.88.17] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id NAA15649
	for <ieprep@ietf.org>; Wed, 18 Dec 2002 13:50:14 -0500 (EST)
Received: from mtassp3.ncr.disa.mil by [209.22.88.17]
          via smtpd (for ietf-mx.ietf.org [132.151.6.1]) with SMTP; 18 Dec 2002 18:45:53 UT
Received: by mtassp3.ncr.disa.mil with Internet Mail Service (5.5.2653.19)
	id <Z1KYMGWG>; Wed, 18 Dec 2002 13:53:16 -0500
Message-ID: <7F18415E4D63CB45BB9B3A591F68D12D02EF4C6F@emshqs1.ncr.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'Bill Woodcock'" <woody@pch.net>
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
Date: Wed, 18 Dec 2002 13:53:09 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Bill,


>-----Original Message-----
>From: Bill Woodcock [mailto:woody@pch.net]
>Sent: Wednesday, December 18, 2002 11:54 AM
>To: King, Kimberly S.
>Cc: 'ieprep@ietf.org'
>Subject: RE: [Ieprep] Understanding Congestion
>
>
>      On Wed, 18 Dec 2002, King, Kimberly  S. wrote:
>    > Firewalls...how many user sessions are behind a single /32?
>
>That's a NAT function rather than a firewall function, of course.  If
>someone is foolish enough to use a NAT, there's little that anyone can do
>to help them.


NAT comes in two flavors: static and dynamic. If using NAT is not a good
thing to do, what can be used instead for network security?

An

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 18 13:54:03 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA15801
	for <ieprep-archive@odin.ietf.org>; Wed, 18 Dec 2002 13:54:03 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBIIubo23919
	for ieprep-archive@odin.ietf.org; Wed, 18 Dec 2002 13:56:37 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIIubv23916
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 18 Dec 2002 13:56:37 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA15798
	for <ieprep-web-archive@ietf.org>; Wed, 18 Dec 2002 13:53:32 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIIt1v23824;
	Wed, 18 Dec 2002 13:55:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIIsdv23738
	for <ieprep@optimus.ietf.org>; Wed, 18 Dec 2002 13:54:39 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA15716
	for <ieprep@ietf.org>; Wed, 18 Dec 2002 13:51:33 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id EE1A467103; Wed, 18 Dec 2002 13:55:38 -0500 (EST)
Date: Wed, 18 Dec 2002 13:54:28 -0500
Subject: Re: [Ieprep] Understanding Congestion
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
To: "Nguyen, An" <nguyena@ncs.gov>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <7F18415E4D63CB45BB9B3A591F68D12D02EF4C6A@emshqs1.ncr.disa.mil>
Message-Id: <229CEE81-12BA-11D7-AE7D-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Wednesday, Dec 18, 2002, at 10:23 America/Montreal, Nguyen, An wrote:
> My understanding of Unicast Reverse Path Forwarding is that uRPF 
> provides a
> way to do a trace back to make sure the source address is not spoofed
> according to ACL's and route map's. This tool is only good for DoS, not
> Distributed DoS in my opinion. It has some drawbacks, though:  1. It 
> does
> not work well on interfaces internal to the network because of its 
> routing
> asymmetry; and 2. It does not work well on multihomed environments.

uRPF only addresses one issue:		Forged Source IP addresses

uRPF does not help with a DoS attack if the attacker is using a 
legitimate
source IP address (or one that is forged, but forged to a valid host IP 
on a
subnet behind the first point that checking of the source address 
happens).

uRPF does not help with a DDoS attack if the attacker is using 
legitimate
source IP addresses (or forged addresses that are forged to a valid 
host IP
on a subnet behind the first point that checking of the source address 
happens).

On any broadcast shared link media using ARP (e.g. Ethernet), it is 
possible
for one host to "steal" another host's IP address -- because ARP is not
authenticated.  IPv6 ND can be authenticated with AH, but the key 
management
is a nightmare to deploy (as with most forms of key management), so 
IPv6 ND
authentication is rarely enabled.

On many CPU-based routers, turning on uRPF features causes the 
forwarding
performance to drop below line rate, so not all operators enable this 
feature;
even operators who do deploy this feature often do so very selectively,
rather than all over the place.

uRPF also assumes that the local understanding of the BGP topology is 
correct,
legitimate, and that other ASs are only advertising prefixes that they 
are
authorised to advertise.  Right now, there is no way to authenticate 
any of
that.  (This is also one of the reasons that S-BGP is not actually 
deployable
right now, btw.)

> By the way, sinking too much traffic would cause a DoS as well because 
> you
> spend so much CPU resource on discarding suspicious packets instead of
> forwarding packets.

True only if the router uses CPU for discarding packets.  Many routers
(e.g. ours and some of our competitors') implement packet filtering in
an ASIC.  This means that ACLs are free and have zero adverse impact on
the system behaviour.  The tradeoff with an ASIC is that one can't add
new features as quickly as one can add them in software.  If a product
has all the features one needs implemented in dedicated hardware (e.g. 
ASIC),
the product will generally be a better choice than products that don't 
have
such dedicated hardware support.

> I have a question for you: if we have a DDoS of a magnitude , for the 
> lack
> of better word, of (NIMDA + CODE RED)^n, where n is a large number, 
> would
> the currently configured IP backbone be ready to withstand it?

The correctly configured hardware-based core routers had a relatively 
easy time
handling past DDOS attacks.  Correctly configured CPU-based core 
routers had a
a very difficult time handling past DDOS attacks.

Clever operators experimentally verify router behaviour under a wide 
range
of operating conditions (including adverse conditions) before buying 
routers.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 18 15:33:38 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18114
	for <ieprep-archive@odin.ietf.org>; Wed, 18 Dec 2002 15:33:38 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBIKaEI29004
	for ieprep-archive@odin.ietf.org; Wed, 18 Dec 2002 15:36:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIKaEv29001
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 18 Dec 2002 15:36:14 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18109
	for <ieprep-web-archive@ietf.org>; Wed, 18 Dec 2002 15:33:07 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIKYRv28931;
	Wed, 18 Dec 2002 15:34:27 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBIKU5v28781
	for <ieprep@optimus.ietf.org>; Wed, 18 Dec 2002 15:30:05 -0500
Received: from revere.sonusnet.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA18005
	for <ieprep@ietf.org>; Wed, 18 Dec 2002 15:26:58 -0500 (EST)
Received: from sonusmail01.sonusnet.com (sonusmail01.sonusnet.com [10.128.32.75])
	by revere.sonusnet.com (Switch-2.1.0/Switch-2.1.0) with ESMTP id gBIKTYS04309
	for <ieprep@ietf.org>; Wed, 18 Dec 2002 15:29:34 -0500 (EST)
Received: by sonusmail01.sonusnet.com with Internet Mail Service (5.5.2653.19)
	id <Y67ZQ746>; Wed, 18 Dec 2002 15:29:58 -0500
Message-ID: <CAC8ED3E4E27EE4BBCF7030CC74A267D62AC3F@sonusmail01.sonusnet.com>
From: "Anzivino, Anthony" <aanzivino@sonusnet.com>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
Date: Wed, 18 Dec 2002 15:29:58 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] Ieprep - Looking for a colleague...
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Has anyone seen Dennis Berg, Perry Fergus, or Gabor Luka on this forum?
Just when you think you're done with GETS, it comes back in another
instantiation...

Sorry for the disruption....

Tony (x-telcordia GETS PM 1990s)
Anthony (Tony) Anzivino, PMP 
Professional Services - Program Manager 
Sonus Networks 
978-204-9156 (cell) 978-589-8654 (office)
Yahoo IM "anthony_anzivino"                     aanzivino@sonusnet.com 


-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Wednesday, December 18, 2002 1:54 PM
To: Nguyen, An
Cc: 'ieprep@ietf.org'
Subject: Re: [Ieprep] Understanding Congestion



On Wednesday, Dec 18, 2002, at 10:23 America/Montreal, Nguyen, An wrote:
> My understanding of Unicast Reverse Path Forwarding is that uRPF 
> provides a
> way to do a trace back to make sure the source address is not spoofed
> according to ACL's and route map's. This tool is only good for DoS, not
> Distributed DoS in my opinion. It has some drawbacks, though:  1. It 
> does
> not work well on interfaces internal to the network because of its 
> routing
> asymmetry; and 2. It does not work well on multihomed environments.

uRPF only addresses one issue:		Forged Source IP addresses

uRPF does not help with a DoS attack if the attacker is using a 
legitimate
source IP address (or one that is forged, but forged to a valid host IP 
on a
subnet behind the first point that checking of the source address 
happens).

uRPF does not help with a DDoS attack if the attacker is using 
legitimate
source IP addresses (or forged addresses that are forged to a valid 
host IP
on a subnet behind the first point that checking of the source address 
happens).

On any broadcast shared link media using ARP (e.g. Ethernet), it is 
possible
for one host to "steal" another host's IP address -- because ARP is not
authenticated.  IPv6 ND can be authenticated with AH, but the key 
management
is a nightmare to deploy (as with most forms of key management), so 
IPv6 ND
authentication is rarely enabled.

On many CPU-based routers, turning on uRPF features causes the 
forwarding
performance to drop below line rate, so not all operators enable this 
feature;
even operators who do deploy this feature often do so very selectively,
rather than all over the place.

uRPF also assumes that the local understanding of the BGP topology is 
correct,
legitimate, and that other ASs are only advertising prefixes that they 
are
authorised to advertise.  Right now, there is no way to authenticate 
any of
that.  (This is also one of the reasons that S-BGP is not actually 
deployable
right now, btw.)

> By the way, sinking too much traffic would cause a DoS as well because 
> you
> spend so much CPU resource on discarding suspicious packets instead of
> forwarding packets.

True only if the router uses CPU for discarding packets.  Many routers
(e.g. ours and some of our competitors') implement packet filtering in
an ASIC.  This means that ACLs are free and have zero adverse impact on
the system behaviour.  The tradeoff with an ASIC is that one can't add
new features as quickly as one can add them in software.  If a product
has all the features one needs implemented in dedicated hardware (e.g. 
ASIC),
the product will generally be a better choice than products that don't 
have
such dedicated hardware support.

> I have a question for you: if we have a DDoS of a magnitude , for the 
> lack
> of better word, of (NIMDA + CODE RED)^n, where n is a large number, 
> would
> the currently configured IP backbone be ready to withstand it?

The correctly configured hardware-based core routers had a relatively 
easy time
handling past DDOS attacks.  Correctly configured CPU-based core 
routers had a
a very difficult time handling past DDOS attacks.

Clever operators experimentally verify router behaviour under a wide 
range
of operating conditions (including adverse conditions) before buying 
routers.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 18 23:29:18 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA28590
	for <ieprep-archive@odin.ietf.org>; Wed, 18 Dec 2002 23:29:18 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBJ4Vvs20459
	for ieprep-archive@odin.ietf.org; Wed, 18 Dec 2002 23:31:57 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJ4Vvv20456
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 18 Dec 2002 23:31:57 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA28587
	for <ieprep-web-archive@ietf.org>; Wed, 18 Dec 2002 23:28:45 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJ4UFv20419;
	Wed, 18 Dec 2002 23:30:15 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJ4TPv20376
	for <ieprep@optimus.ietf.org>; Wed, 18 Dec 2002 23:29:25 -0500
Received: from imo-d06.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA28568
	for <ieprep@ietf.org>; Wed, 18 Dec 2002 23:26:14 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d06.mx.aol.com (mail_out_v34.13.) id i.49.283999b8 (18251);
	Wed, 18 Dec 2002 23:29:04 -0500 (EST)
Message-ID: <49.283999b8.2b32a510@aol.com>
Date: Wed, 18 Dec 2002 23:29:04 EST
Subject: Re: [Ieprep] updated requirements documents (general)
To: K.Carlberg@cs.ucl.ac.uk
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
X-Mailer: AOL 6.0 for Windows XP US sub 51
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

In a message dated 12/17/2002 3:45:56 PM Eastern Standard Time, 
K.Carlberg@cs.ucl.ac.uk writes:


> [MAP] No, my presentation was saying that the requirement for accounting
> (auditing) at the application layer (not the packet transport layer) is
> a common requirement for all "priority" services. That is, any

[KC] to be precise, neither the General Requirements nor the Telephony
requirements document use the term '"priority" services'.  as you will
recall the term "priority" has caused a flurry of arguements on this
list, the previous list, and at meetings.  so if you are speaking
in terms of how the Pierce-draft(s) describe "priority", then the
above subject line you have used is incorrect and we are speaking 
past each other.


