From owner-l2tp@ipsec.org  Wed May  3 17:57:35 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA11864
	for <l2tpext-archive@odin.ietf.org>; Wed, 3 May 2000 17:57:33 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id LAA17434; Wed, 3 May 2000 11:17:16 -0700 (PDT)
Received: from amberex01.ambernetworks.com (outgate01.ambernetworks.com [207.20.187.199]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id LAA17430 for <l2tp@ipsec.org>; Wed, 3 May 2000 11:15:02 -0700 (PDT)
Received: by amberex01.ambernetworks.com with Internet Mail Service (5.5.2448.0)
	id <JBQ2T1WV>; Wed, 3 May 2000 13:39:21 -0700
Message-ID: <405CAE450410D311B26C0050047326DC4F51E5@amberex01.ambernetworks.com>
From: Danny McPherson <Danny@ambernetworks.com>
To: "'l2tp@ipsec.org'" <l2tp@ipsec.org>
Subject: L2TP Circuit Emulation Services Extension
Date: Wed, 3 May 2000 13:39:21 -0700 
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2448.0)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

Comments welcome.

-danny

[]
L2TP Working Group                                       Danny McPherson
Internet Draft                                            Ravi Bail Bhat
                                                          Andy Koscinski
                                                              Chi Fai Ho
                                                          Amber Networks
draft-mcpherson-l2tpext-es-00.txt                               May 2000




               L2TP Circuit Emulation Services Extension


Status of this Memo

   This document is an Internet-Draft and is in full conformance with
   all provisions of Section 10 of RFC2026.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF), its areas, and its working groups.  Note that
   other groups may also distribute working documents as Internet-
   Drafts.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   The list of current Internet-Drafts can be accessed at
   http://www.ietf.org/ietf/1id-abstracts.txt

   The list of Internet-Draft Shadow Directories can be accessed at
   http://www.ietf.org/shadow.html.


Abstract

   The Layer 2 Tunneling Protocol (L2TP) [RFC2661] defines a mechanism
   for tunneling PPP sessions.  This document proposes mechanisms by
   which the L2TP tunneling scheme can be used to provide circuit
   emulation support for layer 2 circuits (i.e. Frame Relay or ATM), as
   well as TDM circuits (i.e. DS1 or DS3).  L2TP is used to provide
   tunneling support and each circuit is encapsulated over a session
   inside the Tunnel.

   An Encapsulation Services Protocol [RefESP] is used on top of the
   individual L2TP sessions to support the circuit emulation of layer
   2 VCs or TDM circuits.  The purpose of this document is to explain
   the L2TP modifications done to facilitate support of circuit 
   emulation services, as well as to define the additional AVPs that can
   be used to provide the service.









Specification of Requirements

   The key words "MUST", "MUST NOT", "REQUIRED", "SHALL", "SHALL NOT",
   "SHOULD", "SHOULD NOT", "RECOMMENDED",  "MAY", and "OPTIONAL" in this
   document are to be interpreted as described in [RFC2119].


Table of Contents

      1. Introduction
      2. Topology Model
         2.1 Modified L2TP Topooofy Model
      3. Protocol Overview
      4. Proposed Protocol Operation
         4.1 Service Type AVP Format
         4.2 Proposed Sub-Address in ICRQ
      5. Quality of Service Considerations
      6. Security Considerations
      7. Intellectual Property Considerations
      8. References
      9. Authors' Addresses


1. Introduction

   The Layer 2 Tunneling Protocol (L2TP) [RFC2661] defines a mechanism
   for tunneling PPP sessions.  This document describes mechanisms by
   which L2TP tunneling scheme can be used to provide circuit emulation
   support for layer 2 circuits (i.e. Frame Relay or ATM), as well as
   TDM circuits (i.e. DS1 or DS3).  L2TP is used to provide tunneling
   support and each circuit is encapsulated over a session inside the
   Tunnel.

   An Encapsulation Services Protocol [RefESP] is used on top of the
   individual L2TP sessions to support the circuit emulation of
   layer 2 Virtual Circuits or TDM circuits.  The purpose of this
   document is to explain the L2TP modifications done to facilitate
   support of circuit emulation services, as well as to define the
   additional AVPs required to provide the service.


2. Topology Model

   The current L2TP model assumes a client/server architecture between
   the LAC and LNS.  To support encapsulations services, a symmetric
   LAC-to-LAC model is proposed.  In this model, a tunnel (with one or
   more sessions sessions) can be established between two LACs.

   A tunnel is setup for a particular service or set of services (e.g.
   FR_GOLD) between a pair of LACs that support circuit emulation
   services.  The Service Type is sent to the peer LAC via a newly
   defined vendor- specific AVP.  Then for each connection within the
   service group (e.g. FR_GOLD), a session is initiated (possible from
   either side).  The connection identifiers for the two legs of the
   connection (e.g.  the Frame Relay interface and the Frame-Relay DLCI)
   are sent in the Sub-Address AVP.