[MAP] I understand this problem with the word "priority", which I used in a 
general sense to include the 4 examples I provided in my presentation. That 
is also why my presentation includes the explicit statement that "priority 
here does not imply any particular mechanism, e.g., priority queuing" and why 
I did not use the word "priority" in the text I proposed for the "general" 
draft. The use of "priority" in the Pierce drafts is its normal English 
meaning, nothing more, the same as in the previously proposed "Resource 
Priority Header". It is not used in the sense of priority queuing, that is, 
transmitting one packet ahead of others.

> [MAP] capability that gives the user the ability to get a better service, at
> the possible disadvantage of others, requires auditing to detect
> misuse. This includes ETS, MLPP, and a couple that are out-of-scope of
> IEPREP such as 911/112.
>
> I believe the objections that have been raised in the past were mostly
> related to beliefs that I was advocating counting packets, a notion
> which I hope I dispelled once and for all in my presentation. Also, I
> believe it has been clearly stated that IEPREP listing something as a
> "requirement" does not in any way mandate that any specific vendor or
> carrier must provide it. Given this, I would like to hear objections to
> identifying "Accounting" as a requirement that must be included in any
> specification for providing "priority services", that is, objections to
> the specific text I've proposed below.

[KC] One of Ran's objections stated that IAM doesn't not have a requirement
for accounting.  This in line with a conversation I had with one of
the IAM designers a while back.  This by itself means that there is a
type of application in which an accounting requirement is simply
out of scope.


[MAP] I presume you meant IAA. I don't consider IAA to be a "priority 
service" in the context which I was talking about requirements. The IAA 
service as described does not contain any capability for one person to get 
better service over another. Besides, we have been concentrating on "call 
setup" characteristics, such as probability of setup, probability of 
disconnect, etc. These do not relate to IAA. If the intention is for the 
"general" draft to only talk about thing that are common to every "ETS" 
service, then the above argument would mean that "Authentication" should not 
be listed as a requirement, either.

If you mean something else by IAM, I'm sorry that I missed it.


[KC] the General requirements draft is meant to be minimal.  If there is a
need to go beyond it, one produces a draft that addresses a more 
specific case.  an example of this was done with the telephony
requiremenst draft.  in this case a statement about accounting was
placed in that draft's requirements section.

if you wish to continue this thread, please provide a response to the 
following:

1. address Ran's objection above.  by adding a requirement that
articulates a statement about accounting, how do you deal with an
application that has no concept about accounting?


[MAP] See comments above. The final sentence of the text that I proposed 
should include a notion of it applying to cases in which one user gets better 
service, such as "However, solutions used to realize ETS capabilities which 
provide one user with better service at the possible disadvantage of others 
should include the capability of an accounting mechanism to support auditing 
as defined in [RFC2975]." A similar qualification should be added to 
"Authentication".


2. provide a reason why the current structure of the two carlberg-ets-*
requirements (general followed by more specific) does not work.



[MAP] It can work fine in this regard if the Accounting requirement is in the 
appropriate one. Sorry that I was concentrating on the "general" draft. See 
below.


3) why doesn't a statement about an accouting requirement in the
telephony requiremenst draft satisfy your perceived needs as they
relate to MLPP?  Is there another case that you are contemplating?
if so, what is it?


[MAP] A statement about a requirement for Accounting in the Telephony draft 
could satisfy my perceived need, if that is in line with the structure of the 
two documents. It pertains to more than "MLPP". However, the current 
telephony draft only contains one paragraph about Accounting, and that is 
related to "Hybrid infrastructures that include and support connectivity to 
the PSTN". This sentence should be changed to not be restricted to the 
"interworking with the PSTN" case. The following is proposed:

   4) Application layer IP telephony capabilities MUST NOT preclude
     the ability to do accounting.

Mike Pierce
Artel
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 19 07:17:16 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA14910
	for <ieprep-archive@odin.ietf.org>; Thu, 19 Dec 2002 07:17:16 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBJCJxI22860
	for ieprep-archive@odin.ietf.org; Thu, 19 Dec 2002 07:19:59 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJCJwv22857
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 19 Dec 2002 07:19:58 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA14905
	for <ieprep-web-archive@ietf.org>; Thu, 19 Dec 2002 07:16:44 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJCICv22806;
	Thu, 19 Dec 2002 07:18:12 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJCH6v22770
	for <ieprep@optimus.ietf.org>; Thu, 19 Dec 2002 07:17:06 -0500
Received: from revere.sonusnet.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA14851
	for <ieprep@ietf.org>; Thu, 19 Dec 2002 07:13:52 -0500 (EST)
Received: from sonusmail01.sonusnet.com (sonusmail01.sonusnet.com [10.128.32.75])
	by revere.sonusnet.com (Switch-2.1.0/Switch-2.1.0) with ESMTP id gBJCGRS19045
	for <ieprep@ietf.org>; Thu, 19 Dec 2002 07:16:28 -0500 (EST)
Received: by sonusmail01.sonusnet.com with Internet Mail Service (5.5.2653.19)
	id <Y67ZQ9FG>; Thu, 19 Dec 2002 07:16:52 -0500
Message-ID: <CAC8ED3E4E27EE4BBCF7030CC74A267D62AC4F@sonusmail01.sonusnet.com>
From: "Anzivino, Anthony" <aanzivino@sonusnet.com>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Ieprep - Looking for a colleague...
Date: Thu, 19 Dec 2002 07:16:51 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Thanks everyone....

found all three thanks to you....

Cheers, Tony

-----Original Message-----
From: Cochran, Shawn P [mailto:Shawn.Cochran@bellsouth.com]
Sent: Wednesday, December 18, 2002 9:17 PM
To: Anzivino, Anthony
Subject: RE: [Ieprep] Ieprep - Looking for a colleague...


I think Gabor left the NCS about a year ago. May be wrong.

-----Original Message-----
From: Anzivino, Anthony [mailto:aanzivino@sonusnet.com]
Sent: Wednesday, December 18, 2002 3:30 PM
To: 'ieprep@ietf.org'
Subject: [Ieprep] Ieprep - Looking for a colleague...


Has anyone seen Dennis Berg, Perry Fergus, or Gabor Luka on this forum?
Just when you think you're done with GETS, it comes back in another
instantiation...

Sorry for the disruption....

Tony (x-telcordia GETS PM 1990s)
Anthony (Tony) Anzivino, PMP 
Professional Services - Program Manager 
Sonus Networks 
978-204-9156 (cell) 978-589-8654 (office)
Yahoo IM "anthony_anzivino"                     aanzivino@sonusnet.com 


-----Original Message-----
From: RJ Atkinson [mailto:rja@extremenetworks.com]
Sent: Wednesday, December 18, 2002 1:54 PM
To: Nguyen, An
Cc: 'ieprep@ietf.org'
Subject: Re: [Ieprep] Understanding Congestion



On Wednesday, Dec 18, 2002, at 10:23 America/Montreal, Nguyen, An wrote:
> My understanding of Unicast Reverse Path Forwarding is that uRPF 
> provides a
> way to do a trace back to make sure the source address is not spoofed
> according to ACL's and route map's. This tool is only good for DoS, not
> Distributed DoS in my opinion. It has some drawbacks, though:  1. It 
> does
> not work well on interfaces internal to the network because of its 
> routing
> asymmetry; and 2. It does not work well on multihomed environments.

uRPF only addresses one issue:		Forged Source IP addresses

uRPF does not help with a DoS attack if the attacker is using a 
legitimate
source IP address (or one that is forged, but forged to a valid host IP 
on a
subnet behind the first point that checking of the source address 
happens).

uRPF does not help with a DDoS attack if the attacker is using 
legitimate
source IP addresses (or forged addresses that are forged to a valid 
host IP
on a subnet behind the first point that checking of the source address 
happens).

On any broadcast shared link media using ARP (e.g. Ethernet), it is 
possible
for one host to "steal" another host's IP address -- because ARP is not
authenticated.  IPv6 ND can be authenticated with AH, but the key 
management
is a nightmare to deploy (as with most forms of key management), so 
IPv6 ND
authentication is rarely enabled.

On many CPU-based routers, turning on uRPF features causes the 
forwarding
performance to drop below line rate, so not all operators enable this 
feature;
even operators who do deploy this feature often do so very selectively,
rather than all over the place.

uRPF also assumes that the local understanding of the BGP topology is 
correct,
legitimate, and that other ASs are only advertising prefixes that they 
are
authorised to advertise.  Right now, there is no way to authenticate 
any of
that.  (This is also one of the reasons that S-BGP is not actually 
deployable
right now, btw.)

> By the way, sinking too much traffic would cause a DoS as well because 
> you
> spend so much CPU resource on discarding suspicious packets instead of
> forwarding packets.

True only if the router uses CPU for discarding packets.  Many routers
(e.g. ours and some of our competitors') implement packet filtering in
an ASIC.  This means that ACLs are free and have zero adverse impact on
the system behaviour.  The tradeoff with an ASIC is that one can't add
new features as quickly as one can add them in software.  If a product
has all the features one needs implemented in dedicated hardware (e.g. 
ASIC),
the product will generally be a better choice than products that don't 
have
such dedicated hardware support.

> I have a question for you: if we have a DDoS of a magnitude , for the 
> lack
> of better word, of (NIMDA + CODE RED)^n, where n is a large number, 
> would
> the currently configured IP backbone be ready to withstand it?

The correctly configured hardware-based core routers had a relatively 
easy time
handling past DDOS attacks.  Correctly configured CPU-based core 
routers had a
a very difficult time handling past DDOS attacks.

Clever operators experimentally verify router behaviour under a wide 
range
of operating conditions (including adverse conditions) before buying 
routers.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep


*****
"The information transmitted is intended only for the person or entity to
which it is addressed and may contain confidential, proprietary, and/or
privileged material. Any review, retransmission, dissemination or other use
of, or taking of any action in reliance upon, this information by persons or
entities other than the intended recipient is prohibited. If you received
this in error, please contact the sender and delete the material from all
computers."
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 19 09:27:04 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA16782
	for <ieprep-archive@odin.ietf.org>; Thu, 19 Dec 2002 09:27:04 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBJETnO29225
	for ieprep-archive@odin.ietf.org; Thu, 19 Dec 2002 09:29:49 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJETmv29222
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 19 Dec 2002 09:29:48 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA16764
	for <ieprep-web-archive@ietf.org>; Thu, 19 Dec 2002 09:26:33 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJES7v29171;
	Thu, 19 Dec 2002 09:28:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJERQv29139
	for <ieprep@optimus.ietf.org>; Thu, 19 Dec 2002 09:27:26 -0500
Received: from bells.cs.ucl.ac.uk (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id JAA16745
	for <ieprep@ietf.org>; Thu, 19 Dec 2002 09:24:11 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.09044-0@bells.cs.ucl.ac.uk>; Thu, 19 Dec 2002 14:27:03 +0000
To: Mpierce1@aol.com
cc: ieprep@ietf.org
Subject: Re: [Ieprep] updated requirements documents (general)
In-reply-to: Your message of "Wed, 18 Dec 2002 23:29:04 EST." <49.283999b8.2b32a510@aol.com>
Date: Thu, 19 Dec 2002 14:27:02 +0000
Message-ID: <4221.1040308022@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


> [MAP] I understand this problem with the word "priority", which I used
> in a general sense to include the 4 examples I provided in my
> presentation. That is also why my presentation includes the explicit
> statement that "priority here does not imply any particular mechanism,
> e.g., priority queuing" and why I did not use the word "priority" in the
> text I proposed for the "general" draft. The use of "priority" in the
> Pierce drafts is its normal English meaning, nothing more, the same as
> in the previously proposed "Resource Priority Header". It is not used in
> the sense of priority queuing, that is, transmitting one packet ahead of
> others.

thats fine.  I won't grill you about mechanisms :-)
it was just important to clarify the context and its association
of the term you brought up.

> [MAP] I presume you meant IAA. 

yes, my apologies, i meant IAA.

> I don't consider IAA to be a "priority
> service" in the context which I was talking about requirements. 

ok.

> The IAA
> service as described does not contain any capability for one person to
> get better service over another. 

i guess that depends on what you define as 'better'.  my understanding
is that they have been looking at using DSCP.  in addition, they have
put forth the reflexive DSCP draft.  how and the extent they wish to
deploy it is fodder for another thread.  i have not taken a closer
look at how the use application layer signaling.

> Besides, we have been concentrating on
> "call setup" characteristics, such as probability of setup, probability
> of disconnect, etc. These do not relate to IAA. If the intention is for
> the "general" draft to only talk about thing that are common to every
> "ETS" service, then the above argument would mean that "Authentication"
> should not be listed as a requirement, either.

in section 1.2 (the Problem statement) of the General requirements
draft, we specifically state:

  "A subsequent problem is to ensure that this support is not focused
   just on IP telephony applications. The I-Am-Alive (IAA) database
   system is an example of an emergency related application used in
   Japan that supports both signaled and non-signaled access by
   users[10]."

I feel this statement makes it clear that the draft is going beyond just
IP telephony.  I also totally disagree with the opinion that
Authentication should not be listed as a requirement.  If we stick with
the IAA example, then users that log on and stipulate ("I am alive")
must be authenticated so that the stored information has a measure of
validity.  And I certainly sympathize with the lack of interest in using
accounting to determine whether I am "abusing" the system by registering
that I am alive.


> 2. provide a reason why the current structure of the two carlberg-ets-*
> requirements (general followed by more specific) does not work.
> 
> [MAP] It can work fine in this regard if the Accounting requirement is
> in the appropriate one. Sorry that I was concentrating on the "general"
> draft. See below.

ok

> 3) why doesn't a statement about an accouting requirement in the
> telephony requiremenst draft satisfy your perceived needs as they
> relate to MLPP?  Is there another case that you are contemplating?
> if so, what is it?
 
> [MAP] A statement about a requirement for Accounting in the Telephony
> draft could satisfy my perceived need, if that is in line with the
> structure of the two documents. It pertains to more than
> "MLPP". However, the current telephony draft only contains one paragraph
> about Accounting, and that is related to "Hybrid infrastructures that
> include and support connectivity to the PSTN". This sentence should be
> changed to not be restricted to the "interworking with the PSTN"
> case. The following is proposed:
>
>    4) Application layer IP telephony capabilities MUST NOT preclude
>      the ability to do accounting.

ok.  barring objections from other folks, I can change the 4'th 
requirement to what you propose above.

-ken

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 19 12:55:26 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA21606
	for <ieprep-archive@odin.ietf.org>; Thu, 19 Dec 2002 12:55:26 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBJHwBo09462
	for ieprep-archive@odin.ietf.org; Thu, 19 Dec 2002 12:58:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJHwBv09459
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 19 Dec 2002 12:58:11 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA21574
	for <ieprep-web-archive@ietf.org>; Thu, 19 Dec 2002 12:54:55 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJHuEv09236;
	Thu, 19 Dec 2002 12:56:15 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJHrcv09128
	for <ieprep@optimus.ietf.org>; Thu, 19 Dec 2002 12:53:38 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA21416
	for <ieprep@ietf.org>; Thu, 19 Dec 2002 12:50:22 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id B50A767103; Thu, 19 Dec 2002 12:54:39 -0500 (EST)
Date: Thu, 19 Dec 2002 12:53:19 -0500
Subject: Re: [Ieprep] updated requirements documents (general)
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <49.283999b8.2b32a510@aol.com>
Message-Id: <C20509E0-137A-11D7-8949-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Wednesday, Dec 18, 2002, at 23:29 America/Montreal, Mpierce1@aol.com 
wrote:
> such as "However, solutions used to realize ETS capabilities which
> provide one user with better service at the possible disadvantage of 
> others
> should include the capability of an accounting mechanism to support 
> auditing
> as defined in [RFC2975]." A similar qualification should be added to
> "Authentication".

Aside:
	RFC-2975 is an Informational RFC.  It is NOT standards-track.  Anyone
can publish nearly anything as an Informational RFC (e.g. IP over 
carrier
pigeon is an Informational RFC).  All Internet Standards are RFCs, but
many many RFCs are NOT Internet Standards.

	So the existence of RFC-2975 is not by itself justification for an
IETF WG to include accounting in any standards-track document or 
informational
document.  The justification for including such needs to be made on its
own merits, not relying on mis-reference to a non-standards-track 
document
for justification.

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 19 12:55:59 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA21628
	for <ieprep-archive@odin.ietf.org>; Thu, 19 Dec 2002 12:55:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBJHwis09495
	for ieprep-archive@odin.ietf.org; Thu, 19 Dec 2002 12:58:44 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJHwiv09492
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 19 Dec 2002 12:58:44 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA21620
	for <ieprep-web-archive@ietf.org>; Thu, 19 Dec 2002 12:55:28 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJHv2v09356;
	Thu, 19 Dec 2002 12:57:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBJHukv09310
	for <ieprep@optimus.ietf.org>; Thu, 19 Dec 2002 12:56:46 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA21513
	for <ieprep@ietf.org>; Thu, 19 Dec 2002 12:53:30 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 3E42467103; Thu, 19 Dec 2002 12:57:50 -0500 (EST)
Date: Thu, 19 Dec 2002 12:56:29 -0500
Subject: Re: [Ieprep] updated requirements documents (general)
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v548)
Cc: ieprep <ieprep@ietf.org>
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <49.283999b8.2b32a510@aol.com>
Message-Id: <3374EEA1-137B-11D7-8949-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.548)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Wednesday, Dec 18, 2002, at 23:29 America/Montreal, Mpierce1@aol.com 
wrote:
> [MAP] A statement about a requirement for Accounting in the Telephony 
> draft
> could satisfy my perceived need, if that is in line with the structure 
> of the
> two documents. It pertains to more than "MLPP". However, the current
> telephony draft only contains one paragraph about Accounting, and that 
> is
> related to "Hybrid infrastructures that include and support 
> connectivity to
> the PSTN". This sentence should be changed to not be restricted to the
> "interworking with the PSTN" case. The following is proposed:
>
>    4) Application layer IP telephony capabilities MUST NOT preclude
>      the ability to do accounting.

The text after "4)" is OK with me.

I'd be a bit happier if we said "application-layer accounting" rather 
than
"accounting", for the benefit of folks who read the document later and
haven't seen any presentations or list discussion where IP-layer 
accounting
was disclaimed.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 20 09:56:14 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA28567
	for <ieprep-archive@odin.ietf.org>; Fri, 20 Dec 2002 09:56:14 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBKExB620005
	for ieprep-archive@odin.ietf.org; Fri, 20 Dec 2002 09:59:11 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBKExAv20002
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 20 Dec 2002 09:59:10 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA28562
	for <ieprep-web-archive@ietf.org>; Fri, 20 Dec 2002 09:55:43 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBKEvVv19940;
	Fri, 20 Dec 2002 09:57:31 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBKEumv19912
	for <ieprep@optimus.ietf.org>; Fri, 20 Dec 2002 09:56:48 -0500
Received: from imo-d09.mx.aol.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA28502
	for <ieprep@ietf.org>; Fri, 20 Dec 2002 09:53:21 -0500 (EST)
From: Mpierce1@aol.com
Received: from Mpierce1@aol.com
	by imo-d09.mx.aol.com (mail_out_v34.13.) id j.bc.312bb8ec (16335);
	Fri, 20 Dec 2002 09:56:16 -0500 (EST)
Message-ID: <bc.312bb8ec.2b348990@aol.com>
Date: Fri, 20 Dec 2002 09:56:16 EST
Subject: Re: [Ieprep] updated requirements documents (general)
To: rja@extremenetworks.com
CC: ieprep@ietf.org
MIME-Version: 1.0
Content-Type: text/plain; charset="US-ASCII"
Content-Transfer-Encoding: 7bit
X-Mailer: AOL 6.0 for Windows XP US sub 51
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

In a message dated 12/19/2002 12:54:08 PM Eastern Standard Time, 
rja@extremenetworks.com writes:

Aside:
    RFC-2975 is an Informational RFC.  It is NOT standards-track.  Anyone
can publish nearly anything as an Informational RFC (e.g. IP over carrier
pigeon is an Informational RFC).  All Internet Standards are RFCs, but
many many RFCs are NOT Internet Standards.

[MAP] Yes, I am fully aware of that.


[RJA]    So the existence of RFC-2975 is not by itself justification for an
IETF WG to include accounting in any standards-track document or informational
document.  The justification for including such needs to be made on its
own merits, not relying on mis-reference to a non-standards-track document
for justification.

[MAP] I was not intending RFC2975 as "justification" for including 
"accounting" in this requirements draft. It simply provides more information 
on what "accounting" (and "auditing" in particular) means.

I presume that the draft under discussion is not a standards-track RFC either.


In a message dated 12/19/2002 12:59:18 PM Eastern Standard Time, 
rja@extremenetworks.com writes:

I'd be a bit happier if we said "application-layer accounting" rather than
"accounting", for the benefit of folks who read the document later and
haven't seen any presentations or list discussion where IP-layer accounting
was disclaimed.

[MAP] That fine.


Mike Pierce
Artel
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Fri Dec 20 10:39:46 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA29654
	for <ieprep-archive@odin.ietf.org>; Fri, 20 Dec 2002 10:39:46 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBKFgh022545
	for ieprep-archive@odin.ietf.org; Fri, 20 Dec 2002 10:42:43 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBKFghv22542
	for <ieprep-web-archive@optimus.ietf.org>; Fri, 20 Dec 2002 10:42:43 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA29617
	for <ieprep-web-archive@ietf.org>; Fri, 20 Dec 2002 10:39:15 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBKFf5v22481;
	Fri, 20 Dec 2002 10:41:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBKFeIv22449
	for <ieprep@optimus.ietf.org>; Fri, 20 Dec 2002 10:40:18 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA29515
	for <ieprep@ietf.org>; Fri, 20 Dec 2002 10:36:51 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 694E767103; Fri, 20 Dec 2002 10:41:22 -0500 (EST)
Date: Fri, 20 Dec 2002 10:39:51 -0500
Subject: Re: [Ieprep] updated requirements documents (general)
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v551)
Cc: ieprep@ietf.org
To: Mpierce1@aol.com
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <bc.312bb8ec.2b348990@aol.com>
Message-Id: <477292C6-1431-11D7-81DD-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.551)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Friday, Dec 20, 2002, at 09:56 America/Montreal, Mpierce1@aol.com 
wrote:
> I presume that the draft under discussion is not a standards-track RFC 
> either.

	I don't know how that draft will end up, since that is up to the WG.  
One
possibility would be that it would end up as BCP.  BCP is a 
standards-track state.

Ran
rja@extremenetworks.com

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 21 10:25:28 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA07662
	for <ieprep-archive@odin.ietf.org>; Sat, 21 Dec 2002 10:25:27 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBLFSag02298
	for ieprep-archive@odin.ietf.org; Sat, 21 Dec 2002 10:28:36 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBLFSav02295
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 21 Dec 2002 10:28:36 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA07659
	for <ieprep-web-archive@ietf.org>; Sat, 21 Dec 2002 10:24:56 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBLFQJv02230;
	Sat, 21 Dec 2002 10:26:19 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBHJhWv31972
	for <ieprep@optimus.ietf.org>; Tue, 17 Dec 2002 14:43:33 -0500
Received: from cougar.icir.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA21276
	for <ieprep@ietf.org>; Tue, 17 Dec 2002 14:40:27 -0500 (EST)
Received: from cougar.icir.org (localhost [127.0.0.1])
	by cougar.icir.org (8.12.3/8.12.3) with ESMTP id gBHJhQgn063784;
	Tue, 17 Dec 2002 11:43:26 -0800 (PST)
	(envelope-from floyd@cougar.icir.org)
Message-Id: <200212171943.gBHJhQgn063784@cougar.icir.org>
To: "King, Kimberly S." <KIMBERLY.S.KING@saic.com>
cc: "Ieprep (E-mail)" <ieprep@ietf.org>,
        "'RJ Atkinson'" <rja@extremenetworks.com>
From: Sally Floyd <floyd@icir.org>
Subject: Re: [Ieprep] Summary of ieprep work and VoIP--Take 2 
Date: Tue, 17 Dec 2002 11:43:26 -0800
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Kimberly -

>"If other non-VoIP traffic is also carried, then there is a 
>tradeoff to make between the performance of voice and non-voice 
>applications.  For example, enough VoIP on a congested link 
>(e.g., 10-40% packet drop rates) can be sufficient to make the
>link essentially unusable for TCP traffic. "

So I went back and read the email that this came from.