2.1. Modified L2TP Topolgy Model (LAC/LAC)


   The following diagram depicts a typical L2TP LAC/LAC scenario.  The
   goal is to tunnel lower-layer (layer 1 or layer 2) frames from the
   one LAC to the another LAC.


       FR-------+             +-------------+            +---- FR
                |----         |             |        ----|
                      LAC1 ---| Internet    |--- LAC2
                |----         |             |        ----|
       TDM------+             +-------------+            +---- TDM
                                     |
                                     |
                                    LAC3
                                    |  |
                                 ---+  +---
                                 |        |
                                 |        |
                                FR       TDM




3. Protocol Overview

   As shown in the figure below, the L2TP protocol structure is modified
   to carry any protocol (Note: The current specification only
   identifies support for PPP frames). An encapsulation protocol with
   both data and control messages is carried over L2TP protocol.

                               +-----------------------+
      +-------------------+    | any protocol control  |
      | any payload       |    | messages              |
      +-------------------+    +-----------------------+
      | L2TP Data Messages|    | L2TP Control Messages |
      +-------------------+    +-----------------------+
      | L2TP Data Channel |    | L2TP Control Channel  |
      | (unreliable)      |    | (reliable)            |
      +------------------------------------------------+
      |      Packet Transport (UDP, FR, ATM, etc.)     |
      +------------------------------------------------+




4. Proposed Protocol Operation

   Encapsulation Services (ES) enable an IP network to support emulation
   of connection-oriented networks like FR and ATM, as well as TDM
   circuit emulation.  ES supports different Emulation Types and also
   supports multiple service profiles. The Emulation Type and the
   Service Type are represented by an optional AVP "Service Type". The
   AVP MUST be present in both SCCRQ and SCCRP messages and MUST match
   one another, otherwise the tunnel is dropped.

   At least one L2TP tunnel is opened for every Emulation Type. Multiple
   tunnels may be opened for each Emulation Type if different Service
   Types are needed.  Each circuit (e.g. DLCI in FR) is mapped into a
   L2TP session and carried transparently to the other end.  During the
   session request, the endpoint connection identifiers are transported
   in the Sub-Address AVP.

   Once the L2TP session is established, the upper layer encapsulation
   protocol MAY exchange additional information to complete the circuit
   emulation establishment.  After that, the LACs start to transmit data
   between one another.


4.1. Service Type AVP Format

   The Service Type AVP MUST be present in SCCRQ and SCCRP message.  This
   AVP is used to inform the tunnel peer that a specific Service Type
   (e.g. FR_GOLD) is used.

   Vendor ID = 4741 Attribute Type = 1

       0                   1                   2                   3
       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |0|0|0|0|0|0|  Length           |            4741               |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |                1              |Service Type (arbitrary length)|
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+


   This AVP is encoded as a Vendor ID of 4741, which reflects Amber
   Networks, the initial developer of this specification. The Vendor ID
   SHOULD be changed to "0" and an official attribute value chosen, if
   this specification advances on the standard tracks. The Attribute
   Type is the 16 bit quantity "1". This AVP MAY be hidden and is
   mandatory (mandatory bit = 1). The L2TP peer is indicating that 
   resources adequate to the Service Type identified by the AVP are 
   required.

   In the event that the peer does not accept the requested Service
   Type, a stopCCN is returned to the peer. Such stopCCN message will
   include the Service Type AVP as provided in the message that caused
   the stopCCN.


4.2. Proposed Sub-Address in ICRQ

   The Sub-Address AVP, Attribute Type 23, encodes additional connection
   identifier information.

   The Attribute Value field for this AVP has the following format:

       0                   1                   2                   3
       0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      | Sub-Address ... (arbitrary number of octets)                  |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

   The Sub-Address is an opaque sequence of octets transmitted
   transparently by the network. The tunnel end points MUST, a priori,
   understand the meaning of the value in this AVP.

   The general format of the information in this AVP is:

   1.  Calling party Sub-Address (iterations of: InfoType, Length,
       Connection Information, e.g. interface and DLCI for FR);

   2.  Called party Sub-Address (iterations of: InfoType, Length,
       Connection Information, e.g. interface and DLCI for FR);


   This AVP MAY be hidden (the H-bit may be 0 or 1). The M-bit for this
   AVP MUST be set to 1.  The Length (before hiding) of this AVP is 6
   octets plus the length of the Sub-Address.