I think I will steer clear of a discussion of what is "usable" or
"unusable" for TCP traffic, as I would agree that that is not really
the point.  (I don't know if the word "unusable" originally came
from me or not...)

With 40% packet drop rates, TCP traffic is undergoing frequent
retransmit timeouts (because when the congestion window is one
packet, and that packet is dropped, a retransmit timeout is the
only recovery mechanism).  With the recommended minimum retransmit
timeout value of one second, and the occasional exponential backoff
of the retransmit timer, a TCP connection is getting through less
than one packet per second.  And if there were multiple congested
routers along the path, then potentially-useful bandwidth was wasted
carrying packets that would only be dropped downstream.

With 10% packet drop rates, a TCP connection could conceivably be
seeing a throughout of several packets per RTT (e.g., if the TCP
implementation is using Limited Transmit, and thereby manages to
avoid too many retransmit timeouts).  This is not too bad.  Of
course, there still are those connections that have long initial
delays (several seconds) retransmitting their dropped SYN packets.

Since everyone agrees on this:

>- Letting non-congestion-avoiding traffic crowd out congestion-avoiding
>   traffic is a very bad thing from an architectural or system engineering
>   perspective.
>- VoIP or any other application that doesn't include congestion-avoidance
>   needs to be modified to implement and use congestion-avoidance.

that seems to me to be what really matters.

(The TCP traffic *could* be carrying email or instant messaging
traffic or web traffic that is just as important for the emergency
communications as the VoIP traffic, it would seem to me...)

(And since this was just about an email message, the original email
message doesn't even need to be revised and resubmitted...)

- Sally


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 23 08:58:50 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA02313
	for <ieprep-archive@odin.ietf.org>; Mon, 23 Dec 2002 08:58:50 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBNE2Kn12627
	for ieprep-archive@odin.ietf.org; Mon, 23 Dec 2002 09:02:20 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBNE2Kv12624
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 23 Dec 2002 09:02:20 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA02301
	for <ieprep-web-archive@ietf.org>; Mon, 23 Dec 2002 08:58:19 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBNE0Ev12505;
	Mon, 23 Dec 2002 09:00:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBNDwwv12378
	for <ieprep@optimus.ietf.org>; Mon, 23 Dec 2002 08:58:58 -0500
Received: from pfwhqs1.ncr.disa.mil (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA01978
	for <ieprep@ietf.org>; Mon, 23 Dec 2002 08:54:57 -0500 (EST)
Received: from mtahqs3.ncr.disa.mil by pfwhqs1.ncr.disa.mil
          via smtpd (for ietf-mx.ietf.org [132.151.6.1]) with SMTP; 23 Dec 2002 13:54:02 UT
Received: by mtahqs3.ncr.disa.mil with Internet Mail Service (5.5.2653.19)
	id <Z29JGGYK>; Mon, 23 Dec 2002 08:59:49 -0500
Message-ID: <7F18415E4D63CB45BB9B3A591F68D12D02EF4C77@emshqs1.ncr.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'Sally Floyd'" <floyd@icir.org>
Cc: "Ieprep (E-mail)" <ieprep@ietf.org>,
        "'RJ Atkinson'"
	 <rja@extremenetworks.com>,
        "King, Kimberly S."
	 <KIMBERLY.S.KING@saic.com>
Subject: RE: [Ieprep] Summary of ieprep work and VoIP--Take 2 
Date: Mon, 23 Dec 2002 08:57:51 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Sally,

I am just curious. What is the transport-layer mix? 
90% ~ TCP 
10% ~ UDP and others 

With a ratio of, say, 80% for TCP and 20% for UDP and packet drop rates of
10-40%, would the link be more or less unusable for TCP traffic?

Thanks,

An

-----Original Message-----
From: Sally Floyd [mailto:floyd@icir.org]
Sent: Tuesday, December 17, 2002 2:43 PM
To: King, Kimberly S.
Cc: Ieprep (E-mail); 'RJ Atkinson'
Subject: Re: [Ieprep] Summary of ieprep work and VoIP--Take 2 


Kimberly -

>"If other non-VoIP traffic is also carried, then there is a 
>tradeoff to make between the performance of voice and non-voice 
>applications.  For example, enough VoIP on a congested link 
>(e.g., 10-40% packet drop rates) can be sufficient to make the
>link essentially unusable for TCP traffic. "

So I went back and read the email that this came from.

I think I will steer clear of a discussion of what is "usable" or
"unusable" for TCP traffic, as I would agree that that is not really
the point.  (I don't know if the word "unusable" originally came
from me or not...)

With 40% packet drop rates, TCP traffic is undergoing frequent
retransmit timeouts (because when the congestion window is one
packet, and that packet is dropped, a retransmit timeout is the
only recovery mechanism).  With the recommended minimum retransmit
timeout value of one second, and the occasional exponential backoff
of the retransmit timer, a TCP connection is getting through less
than one packet per second.  And if there were multiple congested
routers along the path, then potentially-useful bandwidth was wasted
carrying packets that would only be dropped downstream.

With 10% packet drop rates, a TCP connection could conceivably be
seeing a throughout of several packets per RTT (e.g., if the TCP
implementation is using Limited Transmit, and thereby manages to
avoid too many retransmit timeouts).  This is not too bad.  Of
course, there still are those connections that have long initial
delays (several seconds) retransmitting their dropped SYN packets.

Since everyone agrees on this:

>- Letting non-congestion-avoiding traffic crowd out congestion-avoiding
>   traffic is a very bad thing from an architectural or system engineering
>   perspective.
>- VoIP or any other application that doesn't include congestion-avoidance
>   needs to be modified to implement and use congestion-avoidance.

that seems to me to be what really matters.

(The TCP traffic *could* be carrying email or instant messaging
traffic or web traffic that is just as important for the emergency
communications as the VoIP traffic, it would seem to me...)

(And since this was just about an email message, the original email
message doesn't even need to be revised and resubmitted...)

- Sally


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 23 09:02:21 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA02693
	for <ieprep-archive@odin.ietf.org>; Mon, 23 Dec 2002 09:02:21 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBNE5pe12856
	for ieprep-archive@odin.ietf.org; Mon, 23 Dec 2002 09:05:51 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBNE5pv12853
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 23 Dec 2002 09:05:51 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA02681
	for <ieprep-web-archive@ietf.org>; Mon, 23 Dec 2002 09:01:50 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBNE41v12701;
	Mon, 23 Dec 2002 09:04:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBNE3Ev12654
	for <ieprep@optimus.ietf.org>; Mon, 23 Dec 2002 09:03:14 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA02360
	for <ieprep@ietf.org>; Mon, 23 Dec 2002 08:59:13 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 23 Dec 2002 09:01:32 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002122309015823311
 for <ieprep@ietf.org>; Mon, 23 Dec 2002 09:01:58 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <ZG6KA8WT>; Mon, 23 Dec 2002 09:01:44 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77AE5@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "Ieprep (E-mail)" <ieprep@ietf.org>
Date: Mon, 23 Dec 2002 09:01:55 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] Emergency Preparedness--Example
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

"For the latest exercise in October, emergency management personnel hooked
up a laptop computer - loaded with a mapping program - to a projector
displaying the nine-county area surrounding the depot. Underlying the maps
were data on evacuation routes, the location of daycare facilities, nursing
homes and assisted living sites, hospitals and schools, and other pertinent
information."

One way that they use (presumably over an IP network) is to email updated
information and other data.

more at 
http://www.fcw.com/geb/articles/2002/1223/web-cssep-12-23-02.asp

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 23 13:15:22 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA11334
	for <ieprep-archive@odin.ietf.org>; Mon, 23 Dec 2002 13:15:22 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBNIItr27915
	for ieprep-archive@odin.ietf.org; Mon, 23 Dec 2002 13:18:55 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBNIItv27912
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 23 Dec 2002 13:18:55 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA11317
	for <ieprep-web-archive@ietf.org>; Mon, 23 Dec 2002 13:14:51 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBNIH6v27857;
	Mon, 23 Dec 2002 13:17:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBNIGrv27826
	for <ieprep@optimus.ietf.org>; Mon, 23 Dec 2002 13:16:53 -0500
Received: from kc-msxproto4.kc.umkc.edu (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA11285
	for <ieprep@ietf.org>; Mon, 23 Dec 2002 13:12:49 -0500 (EST)
Received: from KC-MAIL2.kc.umkc.edu ([134.193.143.162] RDNS failed) by kc-msxproto4.kc.umkc.edu with Microsoft SMTPSVC(5.0.2195.5329);
	 Mon, 23 Dec 2002 12:15:52 -0600
X-MimeOLE: Produced By Microsoft Exchange V6.0.6249.0
Content-Class: urn:content-classes:message
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: RE: [Ieprep] Summary of ieprep work and VoIP--Take 2 
Date: Mon, 23 Dec 2002 12:15:52 -0600
Message-ID: <A29143A40AEAE3459FF829D1DD43316101D9B0B8@KC-MAIL2.kc.umkc.edu>
Thread-Topic: [Ieprep] Summary of ieprep work and VoIP--Take 2 
Thread-Index: AcKqi4Uoq83URzgUTpOn4P+NzGa18wAHsKXA
From: "Ayyasamy, Senthilkumar  (UMKC-Student)" <saq66@umkc.edu>
To: "Nguyen, An" <nguyena@ncs.gov>, "Sally Floyd" <floyd@icir.org>
Cc: "Ieprep (E-mail)" <ieprep@ietf.org>,
        "RJ Atkinson" <rja@extremenetworks.com>,
        "King, Kimberly S." <KIMBERLY.S.KING@saic.com>
X-OriginalArrivalTime: 23 Dec 2002 18:15:52.0910 (UTC) FILETIME=[543D56E0:01C2AAAF]
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBNIGsv27827
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

> What is the transport-layer mix? 
> 90% ~ TCP 
> 10% ~ UDP and others 

For a more updated TCP/UDP ratio:
http://ipmon.sprintlabs.com/packstat/viewresult.php?8:protobreakdown:nyc-09.0-021121
compare with the old data from caida collected at AIX exchange.

You can observe 90:10 from a raw  tcpdump data taken from your system too.

The 90:10-tcp:udp is the standard observation.
But, more interesting will be knowing who contributes to the UDP traffic.
Obviously file sharing..
http://ipmon.sprintlabs.com/packstat/viewresult.php?2:appsbreakdown:nyc-02.0-021009

Also, an interesting work will be pumping a good traffic mix across a
congested link (at some IX point) and see, how much UDP  
affects TCP based applications? or a trace driven simulation.
This will give some clue on when to terminate the application-limited flows
to leave way for tcp/tcp-friendly traffic.

btw, you can also see the link utilization and how well provisioned the sprint 
backbones from those pages.





> 
> With a ratio of, say, 80% for TCP and 20% for UDP 

I guess..90/10 is the standard measured one. If your dump data shows 80/20 or
less, we really need to think about ways to protect tcp/tcp-friendly traffic
from UDP traffic (like when to terminate a UDP flow and use of ECN to do
such work.) 

> and packet 
> drop rates of
> 10-40%, would the link be more or less unusable for TCP traffic?
> 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Tue Dec 24 09:29:31 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA08009
	for <ieprep-archive@odin.ietf.org>; Tue, 24 Dec 2002 09:29:31 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBOEXPM20988
	for ieprep-archive@odin.ietf.org; Tue, 24 Dec 2002 09:33:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBOEXPJ20985
	for <ieprep-web-archive@optimus.ietf.org>; Tue, 24 Dec 2002 09:33:25 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07992
	for <ieprep-web-archive@ietf.org>; Tue, 24 Dec 2002 09:29:00 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBOEVXJ20948;
	Tue, 24 Dec 2002 09:31:33 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBOEUFJ20911
	for <ieprep@optimus.ietf.org>; Tue, 24 Dec 2002 09:30:15 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07892
	for <ieprep@ietf.org>; Tue, 24 Dec 2002 09:25:50 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Tue, 24 Dec 2002 09:28:11 -0500
Received: from mcl-its-exig01.mail.saic.com ([149.8.64.12])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002122409283811651
 for <ieprep@ietf.org>; Tue, 24 Dec 2002 09:28:38 -0500
Received: by mcl-its-exig01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <ZG6KD1H2>; Tue, 24 Dec 2002 09:28:24 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77AF6@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "Ieprep (E-mail)" <ieprep@ietf.org>
Date: Tue, 24 Dec 2002 09:28:31 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBOEUFJ20912
Subject: [Ieprep] Internet emergency preparedness--more examples
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit


I continue to research Internet emergency preparedness.  Here are more of my
findings...

Internet access isn't a given, even in the U.S.  I don't mean the case where
an emergency occurs in an unexpected place and first-responders have no
access connection.  Rather, I refer to pre-determined locations.  

Wisconsin 2002 

"A recent study of Critical Access Hospitals revealed that only half had
internet access and many have computer equipment that does not accommodate
CD-ROMs. Many have only one e-mail address for the entire hospital. Staff
lack training and time to locate and interpret information distributed on
web pages. Although CAHs have received recent state funding for computers
and software, the adequacy of these resources for disease surveillance,
emergency response and disaster communications has not yet been evaluated or
tested."

http://www.dhfs.state.wi.us/dph_bcd/Bioterrorism/pdf_files/WisconsinBioterro
rismGrant_HRSA.pdf

"Mohammad Akhter, MD, MPH, director
of the American Public Health
Association and a former state EMS director
and state EMS medical director, told
another Senate panel Oct. 9 that 10% of
the nation's public health departments
don't even have Internet access." (in 2001)

http://www.jems.com/911/pdf/ins1101.pdf

****************************************************************************
*************

Meanwhile, other states such as Missouri and Massachusetts or cities like
Dallas among others seem to be communicating using the Internet


"In the event of a disaster or terrorist attack, Missouri now has a
statewide, Internet-based communications system linking hospitals, major
metropolitan emergency services providers, local public health agencies and
the Missouri Department of Health and Senior Services.
Known as the EMSystem®, the communications network allows users to
communicate in real time their abilities to manage various levels of
casualties. If ground based telecommunications lines are amaged or
overloaded, the system can use satellite-based Internet telecommunications
as a back-up. Users need only standard computer capabilities and Internet
access."

http://web.mhanet.com/asp/reporters/news_releases/emsystem1002.asp





"To improve communication and help eliminate this problem in the Dallas-Fort
Worth area, efforts are under way to install EMSystem, a Web-based dispatch
system allowing EMS agencies immediately to know which hospitals can provide
suitable care. 
Information about the number and types of patients a particular hospital can
accept is updated at least every few hours. Emergency personnel, using a
computer with Internet access and a Web browser, determine the appropriate
hospital. "  

http://dallas.bizjournals.com/dallas/stories/2001/10/22/focus1.html

http://www.emsmdc.com/emsystem_overview.htm
 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 25 21:26:08 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24133
	for <ieprep-archive@odin.ietf.org>; Wed, 25 Dec 2002 21:26:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBQ2Ums19156
	for ieprep-archive@odin.ietf.org; Wed, 25 Dec 2002 21:30:48 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQ2UlJ19153
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 25 Dec 2002 21:30:47 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24130
	for <ieprep-web-archive@ietf.org>; Wed, 25 Dec 2002 21:25:37 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQ2SWJ19104;
	Wed, 25 Dec 2002 21:28:33 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQ2RaJ19067
	for <ieprep@optimus.ietf.org>; Wed, 25 Dec 2002 21:27:36 -0500
Received: from multicasttech.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24082
	for <ieprep@ietf.org>; Wed, 25 Dec 2002 21:22:26 -0500 (EST)
Received: from [68.100.8.94] (account marshall_eubanks HELO localhost)
  by multicasttech.com (CommuniGate Pro SMTP 3.4.8)
  with ESMTP-TLS id 1561537 for ieprep@ietf.org; Wed, 25 Dec 2002 21:25:30 -0500
Date: Wed, 25 Dec 2002 21:25:31 -0500
Content-Type: text/plain; charset=WINDOWS-1252; format=flowed
Mime-Version: 1.0 (Apple Message framework v482)
From: Marshall Eubanks <tme@multicasttech.com>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
In-Reply-To: <Pine.GSO.4.44.0212180851300.14120-100000@paixhost.pch.net>
Message-Id: <4E3406B2-1879-11D7-B24F-003065BA697E@multicasttech.com>
X-Mailer: Apple Mail (2.482)
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by www1.ietf.org id gBQ2RaJ19068
Subject: [Ieprep] Globe and Mail on Internet Attacks
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 8bit
Content-Transfer-Encoding: 8bit

I thought this might be of interest to group members - it's reasonably 
thoughtful :

http://www.globeandmail.com/servlet/ArticleNews/front/RTGAM/20021225/wcybe1225a/
Front/homeBN/breakingnews

Regards & Merry Christmas to All;
Marshall Eubanks

POSTED AT 5:54 PM EST    Wednesday, December 25
space
Cyberspace sentinels brace for trouble

Canadian Press

Vancouver — The fight against terrorism and the prospect of hostilities 
with Iraq have the sentinels of cyberspace bracing for trouble.

Experts say it's only a matter of time before someone mounts a 
concerted, politically motivated attack on the Internet or a piece of 
computer-dependent infrastructure such as the electrical grid.

Despite growing security awareness, especially in the wake of the Sept. 
11, 2001, terrorist attacks, many critical systems remain open to 
intrusion and disruption, authorities in both the private and public 
sectors agree.

"The problem at this point is that the vulnerabilities are so numerous 
one has a hard time trying to decide where to start," said Andrew 
McAllister, director of cyber protection at the federal Office of 
Critical Infrastructure and Emergency Preparedness.

<snip> - More follows

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Wed Dec 25 21:29:00 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24160
	for <ieprep-archive@odin.ietf.org>; Wed, 25 Dec 2002 21:28:59 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBQ2Xdu19236
	for ieprep-archive@odin.ietf.org; Wed, 25 Dec 2002 21:33:39 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQ2XdJ19233
	for <ieprep-web-archive@optimus.ietf.org>; Wed, 25 Dec 2002 21:33:39 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24156
	for <ieprep-web-archive@ietf.org>; Wed, 25 Dec 2002 21:28:28 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQ2W1J19206;
	Wed, 25 Dec 2002 21:32:01 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQ2VYJ19184
	for <ieprep@optimus.ietf.org>; Wed, 25 Dec 2002 21:31:34 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA24147
	for <ieprep@ietf.org>; Wed, 25 Dec 2002 21:26:19 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBQ2Mbo04528;
	Wed, 25 Dec 2002 18:23:05 -0800 (PST)
Date: Wed, 25 Dec 2002 18:22:37 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "Nguyen, An" <nguyena@ncs.gov>
cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
In-Reply-To: <7F18415E4D63CB45BB9B3A591F68D12D02EF4C6F@emshqs1.ncr.disa.mil>
Message-ID: <Pine.GSO.4.44.0212251808050.3122-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Wed, 18 Dec 2002, Nguyen, An wrote:
    > NAT comes in two flavors: static and dynamic. If using NAT is not a good
    > thing to do, what can be used instead for network security?

That question implies that NATs have something to do with security, which
they don't.  NATs are a hack which equipment vendors created in response
to artificial scarcity of IP addresses which marketing idiots in some ISPs
dreamed up as a way of getting back towards profitability after they
priced transit too low in hopes of attracting customers away from
competitors who were themselves already selling transit below cost.  It's
just an arms-race between ISP marketing folks and equipment vendors.
That's where dynamic NAT came from as well...  It was a successful attempt
to break the already-deployed NAT boxes.  However, it backfired, because
equipment vendors _like_ it when you break their already-deployed
equipment, since it means that customers have to come back and buy more.
The ISP marketing idiots have since moved on to trying to use legal means,
writing and then attempting to enforce a contractual relationship between
price paid and number of users on the end of a line.  Which is equally
hopeless, but they still refuse to move away from flat-rate pricing, which
is really the root of their problem.  Not that I encourage such a move,
I think their whole posiition is ill-conceived.

If one were to try to dream up some hypothetical relationship between NATs
and security, it would be along the same lines as air-gap.  If you break
the network, traffic can't get through it, and the stuff on one side of
the gap is thus secure against attack from stuff on the other side of the
gap.  Not that that actually secures any of the machines.

If you want security, write some firewall rules, like everybody else.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 26 08:25:56 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10135
	for <ieprep-archive@odin.ietf.org>; Thu, 26 Dec 2002 08:25:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBQDUm025984
	for ieprep-archive@odin.ietf.org; Thu, 26 Dec 2002 08:30:48 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQDUmJ25981
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 26 Dec 2002 08:30:48 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10131
	for <ieprep-web-archive@ietf.org>; Thu, 26 Dec 2002 08:25:25 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQDOAJ25864;
	Thu, 26 Dec 2002 08:24:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQDNPJ25821
	for <ieprep@optimus.ietf.org>; Thu, 26 Dec 2002 08:23:25 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA10087
	for <ieprep@ietf.org>; Thu, 26 Dec 2002 08:18:02 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Thu, 26 Dec 2002 08:20:25 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002122608205122556
 ; Thu, 26 Dec 2002 08:20:51 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <Z2TNFGQX>; Thu, 26 Dec 2002 08:22:30 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77AFD@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'Bill Woodcock'" <woody@pch.net>, "Nguyen, An" <nguyena@ncs.gov>
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
Date: Thu, 26 Dec 2002 08:20:50 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Bill said:
>NATs are a hack which equipment vendors created 
>in response
>to artificial scarcity of IP addresses which marketing idiots 
>in some ISPs
>dreamed up as a way of getting back towards profitability after they
>priced transit too low ...

For an alternate perspective on addresses,

"Everyone who wants one is not able to get one
Example: People's Republic of China
1.3 Billion people
Order of magnitude growth in Internet usage year over year
~9M addresses in 1999
~16M addresses in 2001
Do we simply assume that anyone who has not already asked never will? 
Africa, South America, India, Arab world..."
http://www.isoc-au.org.au/Events/IPv6Direction.ppt

Kimberly
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 26 09:03:48 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA10553
	for <ieprep-archive@odin.ietf.org>; Thu, 26 Dec 2002 09:03:47 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBQE8fm28093
	for ieprep-archive@odin.ietf.org; Thu, 26 Dec 2002 09:08:41 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQE8fJ28090
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 26 Dec 2002 09:08:41 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA10550
	for <ieprep-web-archive@ietf.org>; Thu, 26 Dec 2002 09:03:16 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQE23J27285;
	Thu, 26 Dec 2002 09:02:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQE1eJ27271
	for <ieprep@optimus.ietf.org>; Thu, 26 Dec 2002 09:01:40 -0500
Received: from [209.22.88.17] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id IAA10406
	for <ieprep@ietf.org>; Thu, 26 Dec 2002 08:56:16 -0500 (EST)
Received: from mtassp3.ncr.disa.mil by [209.22.88.17]
          via smtpd (for ietf-mx.ietf.org [132.151.6.1]) with SMTP; 26 Dec 2002 13:51:40 UT
Received: by mtassp3.ncr.disa.mil with Internet Mail Service (5.5.2653.19)
	id <Z24WJAG4>; Thu, 26 Dec 2002 08:59:24 -0500
Message-ID: <7F18415E4D63CB45BB9B3A591F68D12D02EF4C7E@emshqs1.ncr.disa.mil>
From: "Nguyen, An" <nguyena@ncs.gov>
To: "'Bill Woodcock'" <woody@pch.net>
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
Date: Thu, 26 Dec 2002 08:59:16 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Bill,

-----Original Message-----
From: Bill Woodcock [mailto:woody@pch.net]
Sent: Wednesday, December 25, 2002 9:23 PM
To: Nguyen, An
Cc: 'ieprep@ietf.org'
Subject: RE: [Ieprep] Understanding Congestion

>If you want security, write some firewall rules, like everybody else.

Actually, we use both firewall rules and NAT as needed.

Happy holidays!