5. Quality of Service Considerations

   Quality of Service (QoS) is a necessity for circuit emulation
   applications.  The QoS mechanisms such as those proposed for L2TP in
   [L2TP-DS] and [L2TP-MPLS] should be considered.  Additional
   discussion of this topic is beyond the scope of this document.


6. Security Considerations

   This document does not introduce security considerations beyond those
   listed in [RFC1661].


7. Intellectual Property Considerations

   Amber Networks may seek patent or other intellectual property
   protection for some of all of the technologies disclosed in this
   document. If any standards arising from this document are or become
   protected by one or more patents assigned to Amber Networks, Amber
   intends to disclose those patents and license them on reasonable and
   non-discriminatory terms.


8. References

      [RFC2661]   W.M. Townsley, A. Valencia, A. Rubens, G. Singh Pall,
                  G. Zorn, B. Palter, "Layer Two Tunneling Protocol
                  (L2TP)", RFC 2661, August 1999.

      [RefESP]    McPherson, D., et al., "Encapsulation Services 
                  Protocol (ESP)", "Work in Progress".

      [RFC2119]   Bradner, S., "Key words for use in RFCs to Indicate
                  Requirement Levels", BCP 14, RFC 2119, March 1997.

      [L2TP-DS]   Calhoun, P., Peirce, K., "Layer Two Tunneling Protocol
                  "L2TP" IP Diferentiated Services Extension", "Work in
                   Progress".

      [L2TP-MPLS] Calhoun, P., Peirce, K., "Layer Two Tunneling Protocol 
                  "L2TP" Multi-Protocol Label Switching Extension", 
                  "Work in Progress".





9. Authors' Addresses

   Danny McPherson
   Amber Networks
   2465 Augustine Drive
   Santa Clara, CA  95054
   Phone: +1 408.486.6336
   Email: danny@ambernetworks.com

   Ravi Bail Bhat
   Amber Networks
   2465 Augustine Drive
   Santa Clara, CA  95054
   Phone: +1 408.845.5597
   Email: rbhat@ambernetworks.com

   Andy Koscinski
   Amber Networks
   2465 Augustine Drive
   Santa Clara, CA  95054
   Phone: +1 408.845.5536
   Email: andyk@ambernetworks.com

   Chi Fai Ho
   Amber Networks
   2465 Augustine Drive
   Santa Clara, CA  95054
   Phone: +1 408.845.5547
   Email: ho@ambernetworks.com
































From owner-l2tp@ipsec.org  Tue May  9 07:53:29 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id HAA15845
	for <l2tpext-archive@odin.ietf.org>; Tue, 9 May 2000 07:53:28 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id BAA01813; Tue, 9 May 2000 01:13:42 -0700 (PDT)
Received: from ietf.org (odin.ietf.org [132.151.1.176]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id BAA01809 for <l2tp@ipsec.org>; Tue, 9 May 2000 01:13:05 -0700 (PDT)
Received: from CNRI.Reston.VA.US (localhost [127.0.0.1])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA13908;
	Tue, 9 May 2000 06:36:24 -0400 (EDT)
Message-Id: <200005091036.GAA13908@ietf.org>
Mime-Version: 1.0
Content-Type: Multipart/Mixed; Boundary="NextPart"
To: IETF-Announce:;
Cc: l2tp@ipsec.org
From: Internet-Drafts@ietf.org
Reply-to: Internet-Drafts@ietf.org
Subject: I-D ACTION:draft-ietf-l2tpext-atmext-02.txt
Date: Tue, 09 May 2000 06:36:23 -0400
Sender: owner-l2tp@ipsec.org
Precedence: bulk

--NextPart

A New Internet-Draft is available from the on-line Internet-Drafts directories.
This draft is a work item of the Layer Two Tunneling Protocol Extensions Working Group of the IETF.

	Title		: Layer Two Tunnelling Protocol : ATM access network 
                          extensions.
	Author(s)	: Y. T'Joens, P. Crivellari,  B. Sales
	Filename	: draft-ietf-l2tpext-atmext-02.txt
	Pages		: 19
	Date		: 08-May-00
	
L2TP [RFC2661] specifies a protocol which permits the tunnelling of
the link layer of PPP over packet based networks, to support remote
access (mainly) by ISDN and PSTN networks. This document augments the
procedures described in [RFC2661] to further support ATM SVC or PVC
based access networks. The extensions defined within this document
allow for asymmetric bi-directional call establishment and service
selection in the ATM access network.