An


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 26 10:58:56 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA11873
	for <ieprep-archive@odin.ietf.org>; Thu, 26 Dec 2002 10:58:56 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBQG3o632455
	for ieprep-archive@odin.ietf.org; Thu, 26 Dec 2002 11:03:50 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQG3oJ32452
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 26 Dec 2002 11:03:50 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA11859
	for <ieprep-web-archive@ietf.org>; Thu, 26 Dec 2002 10:58:24 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQFvNJ32251;
	Thu, 26 Dec 2002 10:57:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQFuMJ32220
	for <ieprep@optimus.ietf.org>; Thu, 26 Dec 2002 10:56:22 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA11777
	for <ieprep@ietf.org>; Thu, 26 Dec 2002 10:50:57 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Thu, 26 Dec 2002 10:53:28 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002122610535522574
 for <ieprep@ietf.org>; Thu, 26 Dec 2002 10:53:55 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <ZVCV9K3J>; Thu, 26 Dec 2002 10:55:34 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77B01@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "Ieprep (E-mail)" <ieprep@ietf.org>
Date: Thu, 26 Dec 2002 10:53:54 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Subject: [Ieprep] Iowa Emergency Management Division--Internet Use
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

1) How does your organization use the Internet during emergency situations
or for emergency recovery?  

The internet is used during emergency situations
to obtain weather information; river flood stages; monitor CNN, NBC and
other major news websites.  The internet is also used for relaying
information from the field to the emergency operations center.



2) Is there a way that the Internet fails to meet your needs in these
 situations?  Will you describe the symptoms of the problem?  

No (problems)


3) Do you envision using Voice over IP applications in such situations?
(For example, Microsoft NetMeeting can support audio conversations between
people like a traditional telephone call except using their computers with
microphones and speakers)  

Not aware of any plans for this use at this time.


4) Have you already tried such voice applications over the Internet to aid
emergency recovery?  

No			

5) In other words, how would you like to use the Internet during an
emergency or how do you think it could be of assistance?  

At this time, the internet is serving the purposes we use for it during
disaster
response/recovery.  As new uses become available, we are constantly looking
at ways to use new technology.
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Thu Dec 26 11:23:29 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA12237
	for <ieprep-archive@odin.ietf.org>; Thu, 26 Dec 2002 11:23:29 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBQGSO201143
	for ieprep-archive@odin.ietf.org; Thu, 26 Dec 2002 11:28:24 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQGSOJ01140
	for <ieprep-web-archive@optimus.ietf.org>; Thu, 26 Dec 2002 11:28:24 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA12232
	for <ieprep-web-archive@ietf.org>; Thu, 26 Dec 2002 11:22:58 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQGM2J01011;
	Thu, 26 Dec 2002 11:22:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBQGLGJ00989
	for <ieprep@optimus.ietf.org>; Thu, 26 Dec 2002 11:21:16 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA12134
	for <ieprep@ietf.org>; Thu, 26 Dec 2002 11:15:50 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBQGCVo08557;
	Thu, 26 Dec 2002 08:12:31 -0800 (PST)
Date: Thu, 26 Dec 2002 08:12:31 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
cc: "Nguyen, An" <nguyena@ncs.gov>, "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] Understanding Congestion
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77AFD@mcl-its-exs01.mail.saic.com>
Message-ID: <Pine.GSO.4.44.0212260810500.4535-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Thu, 26 Dec 2002, King, Kimberly  S. wrote:
    > For an alternate perspective on addresses,
    > "Everyone who wants one is not able to get one

Yes, but the question was why NAT exists.  Nobody was thinking of China
when NAT was created, it's just a gross hack.  IPv6 is the solution to the
"not enough IP addresses" problem.  NAT doesn't create any additional IP
addresses, so it can't solve China's problem.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 28 15:45:20 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA08802
	for <ieprep-archive@odin.ietf.org>; Sat, 28 Dec 2002 15:45:20 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBSKpJO31651
	for ieprep-archive@odin.ietf.org; Sat, 28 Dec 2002 15:51:19 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBSKpJJ31648
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 28 Dec 2002 15:51:19 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA08793
	for <ieprep-web-archive@ietf.org>; Sat, 28 Dec 2002 15:44:49 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBSKnNJ31615;
	Sat, 28 Dec 2002 15:49:23 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBSKmHJ31583
	for <ieprep@optimus.ietf.org>; Sat, 28 Dec 2002 15:48:17 -0500
Received: from joy.songbird.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA08773
	for <ieprep@ietf.org>; Sat, 28 Dec 2002 15:41:46 -0500 (EST)
Received: from bbprime.brandenburg.com (208.184.79.252.songbird.com [208.184.79.252] (may be forged))
	by joy.songbird.com (8.9.3/8.9.3) with ESMTP id MAA31335;
	Sat, 28 Dec 2002 12:44:28 -0800
Date: Sat, 28 Dec 2002 12:43:02 -0800
From: Dave Crocker <dhc2@dcrocker.net>
X-Mailer: The Bat! (v1.62 Christmas Edition) Personal
Organization: Brandenburg InternetWorking
X-Priority: 3 (Normal)
Message-ID: <199883169.20021228124302@brandenburg.com>
To: Bill Woodcock <woody@pch.net>
CC: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>,
        "Nguyen, An" <nguyena@ncs.gov>, "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: Re: [Ieprep] Understanding Congestion
In-Reply-To: <Pine.GSO.4.44.0212260810500.4535-100000@paixhost.pch.net>
References: <Pine.GSO.4.44.0212260810500.4535-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Folks,

Bill> Yes, but the question was why NAT exists.  Nobody was thinking of China
Bill> when NAT was created, it's just a gross hack.  IPv6 is the solution to the
Bill> "not enough IP addresses" problem.  NAT doesn't create any additional IP
Bill> addresses, so it can't solve China's problem.

I do not like NAT, either, but we need to be careful about dismissing it
simply as a hack. It responded to two, real-world issues:

1.  NAT environments has been more convenient for small networks than was
(is?) possible without NAT.  That's not an address space issue, but a
network (and host) administration issue.

2.  In effect, NAT does for a network what common networking software does
for a host.  That is, it provides an integrated point of lower-level
addressing for a collection of processes (machines.)  In effect, one can
call this a kind of addressing data compression.  Data compression is a good
way to increase virtual throughput, so in fact one can argue that NAT does,
in fact, achieve the same address scaling effect as adding new address
space.

d/
-- 
 Dave <mailto:dhc2@dcrocker.net>
 Brandenburg InternetWorking <http://www.brandenburg.com>
 t +1.408.246.8253; f +1.408.850.1850

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 28 20:07:02 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA10984
	for <ieprep-archive@odin.ietf.org>; Sat, 28 Dec 2002 20:07:02 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBT1D6Q09359
	for ieprep-archive@odin.ietf.org; Sat, 28 Dec 2002 20:13:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBT1D6J09356
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 28 Dec 2002 20:13:06 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA10981
	for <ieprep-web-archive@ietf.org>; Sat, 28 Dec 2002 20:06:30 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBT1BSJ09315;
	Sat, 28 Dec 2002 20:11:28 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBT1A3J09295
	for <ieprep@optimus.ietf.org>; Sat, 28 Dec 2002 20:10:03 -0500
Received: from paixhost.pch.net (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA10953
	for <ieprep@ietf.org>; Sat, 28 Dec 2002 20:03:28 -0500 (EST)
Received: from ns1.pch.net (ns1.pch.net [206.220.231.1])
	by paixhost.pch.net (8.11.6/8.11.6) with ESMTP id gBT0xvo24217;
	Sat, 28 Dec 2002 16:59:57 -0800 (PST)
Date: Sat, 28 Dec 2002 16:59:57 -0800 (PST)
From: Bill Woodcock <woody@pch.net>
To: Dave Crocker <dhc2@dcrocker.net>
cc: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>,
        "Nguyen, An" <nguyena@ncs.gov>, "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: Re: [Ieprep] Understanding Congestion
In-Reply-To: <199883169.20021228124302@brandenburg.com>
Message-ID: <Pine.GSO.4.44.0212281658560.24186-100000@paixhost.pch.net>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

      On Sat, 28 Dec 2002, Dave Crocker wrote:
    > 1.  NAT environments has been more convenient for small networks than was
    > (is?) possible without NAT.  That's not an address space issue, but a
    > network (and host) administration issue.

NAT mitigates the problem of renumbering as well as the problem of
artificial scarcity, but at the cost of many many other problems.  That's
hardly beneficial, by any measure.

                                -Bill


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 28 20:57:09 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA11393
	for <ieprep-archive@odin.ietf.org>; Sat, 28 Dec 2002 20:57:08 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBT23Es10886
	for ieprep-archive@odin.ietf.org; Sat, 28 Dec 2002 21:03:14 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBT23EJ10883
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 28 Dec 2002 21:03:14 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA11364
	for <ieprep-web-archive@ietf.org>; Sat, 28 Dec 2002 20:56:37 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBT213J10846;
	Sat, 28 Dec 2002 21:01:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBT20tJ10821
	for <ieprep@optimus.ietf.org>; Sat, 28 Dec 2002 21:00:55 -0500
Received: from bells.cs.ucl.ac.uk (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id UAA11351
	for <ieprep@ietf.org>; Sat, 28 Dec 2002 20:54:19 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.23762-0@bells.cs.ucl.ac.uk>; Sun, 29 Dec 2002 01:45:22 +0000
To: ieprep@ietf.org
Date: Sun, 29 Dec 2002 01:45:20 +0000
Message-ID: <15454.1041126320@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Subject: [Ieprep] NATs
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


lets provide a more full picture.  NATs originally came out when 
address space was indeed scarce, and before CIDR took off.  so they
were not originally a product of hyped marketing.

NATs continued proliferation would suggest that there are a variety
of reasons for their existance.  I'm not sure this is the venue to
debate the usefullness of NATs.  they are painful, i can't stand them, 
but they exist and they aren't going to go away.  as for IPv6 being
the answer for address depletion, i'll jump on that bandwagon when i
see ISPs (beyond the ones i can count on my hand) offer it as a
*production* level service.

-ken

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Sat Dec 28 21:15:15 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA11635
	for <ieprep-archive@odin.ietf.org>; Sat, 28 Dec 2002 21:15:14 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBT2LLf11872
	for ieprep-archive@odin.ietf.org; Sat, 28 Dec 2002 21:21:21 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBT2LLJ11869
	for <ieprep-web-archive@optimus.ietf.org>; Sat, 28 Dec 2002 21:21:21 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA11626
	for <ieprep-web-archive@ietf.org>; Sat, 28 Dec 2002 21:14:43 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBT2K5J11839;
	Sat, 28 Dec 2002 21:20:05 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBT2G9J11724
	for <ieprep@optimus.ietf.org>; Sat, 28 Dec 2002 21:16:09 -0500
Received: from bells.cs.ucl.ac.uk (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with SMTP id VAA11544
	for <ieprep@ietf.org>; Sat, 28 Dec 2002 21:09:32 -0500 (EST)
Received: from sonic.cs.ucl.ac.uk by bells.cs.ucl.ac.uk with local SMTP 
          id <g.24886-0@bells.cs.ucl.ac.uk>; Sun, 29 Dec 2002 02:12:26 +0000
To: "Nguyen, An" <nguyena@ncs.gov>
cc: ieprep@ietf.org
Subject: Re: [Ieprep] Summary of ieprep work and VoIP--Take 2
In-reply-to: Your message of "Mon, 23 Dec 2002 08:57:51 EST." <7F18415E4D63CB45BB9B3A591F68D12D02EF4C77@emshqs1.ncr.disa.mil>
Date: Sun, 29 Dec 2002 02:12:24 +0000
Message-ID: <15931.1041127944@cs.ucl.ac.uk>
From: Ken Carlberg <K.Carlberg@cs.ucl.ac.uk>
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>


> I am just curious. What is the transport-layer mix? 
> 90% ~ TCP 
> 10% ~ UDP and others 
> 
> With a ratio of, say, 80% for TCP and 20% for UDP and packet drop rates
> of 10-40%, would the link be more or less unusable for TCP traffic?

one needs to be careful in how things are phrased.  the mix of
protocols, and in particular the amount of TCP, gives an indication of
that portion of traffic that will automatically attempt to backoff and
avoid congestion (ie, TCP) versus the traffic that is not expected to
backoff (eg, UDP).  so whether the ratio is 90:10 or 50:50 is moot with
respect to the "usability" of TCP traffic.  the key number in the
equation is the *sustained* level of packet loss.

also, the ability of the TCP flow to operate effectively over high loss
(approaching 40%) is dependent on the implementation of an application
-- most commonly the buffer space it requires.  Telnet (ie, sending 1 or
two bytes of app data) is not an application that consumes a lot of
buffer space, hence it can generally operate well in the presence of
high packet loss sustained over long periods.  On the other hand, as an
example, there is a vendor that terminates a TCP connection when its
buffers overflow because it can't forward BGP tables to other BGP
speakers due to loss (less than 40%) sustained over a long period.

ie, one needs to add qualifying statements about loss and TCP.