A URL for this Internet-Draft is:
http://www.ietf.org/internet-drafts/draft-ietf-l2tpext-atmext-02.txt

Internet-Drafts are also available by anonymous FTP. Login with the username
"anonymous" and a password of your e-mail address. After logging in,
type "cd internet-drafts" and then
	"get draft-ietf-l2tpext-atmext-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-l2tpext-atmext-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:	<20000508103949.I-D@ietf.org>

ENCODING mime
FILE /internet-drafts/draft-ietf-l2tpext-atmext-02.txt

--OtherAccess
Content-Type: Message/External-body;
	name="draft-ietf-l2tpext-atmext-02.txt";
	site="ftp.ietf.org";
	access-type="anon-ftp";
	directory="internet-drafts"

Content-Type: text/plain
Content-ID:	<20000508103949.I-D@ietf.org>

--OtherAccess--

--NextPart--




From owner-l2tp@ipsec.org  Wed May 10 21:03:27 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA05677
	for <l2tpext-archive@odin.ietf.org>; Wed, 10 May 2000 21:03:26 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id OAA05583; Wed, 10 May 2000 14:15:50 -0700 (PDT)
Received: from dfssl.exchange.microsoft.com ([131.107.88.59]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id OAA05560 for <l2tp@ipsec.org>; Wed, 10 May 2000 14:14:53 -0700 (PDT)
Received: from 172.30.236.11 by dfssl.exchange.microsoft.com (InterScan E-Mail VirusWall NT); Wed, 10 May 2000 16:31:12 -0700 (Pacific Daylight Time)
Received: from SIT.platinum.corp.microsoft.com ([172.30.236.233]) by yuri.dns.microsoft.com with Microsoft SMTPSVC(5.0.2195.2023);
	 Wed, 10 May 2000 16:38:36 -0700
content-class: urn:content-classes:message
Subject: Incompatibility between Diffserv and PPP based tunneling protocols
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01BFBAD8.04458852"
Date: Wed, 10 May 2000 16:32:33 -0700
X-MimeOLE: Produced By Microsoft Exchange V6.0.4362.4
Message-ID: <19398D273324D3118A2B0008C7E9A5690681E0CB@SIT.platinum.corp.microsoft.com>
Thread-Topic: Incompatibility between Diffserv and PPP based tunneling protocols
Thread-Index: Ab+62L4imXIiRD7ISO+39RMZ4Hml9Q==
From: "Rajesh Sundaram" <rajeshsu@Exchange.Microsoft.com>
To: <diffserv@ietf.org>, <l2tp@ipsec.org>
X-OriginalArrivalTime: 10 May 2000 23:38:36.0234 (UTC) FILETIME=[DC5B5AA0:01BFBAD8]
Sender: owner-l2tp@ipsec.org
Precedence: bulk

This is a multi-part message in MIME format.

------_=_NextPart_001_01BFBAD8.04458852
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

Folks,

There seems to be an incompatibility between tunneling protocols that
encapsulate PPP (like L2TP and PPTP) and Diffserv PHBs.=20

Let's assume that there are two micro-flows of traffic within a single
L2TP tunnel, marked for different PHBs. The DSCP mark has to be
propagated from the original packet to the tunneled header at tunnel
ingress (as per section 2.5 of EF PHB). The restrictions on packet
re-order apply only to a packets belonging to a micro flow within a
particular PHB class. So, it is possible for packets belonging to
different PHBs to get re-ordered by DS aware nodes en-route to tunnel
egress.=20

This re-ordering breaks PPTP and L2TP. PPTP is required to discard
re-ordered packets in order to maintain the ordering semantics of PPP.
L2TP allows for the receiver to restore the original sequence of
re-ordered packets, but doing this negates the benefits of the PHB,
since the packet will potentially remain in the receiver's tunnel buffer
for a long time.=20

One way of solving this problem is to create multiple L2TP or PPTP
sessions and by dedicating one PHB for each session. Since both L2TP and
PPTP allow re-ordering across sessions, the packet re-order problem is
not as nasty.

Any suggestions on how this incompatibility can be addressed ?

Thanks,
Rajesh.

------_=_NextPart_001_01BFBAD8.04458852
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
6.0.4362.4">
<TITLE>Incompatibility between Diffserv and PPP based tunneling =
protocols</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/rtf format -->

<P><FONT SIZE=3D2 FACE=3D"Verdana">Folks,</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Verdana">There seems to be an incompatibility =
between tunneling protocols that encapsulate PPP (like L2TP and PPTP) =
and Diffserv PHBs. </FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Verdana">Let's assume that there are two =
micro-flows of traffic within a single L2TP tunnel, marked for different =
PHBs. The DSCP mark has to be propagated from the original packet to the =
tunneled header at tunnel ingress (as per section 2.5 of EF PHB). The =
restrictions on packet re-order apply only to a packets belonging to a =
micro flow within a particular PHB class. So, it is possible for packets =
belonging to different PHBs to get re-ordered by DS aware nodes en-route =
to tunnel egress. </FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Verdana">This re-ordering breaks PPTP and =
L2TP. PPTP is required to discard re-ordered packets in order to =
maintain the ordering semantics of PPP. L2TP allows for the receiver to =
restore the original sequence of re-ordered packets, but doing this =
negates the benefits of the PHB, since the packet will potentially =
remain in the receiver's tunnel buffer for a long time. </FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Verdana">One way of solving this problem is to =
create multiple L2TP or PPTP sessions and by dedicating one PHB for each =
session. Since both L2TP and PPTP allow re-ordering across sessions, the =
packet re-order problem is not as nasty.</FONT></P>