-ken

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 30 10:54:33 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25898
	for <ieprep-archive@odin.ietf.org>; Mon, 30 Dec 2002 10:54:33 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBUG1P809263
	for ieprep-archive@odin.ietf.org; Mon, 30 Dec 2002 11:01:25 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUG1PJ09260
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 30 Dec 2002 11:01:25 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25895
	for <ieprep-web-archive@ietf.org>; Mon, 30 Dec 2002 10:54:01 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUFwfJ09137;
	Mon, 30 Dec 2002 10:58:41 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUFvUJ09103
	for <ieprep@optimus.ietf.org>; Mon, 30 Dec 2002 10:57:30 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25855
	for <ieprep@ietf.org>; Mon, 30 Dec 2002 10:50:06 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id EF27167105; Mon, 30 Dec 2002 10:56:31 -0500 (EST)
Date: Mon, 30 Dec 2002 10:53:13 -0500
Subject: Re: [Ieprep] Understanding Congestion
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v551)
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77AFD@mcl-its-exs01.mail.saic.com>
Message-Id: <CD64A512-1C0E-11D7-AD1B-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.551)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Thursday, Dec 26, 2002, at 08:20 America/Montreal, King, Kimberly S. 
wrote:
> For an alternate perspective on addresses,
>
> "Everyone who wants one is not able to get one
> Example: People's Republic of China
> 1.3 Billion people
> Order of magnitude growth in Internet usage year over year
> ~9M addresses in 1999
> ~16M addresses in 2001
> Do we simply assume that anyone who has not already asked never will?
> Africa, South America, India, Arab world..."
> http://www.isoc-au.org.au/Events/IPv6Direction.ppt

The above analysis is popular with IPv6 advocates, but is a deeply 
flawed analysis.

If one actually talks with APNIC (or with large Chinese customers) and 
asks
probing questions, what one finds out is that even China is able to get 
all
the IPv4 address space that it needs now and for many years into the 
future.

The popular claim that China "needs 70 Class-A networks today" also 
turns
out to be boldly wrong when one does careful analysis or talks with the
informed folks inside the Chinese administration.  The allegation that
"China needs 70 Class-A networks" has been popularised in China by a 
large
equipment vendor who has the commercial motivation of selling 
IPv6-capable
equipment.  It turns out to not be true today.  It might be true at 
some future point,
of course, since IP address space (either version) is finite.  And 
among the
many flaws in the above analysis is that address reclamation of IPv4 
address
*IS* happening (e.g. 36.0/8 used to belong to Stanford University and
has since been returned to IANA for allocation to registries when 
needed).

Ran
rja@extremenetworks.com

PS:	I've coded up more IPv6 implementations than anyone else on the 
planet,
including cisco's and the NRL implementation for BSD (which was the 
world's first).
Extreme has an IPv6 implementation and has had such for a while now.

PPS:	These views are my own and are in no way reflective of any position
of my employer.  Further, my comments in any IETF context are always 
made
as my own view, whether or not they agree with whatever my employer 
might
think.

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 30 10:56:39 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25922
	for <ieprep-archive@odin.ietf.org>; Mon, 30 Dec 2002 10:56:39 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBUG3WN09326
	for ieprep-archive@odin.ietf.org; Mon, 30 Dec 2002 11:03:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUG3VJ09323
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 30 Dec 2002 11:03:31 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25917
	for <ieprep-web-archive@ietf.org>; Mon, 30 Dec 2002 10:56:08 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUG12J09245;
	Mon, 30 Dec 2002 11:01:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUG0dJ09215
	for <ieprep@optimus.ietf.org>; Mon, 30 Dec 2002 11:00:39 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA25889
	for <ieprep@ietf.org>; Mon, 30 Dec 2002 10:53:15 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id 96D9767105; Mon, 30 Dec 2002 10:59:35 -0500 (EST)
Date: Mon, 30 Dec 2002 10:56:16 -0500
Subject: Re: [Ieprep] NATs
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v551)
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
To: Dave Crocker <dhc2@dcrocker.net>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <199883169.20021228124302@brandenburg.com>
Message-Id: <3AD0114D-1C0F-11D7-AD1B-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.551)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Saturday, Dec 28, 2002, at 15:43 America/Montreal, Dave Crocker 
wrote:
> I do not like NAT, either, but we need to be careful about dismissing 
> it
> simply as a hack. It responded to two, real-world issues:
>
> 1.  NAT environments has been more convenient for small networks than 
> was
> (is?) possible without NAT.  That's not an address space issue, but a
> network (and host) administration issue.
>
> 2.  In effect, NAT does for a network what common networking software 
> does
> for a host.  That is, it provides an integrated point of lower-level
> addressing for a collection of processes (machines.)  In effect, one 
> can
> call this a kind of addressing data compression.  Data compression is 
> a good
> way to increase virtual throughput, so in fact one can argue that NAT 
> does,
> in fact, achieve the same address scaling effect as adding new address
> space.

The above is on target and important for everyone here to understand.

A third bullet (albeit perhaps a bit cynical) is that:

3.  NAT demand was significantly increased by misleading marketing from
     equipment vendors trying to generate demand for equipment that had
     NAT capability.  Once the equipment was in hand, users creatively 
found
     many novel ways to deploy the NAT capability.


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 30 11:06:38 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA26093
	for <ieprep-archive@odin.ietf.org>; Mon, 30 Dec 2002 11:06:37 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBUGDU910281
	for ieprep-archive@odin.ietf.org; Mon, 30 Dec 2002 11:13:30 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUGDUJ10278
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 30 Dec 2002 11:13:30 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA26088
	for <ieprep-web-archive@ietf.org>; Mon, 30 Dec 2002 11:06:06 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUGC3J10222;
	Mon, 30 Dec 2002 11:12:03 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUGB6J10194
	for <ieprep@optimus.ietf.org>; Mon, 30 Dec 2002 11:11:06 -0500
Received: from [140.32.132.66] (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA26044
	for <ieprep@ietf.org>; Mon, 30 Dec 2002 11:03:38 -0500 (EST)
Received: from mail.nps.navy.mil by [140.32.132.66]
          via smtpd (for [132.151.6.1]) with ESMTP; Mon, 30 Dec 2002 08:01:21 -0800
Received: from ziggy ([131.120.179.220])
	by capella.nps.navy.mil (8.12.2/8.12.2) with ESMTP id gBUG1W93019251;
	Mon, 30 Dec 2002 08:01:33 -0800 (PST)
Subject: RE: [Ieprep] Understanding Congestion
From: Rex Buddenberg <budden@nps.navy.mil>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
Cc: "'Bill Woodcock'" <woody@pch.net>, "Nguyen, An" <nguyena@ncs.gov>,
        "'ieprep@ietf.org'" <ieprep@ietf.org>
In-Reply-To: 
	<B8030EB94AF1D51196D70002A589D64207E77AFD@mcl-its-exs01.mail.saic.com>
References: 
	<B8030EB94AF1D51196D70002A589D64207E77AFD@mcl-its-exs01.mail.saic.com>
Content-Type: text/plain
Content-Transfer-Encoding: 7bit
Message-Id: <1040927404.1591.176.camel@ziggy>
Mime-Version: 1.0
X-Mailer: Ximian Evolution 1.0.8 (1.0.8-10) 
Date: 30 Dec 2002 07:55:40 -0800
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit

Kimberly,

Bill is basically right, if a bit low on the PH scale for my taste. NAT
boxes were a kluge to allow a unit (community college in my experience)
to shift ISPs (which required a wholesale change in IP addresses)
without visiting every office. 
 
The symptom of Internet scale problems were the _perceived_ exhaustion
of IP addresses (greatly exacerbated by the tendency of each requester
to ask for a Class B address regardless of real needs).  


The IPng exercises showed us that something like <1% of the addresses
assigned were actually being used.  Enter things like CIDR and DHCP
which enormously improved the efficiency -- and postponed the need for a
larger address space for at least a decade.  (NAT is known as IP
Masquerade in the Linux world for those wishing to track down how to use
it at home).

What was really hurting in the Internet a decade ago was the inefficient
assignment of routes. Each address block  showed up in a lookup table in
the routers' RAM.  This was both devouring RAM at increasing rates and
causing lookup inefficiencies -- something routers do for _every_
datagram through the box.  Classless InterDomain Routing (CIDR) pretty
much repaired this problem through route aggregation.  Of the facets of
this note, CIDR is the one tool that made routers work better, hence has
an impact on congestion.  

Your comment (below) sounds more down DHCP's alley than NAT -- not all
of those addresses will be in use at any point in time. DHCP allows you
to scavenge and reuse.  While routers like addresses in tidy blocks,
DHCP is essentially the sysadmin's sanity saver. (Note that the phone
company has been doing an analog for years: they tend to provision one
port on the switch for every 10 lines in the local loop on the
assumption that no more than 10% of the phones will be in use at a
single time).  

Eventually, IPv6 will indeed be necessary.  My guess is that the
cellphone and cable modem folks will hit the wall first.  Most OSs
(including those in routers) know how to handle it today but it's not
widely used yet.  

Accounting.  Counting IP address demand by counting human noses is a
crude way to do business.  In the two-decade-old timeshare days the
number of hosts was much less than the number of users so that method
was not realistic then.  Today, it's equally hokey because we're
increasingly assigning IP addresses to wholly inanimate objects like
unmanned sensors.  So the number of IP addresses may very well exceed
the number of warm human bodies.  I'd expect this to be the case in
spades in the emergency services world where fire, chemical, bio and a
host of other sensors will be wholly unmanned but have e-mail user
agents in them to send you mail when they detect what they're designed
for ... so they need addresses.  (And I wouldn't want to think about
managing that with a bunch of NAT boxes in the way;)






On Thu, 2002-12-26 at 05:20, King, Kimberly S. wrote:
> Bill said:
> >NATs are a hack which equipment vendors created 
> >in response
> >to artificial scarcity of IP addresses which marketing idiots 
> >in some ISPs
> >dreamed up as a way of getting back towards profitability after they
> >priced transit too low ...
> 
> For an alternate perspective on addresses,
> 
> "Everyone who wants one is not able to get one
> Example: People's Republic of China
> 1.3 Billion people
> Order of magnitude growth in Internet usage year over year
> ~9M addresses in 1999
> ~16M addresses in 2001
> Do we simply assume that anyone who has not already asked never will? 
> Africa, South America, India, Arab world..."
> http://www.isoc-au.org.au/Events/IPv6Direction.ppt
> 
> Kimberly
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep


_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 30 11:18:41 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA26283
	for <ieprep-archive@odin.ietf.org>; Mon, 30 Dec 2002 11:18:41 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBUGPY910774
	for ieprep-archive@odin.ietf.org; Mon, 30 Dec 2002 11:25:34 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUGPXJ10771
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 30 Dec 2002 11:25:33 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA26279
	for <ieprep-web-archive@ietf.org>; Mon, 30 Dec 2002 11:18:10 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUGO2J10676;
	Mon, 30 Dec 2002 11:24:02 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUGNZJ10646
	for <ieprep@optimus.ietf.org>; Mon, 30 Dec 2002 11:23:35 -0500
Received: from mclmx.mail.saic.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA26256
	for <ieprep@ietf.org>; Mon, 30 Dec 2002 11:16:11 -0500 (EST)
Received: from mcl-its-ieg01.mail.saic.com by mclmx.mail.saic.com for ieprep@ietf.org; Mon, 30 Dec 2002 11:18:32 -0500
Received: from mcl-its-exbh01.mail.saic.com ([149.8.64.11])
 by mcl-its-ieg01.mail.saic.com (NAVGW 2.5.2.17) with SMTP id M2002123011190217263
 for <ieprep@ietf.org>; Mon, 30 Dec 2002 11:19:02 -0500
Received: by mcl-its-exbh01.mail.saic.com with Internet Mail Service (5.5.2653.19)
	id <ZVCV01DJ>; Mon, 30 Dec 2002 11:20:42 -0500
Message-Id: <B8030EB94AF1D51196D70002A589D64207E77B1E@mcl-its-exs01.mail.saic.com>
From: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
To: "'ieprep@ietf.org'" <ieprep@ietf.org>
Subject: RE: [Ieprep] NATs
Date: Mon, 30 Dec 2002 11:18:55 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2653.19)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Okay, what have we learned?