<P><FONT SIZE=3D2 FACE=3D"Verdana">Any suggestions on how this =
incompatibility can be addressed ?</FONT>
</P>

<P><FONT SIZE=3D2 FACE=3D"Verdana">Thanks,</FONT>

<BR><FONT SIZE=3D2 FACE=3D"Verdana">Rajesh.</FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01BFBAD8.04458852--


From owner-l2tp@ipsec.org  Wed May 10 22:04:41 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA07084
	for <l2tpext-archive@odin.ietf.org>; Wed, 10 May 2000 22:04:41 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id PAA05657; Wed, 10 May 2000 15:38:58 -0700 (PDT)
Received: from crufty.research.bell-labs.com (crufty.research.bell-labs.com [204.178.16.49]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id PAA05653 for <l2tp@ipsec.org>; Wed, 10 May 2000 15:38:32 -0700 (PDT)
Received: from zubin.dnrc.bell-labs.com ([135.180.130.56]) by crufty; Wed May 10 21:01:56 EDT 2000
Received: from dnrc.bell-labs.com (dhcp-6-168.pa.bell-labs.com [135.250.6.168])
	by zubin.dnrc.bell-labs.com (8.9.3/8.9.3) with ESMTP id VAA10331;
	Wed, 10 May 2000 21:01:52 -0400 (EDT)
Message-ID: <391A0754.FA3C92B2@dnrc.bell-labs.com>
Date: Wed, 10 May 2000 18:05:24 -0700
From: Grenville Armitage <gja@dnrc.bell-labs.com>
Organization: Bell Laboratories, Lucent Technologies
X-Mailer: Mozilla 4.7 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Rajesh Sundaram <rajeshsu@Exchange.Microsoft.com>
CC: diffserv@ietf.org, l2tp@ipsec.org
Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based tunneling 
 protocols
References: <19398D273324D3118A2B0008C7E9A5690681E0CB@SIT.platinum.corp.microsoft.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit


It isn't always the case that the outer DSCP (carried in the tunneling
IP header) has to be copied from the DSCP of the IP packet being
tunneled. In this case, when reordering of the tunnel's contents is
a bad thing, the tunneling header ought to be given some relatively
fixed DSCP (one that ensures 'appropriate' QoS across the tunnel's
length). This would ensure that the aggregate of all traffic
going through the tunnel is treated as a single non-reorderable 'flow'
by the DS routers along the tunnel's path.

cheers,
gja

> Rajesh Sundaram wrote:
> 
> Folks,
> 
> There seems to be an incompatibility between tunneling protocols that
> encapsulate PPP (like L2TP and PPTP) and Diffserv PHBs.
> 
> Let's assume that there are two micro-flows of traffic within a single L2TP
> tunnel, marked for different PHBs. The DSCP mark has to be propagated from the
> original packet to the tunneled header at tunnel ingress (as per section 2.5
> of EF PHB). The restrictions on packet re-order apply only to a packets
> belonging to a micro flow within a particular PHB class. So, it is possible
> for packets belonging to different PHBs to get re-ordered by DS aware nodes
> en-route to tunnel egress.
> 
> This re-ordering breaks PPTP and L2TP. PPTP is required to discard re-ordered
> packets in order to maintain the ordering semantics of PPP. L2TP allows for
> the receiver to restore the original sequence of re-ordered packets, but doing
> this negates the benefits of the PHB, since the packet will potentially remain
> in the receiver's tunnel buffer for a long time.
> 
> One way of solving this problem is to create multiple L2TP or PPTP sessions
> and by dedicating one PHB for each session. Since both L2TP and PPTP allow
> re-ordering across sessions, the packet re-order problem is not as nasty.
> 
> Any suggestions on how this incompatibility can be addressed ?
> 
> Thanks,
> Rajesh.

-- 
________________________________________________________________________
Grenville Armitage                    http://members.home.net/garmitage/
Bell Labs Research Silicon Valley


From owner-l2tp@ipsec.org  Wed May 10 22:13:20 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA07121
	for <l2tpext-archive@odin.ietf.org>; Wed, 10 May 2000 22:13:20 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id PAA05690; Wed, 10 May 2000 15:50:09 -0700 (PDT)
Received: from fcs-nt1.futsoft.com (fcs-nt1.futsoft.com [38.242.189.2]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id PAA05681 for <l2tp@ipsec.org>; Wed, 10 May 2000 15:49:42 -0700 (PDT)
Received: from sanjose.futsoft.com (unverified) by fcs-nt1.futsoft.com
 (Content Technologies SMTPRS 2.0.15) with SMTP id <B0000039139@fcs-nt1.futsoft.com>;
 Wed, 10 May 2000 18:02:11 -0700
Received: from rajeshs ([38.242.189.59])
	by sanjose.futsoft.com (8.9.3/8.8.7) with SMTP id RAA27983;
	Wed, 10 May 2000 17:07:47 -0700
Received: by localhost with Microsoft MAPI; Wed, 10 May 2000 18:25:15 -0700
Message-Id: <01BFBAAD.163F02C0.rajeshs@futsoft.com>
From: Rajesh Kumar <rajeshs@futsoft.com>
Reply-To: "rajeshs@futsoft.com" <rajeshs@futsoft.com>
To: "'Rajesh Sundaram'" <rajeshsu@Exchange.Microsoft.com>,
        "diffserv@ietf.org"
	 <diffserv@ietf.org>,
        "l2tp@ipsec.org" <l2tp@ipsec.org>
Subject: RE: Incompatibility between Diffserv and PPP based tunneling protocols
Date: Wed, 10 May 2000 18:25:14 -0700
Organization: Future Communications Software
X-Mailer: Microsoft Internet E-mail/MAPI - 8.0.0.4211
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit

If you turn off reordering and sequencing (as these are optional
in l2tp) and let tcp handle these issues, l2tp should work
seamlessly. tcp would handle the reliability issues for each
microflow. (I don't know much about pptp).
But this works only for tcp. How about others - like udp? - that's
unreliable service anyway, so why try to introduce reliability
at the link layer?

Does that sound workable?

-----Original Message-----
From:	Rajesh Sundaram [SMTP:rajeshsu@Exchange.Microsoft.com]
Sent:	Wednesday, May 10, 2000 4:33 PM
To:	diffserv@ietf.org; l2tp@ipsec.org
Subject:	Incompatibility between Diffserv and PPP based tunneling protocols

Folks,

There seems to be an incompatibility between tunneling protocols that
encapsulate PPP (like L2TP and PPTP) and Diffserv PHBs. 

Let's assume that there are two micro-flows of traffic within a single
L2TP tunnel, marked for different PHBs. The DSCP mark has to be
propagated from the original packet to the tunneled header at tunnel
ingress (as per section 2.5 of EF PHB). The restrictions on packet
re-order apply only to a packets belonging to a micro flow within a
particular PHB class. So, it is possible for packets belonging to
different PHBs to get re-ordered by DS aware nodes en-route to tunnel
egress. 

This re-ordering breaks PPTP and L2TP. PPTP is required to discard
re-ordered packets in order to maintain the ordering semantics of PPP.
L2TP allows for the receiver to restore the original sequence of
re-ordered packets, but doing this negates the benefits of the PHB,
since the packet will potentially remain in the receiver's tunnel buffer
for a long time. 

One way of solving this problem is to create multiple L2TP or PPTP
sessions and by dedicating one PHB for each session. Since both L2TP and
PPTP allow re-ordering across sessions, the packet re-order problem is
not as nasty.

Any suggestions on how this incompatibility can be addressed ?

Thanks,
Rajesh.
 << File: ATT00002.htm >> 


From owner-l2tp@ipsec.org  Wed May 10 23:10:29 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA08467
	for <l2tpext-archive@odin.ietf.org>; Wed, 10 May 2000 23:10:29 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id QAA05805; Wed, 10 May 2000 16:43:33 -0700 (PDT)
Received: from sales-finance.vpnet.com (mail.vpnet.com [209.1.117.16]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id QAA05800 for <l2tp@ipsec.org>; Wed, 10 May 2000 16:42:47 -0700 (PDT)
Received: by mail.vpnet.com with Internet Mail Service (5.0.1461.28)
	id <KRPJ0450>; Wed, 10 May 2000 19:07:23 -0700
Message-ID: <D899E9E27BE9D211842200805FA67B43923551@mail.vpnet.com>
From: Mahalingam Mani <Mani@vpnet.com>
To: Grenville Armitage <gja@dnrc.bell-labs.com>
Cc: diffserv@ietf.org, l2tp@ipsec.org
Subject: RE: [Diffserv] Incompatibility between Diffserv and PPP based tun
	neling  protocols
Date: Wed, 10 May 2000 19:07:21 -0700
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.0.1461.28)
Content-Type: text/plain;
	charset="windows-1252"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

What if the desired PHB is as required on a per-microflow basis - even over
the
tunnel? Especially the case with IKE tunnels where one may encounter
multiple
DS router hops in the tunnel?