NATs exist.

Their developers claimed lack of address 
space was a motivation.

Different people disagree about whether 
these developers were able to foresee 
problems in address space or whether 
they lied, whether analyses are flawed 
about the lack of address space for 
China, or whether more than a handful 
of ISPs will ever implement IPv6. 

So, what's our IEPREP perspective?  
NATs exist.  NATs mean hundreds of 
user sessions are potential behind a 
single /32.  They can cause problems 
for applications such as VoIP.  
Workarounds exist.  Marketing hype 
exists and people use things in ways 
their creators didn't envision.

What ieprep problem were we trying 
to examine?  As I recall, we were 
trying to understand the nature of 
congestion on the Internet.

Kimberly

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 30 11:44:14 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA26663
	for <ieprep-archive@odin.ietf.org>; Mon, 30 Dec 2002 11:44:14 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBUGp7k12439
	for ieprep-archive@odin.ietf.org; Mon, 30 Dec 2002 11:51:07 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUGp7J12436
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 30 Dec 2002 11:51:07 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA26660
	for <ieprep-web-archive@ietf.org>; Mon, 30 Dec 2002 11:43:43 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUGn6J12386;
	Mon, 30 Dec 2002 11:49:06 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUGmFJ12359
	for <ieprep@optimus.ietf.org>; Mon, 30 Dec 2002 11:48:15 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA26595
	for <ieprep@ietf.org>; Mon, 30 Dec 2002 11:40:50 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP
	id CA47767105; Mon, 30 Dec 2002 11:47:16 -0500 (EST)
Date: Mon, 30 Dec 2002 11:43:57 -0500
Subject: Re: [Ieprep] NATs
Content-Type: text/plain; charset=US-ASCII; format=flowed
Mime-Version: 1.0 (Apple Message framework v551)
Cc: "'ieprep@ietf.org'" <ieprep@ietf.org>
To: "King, Kimberly  S." <KIMBERLY.S.KING@saic.com>
From: RJ Atkinson <rja@extremenetworks.com>
In-Reply-To: <B8030EB94AF1D51196D70002A589D64207E77B1E@mcl-its-exs01.mail.saic.com>
Message-Id: <E400A443-1C15-11D7-AD1B-00039357A82A@extremenetworks.com>
Content-Transfer-Encoding: 7bit
X-Mailer: Apple Mail (2.551)
Content-Transfer-Encoding: 7bit
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


On Monday, Dec 30, 2002, at 11:18 America/Montreal, King, Kimberly S. 
wrote:
> So, what's our IEPREP perspective?
> NATs exist.  NATs mean hundreds of user sessions are potential
> behind a  single /32.  They can cause problems for applications
> such as VoIP.

I take several things away from this discussion.

1) If one wants an application to have the greatest deployability,
	then one should design that application to work well through
	an unexpected or unknown NAT device.
2) Voice might not be the best application to try to make work through
	an unknown or unexpected NAT device (i.e. one where one cannot
	preconfigure stuff to facilitate RTP through the NAT).
3) NATs don't really provide any security properties by themselves,
	so folks ought not deploy a security architecture based on the
	assumption that "NATs provide security".

Ran

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 30 13:39:54 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29472
	for <ieprep-archive@odin.ietf.org>; Mon, 30 Dec 2002 13:39:54 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBUIkn719763
	for ieprep-archive@odin.ietf.org; Mon, 30 Dec 2002 13:46:49 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUIkmJ19760
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 30 Dec 2002 13:46:48 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29469
	for <ieprep-web-archive@ietf.org>; Mon, 30 Dec 2002 13:39:23 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUIiVJ19700;
	Mon, 30 Dec 2002 13:44:32 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUIhBJ19638
	for <ieprep@optimus.ietf.org>; Mon, 30 Dec 2002 13:43:11 -0500
Received: from gnat.inet.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA29300
	for <ieprep@ietf.org>; Mon, 30 Dec 2002 13:35:44 -0500 (EST)
Received: from extremenetworks.com (unknown [10.18.3.100])
	by gnat.inet.org (Postfix) with ESMTP id BAB7467105
	for <ieprep@ietf.org>; Mon, 30 Dec 2002 13:42:11 -0500 (EST)
Date: Mon, 30 Dec 2002 13:38:52 -0500
Mime-Version: 1.0 (Apple Message framework v551)
Content-Type: text/plain; charset=US-ASCII; format=flowed
From: RJ Atkinson <rja@extremenetworks.com>
To: ieprep <ieprep@ietf.org>
Content-Transfer-Encoding: 7bit
Message-Id: <F16C36CF-1C25-11D7-AD1B-00039357A82A@extremenetworks.com>
X-Mailer: Apple Mail (2.551)
Content-Transfer-Encoding: 7bit
Subject: [Ieprep] Fwd:  RTP codec enhancements
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>
Content-Transfer-Encoding: 7bit
Content-Transfer-Encoding: 7bit


	Both of these enhancements seem very directly relevant
to making VoIP work robustly in an IEprep context that includes
VoIP.

Cheers,

Ran

Begin forwarded message:
> From: The IESG <iesg-secretary@ietf.org>
> Date: Mon Dec 30, 2002  13:28:34 America/Montreal
> To: IETF-Announce: ;
> Cc: RFC Editor <rfc-editor@isi.edu>, Internet Architecture Board 
> <iab@iab.org>, avt@ietf.org
> Subject: Protocol Action: RTP Payload Format for Enhanced Variable 
> Rate Codecs (EVRC) and Selectable Mode Vocoders SMV to Proposed 
> Standard
> Message-Id: <200212301828.NAA29111@ietf.org>
> Precedence: bulk
>
> The IESG has approved the Internet-Draft 'RTP Payload Format for
> Enhanced Variable Rate Codecs (EVRC) and Selectable Mode Vocoders SMV'
> <draft-ietf-avt-evrc-smv-03.txt> as a Proposed Standard.  This document
> is the product of the Audio/Video Transport Working Group.
>
> The IESG contact persons are Scott Bradner and Allison Mankin.

_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



From mailnull@www1.ietf.org  Mon Dec 30 15:47:15 2002
Received: from www1.ietf.org (ietf.org [132.151.1.19] (may be forged))
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01583
	for <ieprep-archive@odin.ietf.org>; Mon, 30 Dec 2002 15:47:15 -0500 (EST)
Received: (from mailnull@localhost)
	by www1.ietf.org (8.11.6/8.11.6) id gBUKsDS27017
	for ieprep-archive@odin.ietf.org; Mon, 30 Dec 2002 15:54:13 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUKsDJ27014
	for <ieprep-web-archive@optimus.ietf.org>; Mon, 30 Dec 2002 15:54:13 -0500
Received: from www1.ietf.org (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01575
	for <ieprep-web-archive@ietf.org>; Mon, 30 Dec 2002 15:46:44 -0500 (EST)
Received: from www1.ietf.org (localhost.localdomain [127.0.0.1])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUKnAJ26886;
	Mon, 30 Dec 2002 15:49:10 -0500
Received: from ietf.org (odin.ietf.org [132.151.1.176])
	by www1.ietf.org (8.11.6/8.11.6) with ESMTP id gBUKmZJ26844
	for <ieprep@optimus.ietf.org>; Mon, 30 Dec 2002 15:48:35 -0500
Received: from mailgate.pit.comms.marconi.com (ietf-mx.ietf.org [132.151.6.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA01481
	for <ieprep@ietf.org>; Mon, 30 Dec 2002 15:41:06 -0500 (EST)
Received: from mailman.pit.comms.marconi.com (mailman.pit.comms.marconi.com [169.144.2.12])
	by mailgate.pit.comms.marconi.com (8.9.3/8.9.3) with ESMTP id PAA21448;
	Mon, 30 Dec 2002 15:44:12 -0500 (EST)
Received: from whq-msgrtr-01.pit.comms.marconi.com (whq-msgrtr-01.pit.comms.marconi.com [169.144.2.221])
	by mailman.pit.comms.marconi.com (8.9.3/8.9.3) with ESMTP id PAA27924;
	Mon, 30 Dec 2002 15:44:14 -0500 (EST)
Received: by whq-msgrtr-01.pit.comms.marconi.com with Internet Mail Service (5.5.2650.21)
	id <ZVACFYWM>; Mon, 30 Dec 2002 15:44:13 -0500
Message-ID: <313680C9A886D511A06000204840E1CF030B5C84@whq-msgusr-02.pit.comms.marconi.com>
From: "Rosen, Brian" <Brian.Rosen@marconi.com>
To: "'RJ Atkinson'" <rja@extremenetworks.com>, ieprep <ieprep@ietf.org>
Subject: RE: [Ieprep] Fwd:  RTP codec enhancements
Date: Mon, 30 Dec 2002 15:44:11 -0500
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: ieprep-admin@ietf.org
Errors-To: ieprep-admin@ietf.org
X-BeenThere: ieprep@ietf.org
X-Mailman-Version: 2.0.12
Precedence: bulk
List-Unsubscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=unsubscribe>
List-Id: Internet Emergency Preparedness Working Group <ieprep.ietf.org>
List-Post: <mailto:ieprep@ietf.org>
List-Help: <mailto:ieprep-request@ietf.org?subject=help>
List-Subscribe: <https://www1.ietf.org/mailman/listinfo/ieprep>,
	<mailto:ieprep-request@ietf.org?subject=subscribe>

Actually, I'd think the opposite - don't assume any compression.
It may be that it's impractical to force 3G wireless systems to
deliver uncompressed voice when connected to emergency systems.
They do today of course, but it may be that future architectures
favor direct connection of the compressed stream into the emergency
center.  However, we surely don't advocate forcing every VoIP device
to implement a proprietary codec.  

Of course, forcing every mobile phone system in the world to have 
a common codec would be very interesting ;)

Brian

> -----Original Message-----
> From: RJ Atkinson [mailto:rja@extremenetworks.com]
> Sent: Monday, December 30, 2002 1:39 PM
> To: ieprep
> Subject: [Ieprep] Fwd: RTP codec enhancements
> 
> 
> 
> 	Both of these enhancements seem very directly relevant
> to making VoIP work robustly in an IEprep context that includes
> VoIP.
> 
> Cheers,
> 
> Ran
> 
> Begin forwarded message:
> > From: The IESG <iesg-secretary@ietf.org>
> > Date: Mon Dec 30, 2002  13:28:34 America/Montreal
> > To: IETF-Announce: ;
> > Cc: RFC Editor <rfc-editor@isi.edu>, Internet Architecture Board 
> > <iab@iab.org>, avt@ietf.org
> > Subject: Protocol Action: RTP Payload Format for Enhanced Variable 
> > Rate Codecs (EVRC) and Selectable Mode Vocoders SMV to Proposed 
> > Standard
> > Message-Id: <200212301828.NAA29111@ietf.org>
> > Precedence: bulk
> >
> > The IESG has approved the Internet-Draft 'RTP Payload Format for
> > Enhanced Variable Rate Codecs (EVRC) and Selectable Mode 
> Vocoders SMV'
> > <draft-ietf-avt-evrc-smv-03.txt> as a Proposed Standard.  
> This document
> > is the product of the Audio/Video Transport Working Group.
> >
> > The IESG contact persons are Scott Bradner and Allison Mankin.
> 
> _______________________________________________
> Ieprep mailing list
> Ieprep@ietf.org
> https://www1.ietf.org/mailman/listinfo/ieprep
> 
_______________________________________________
Ieprep mailing list
Ieprep@ietf.org
https://www1.ietf.org/mailman/listinfo/ieprep