This is an issue with IPSec (IKE tunnels) - on similar lines. More so
because
if reordering is more than a certain (RFC-)suggested (which is a window of
32
packets - or more, as implemented) number then it is flagged as a replay
attack.

The suggestion to split the two microflows into 2 different tunnels will
work,
however with thousands of tunnels to manage in an enterprise/carrier-class
VPN
setup this will become an administrative nightmare.

-mani
> -----Original Message-----
> From: Grenville Armitage [mailto:gja@dnrc.bell-labs.com]
> Sent: Wednesday, May 10, 2000 6:05 PM
> To: Rajesh Sundaram
> Cc: diffserv@ietf.org; l2tp@ipsec.org
> Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based
> tunneling protocols
> 
> 
> 
> It isn't always the case that the outer DSCP (carried in the tunneling
> IP header) has to be copied from the DSCP of the IP packet being
> tunneled. In this case, when reordering of the tunnel's contents is
> a bad thing, the tunneling header ought to be given some relatively
> fixed DSCP (one that ensures 'appropriate' QoS across the tunnel's
> length). This would ensure that the aggregate of all traffic
> going through the tunnel is treated as a single non-reorderable 'flow'
> by the DS routers along the tunnel's path.
> 
> cheers,
> gja
> 
> > Rajesh Sundaram wrote:
> > 
> > Folks,
> > 
> > There seems to be an incompatibility between tunneling 
> protocols that
> > encapsulate PPP (like L2TP and PPTP) and Diffserv PHBs.
> > 
> > Let's assume that there are two micro-flows of traffic 
> within a single L2TP
> > tunnel, marked for different PHBs. The DSCP mark has to be 
> propagated from the
> > original packet to the tunneled header at tunnel ingress 
> (as per section 2.5
> > of EF PHB). The restrictions on packet re-order apply only 
> to a packets
> > belonging to a micro flow within a particular PHB class. 
> So, it is possible
> > for packets belonging to different PHBs to get re-ordered 
> by DS aware nodes
> > en-route to tunnel egress.
> > 
> > This re-ordering breaks PPTP and L2TP. PPTP is required to 
> discard re-ordered
> > packets in order to maintain the ordering semantics of PPP. 
> L2TP allows for
> > the receiver to restore the original sequence of re-ordered 
> packets, but doing
> > this negates the benefits of the PHB, since the packet will 
> potentially remain
> > in the receiver's tunnel buffer for a long time.
> > 
> > One way of solving this problem is to create multiple L2TP 
> or PPTP sessions
> > and by dedicating one PHB for each session. Since both L2TP 
> and PPTP allow
> > re-ordering across sessions, the packet re-order problem is 
> not as nasty.
> > 
> > Any suggestions on how this incompatibility can be addressed ?
> > 
> > Thanks,
> > Rajesh.
> 
> -- 
> ______________________________________________________________
> __________
> Grenville Armitage                    
> http://members.home.net/garmitage/
> Bell Labs Research Silicon Valley
> 
> _______________________________________________
> diffserv mailing list
> diffserv@ietf.org
> http://www1.ietf.org/mailman/listinfo/diffserv
> Archive: http://www-nrg.ee.lbl.gov/diff-serv-arch/
> 


From owner-l2tp@ipsec.org  Thu May 11 04:48:04 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA22754
	for <l2tpext-archive@odin.ietf.org>; Thu, 11 May 2000 04:48:03 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id WAA06301; Wed, 10 May 2000 22:32:49 -0700 (PDT)
Received: from ftpbox.mot.com (ftpbox.mot.com [129.188.136.101]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id WAA06297 for <l2tp@ipsec.org>; Wed, 10 May 2000 22:32:14 -0700 (PDT)
Received: [from mothost.mot.com (mothost.mot.com [129.188.137.101]) by ftpbox.mot.com (ftpbox 2.1) with ESMTP id AAA20980; Thu, 11 May 2000 00:55:42 -0700 (MST)]
Received: [from hpux4.miel.mot.com (hpux4.miel.mot.com [217.1.84.89]) by mothost.mot.com (MOT-mothost 2.0) with ESMTP id AAA16694; Thu, 11 May 2000 00:55:35 -0700 (MST)]
Received: from xmail.miel.mot.com (xmail.miel.mot.com [217.1.84.38])
	by miel.mot.com (8.9.3/8.9.3) with ESMTP id NAA27482;
	Thu, 11 May 2000 13:30:02 +0530 (IST)
Received: by xmail.miel.mot.com with Internet Mail Service (5.5.2650.21)
	id <KSPFDCRN>; Thu, 11 May 2000 13:16:16 +0530
Message-ID: <2C202C0F886DD3119B1200A0C9A85FF405DC59@xmail.miel.mot.com>
From: Radhakrishna <rk@miel.mot.com>
To: "'rajeshs@futsoft.com'" <rajeshs@futsoft.com>,
        "'Rajesh Sundaram'"
	 <rajeshsu@Exchange.Microsoft.com>,
        diffserv@ietf.org, l2tp@ipsec.org
Subject: RE: [Diffserv] RE: Incompatibility between Diffserv and PPP based
	 tunneling protocols
Date: Thu, 11 May 2000 13:16:13 +0530
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain
Sender: owner-l2tp@ipsec.org
Precedence: bulk

Same issue is there with GRE tunneling also. Turning off sequencing and
reordering is definetly one solution. But if sequencing is really required,
the reordering can be done based on DSCPs and hence multiple tunnels
are not required.

Regards,
.....RK

> -----Original Message-----
> From:	Rajesh Kumar [SMTP:rajeshs@futsoft.com]
> Sent:	Thursday, May 11, 2000 6:55 AM
> To:	'Rajesh Sundaram'; diffserv@ietf.org; l2tp@ipsec.org
> Subject:	[Diffserv] RE: Incompatibility between Diffserv and PPP
> based tunneling protocols
> 
> If you turn off reordering and sequencing (as these are optional
> in l2tp) and let tcp handle these issues, l2tp should work
> seamlessly. tcp would handle the reliability issues for each
> microflow. (I don't know much about pptp).
> But this works only for tcp. How about others - like udp? - that's
> unreliable service anyway, so why try to introduce reliability
> at the link layer?
> 
> Does that sound workable?
> 
> -----Original Message-----
> From:	Rajesh Sundaram [SMTP:rajeshsu@Exchange.Microsoft.com]
> Sent:	Wednesday, May 10, 2000 4:33 PM
> To:	diffserv@ietf.org; l2tp@ipsec.org
> Subject:	Incompatibility between Diffserv and PPP based tunneling
> protocols
> 
> Folks,
> 
> There seems to be an incompatibility between tunneling protocols that
> encapsulate PPP (like L2TP and PPTP) and Diffserv PHBs. 
> 
> Let's assume that there are two micro-flows of traffic within a single
> L2TP tunnel, marked for different PHBs. The DSCP mark has to be
> propagated from the original packet to the tunneled header at tunnel
> ingress (as per section 2.5 of EF PHB). The restrictions on packet
> re-order apply only to a packets belonging to a micro flow within a
> particular PHB class. So, it is possible for packets belonging to
> different PHBs to get re-ordered by DS aware nodes en-route to tunnel
> egress. 
> 
> This re-ordering breaks PPTP and L2TP. PPTP is required to discard
> re-ordered packets in order to maintain the ordering semantics of PPP.
> L2TP allows for the receiver to restore the original sequence of
> re-ordered packets, but doing this negates the benefits of the PHB,
> since the packet will potentially remain in the receiver's tunnel buffer
> for a long time. 
> 
> One way of solving this problem is to create multiple L2TP or PPTP
> sessions and by dedicating one PHB for each session. Since both L2TP and
> PPTP allow re-ordering across sessions, the packet re-order problem is
> not as nasty.
> 
> Any suggestions on how this incompatibility can be addressed ?
> 
> Thanks,
> Rajesh.
>  << File: ATT00002.htm >> 
> 
> _______________________________________________
> diffserv mailing list
> diffserv@ietf.org
> http://www1.ietf.org/mailman/listinfo/diffserv
> Archive: http://www-nrg.ee.lbl.gov/diff-serv-arch/


From owner-l2tp@ipsec.org  Thu May 11 04:55:29 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA22784
	for <l2tpext-archive@odin.ietf.org>; Thu, 11 May 2000 04:55:29 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id WAA06315; Wed, 10 May 2000 22:46:45 -0700 (PDT)
Received: from dirty.research.bell-labs.com (dirty.research.bell-labs.com [204.178.16.6]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id WAA06311 for <l2tp@ipsec.org>; Wed, 10 May 2000 22:46:28 -0700 (PDT)
Received: from zubin.dnrc.bell-labs.com ([135.180.130.56]) by dirty; Thu May 11 04:09:00 EDT 2000
Received: from dnrc.bell-labs.com (ex-vpn69.pa.bell-labs.com [135.250.1.69])
	by zubin.dnrc.bell-labs.com (8.9.3/8.9.3) with ESMTP id EAA12073;
	Thu, 11 May 2000 04:08:56 -0400 (EDT)
Message-ID: <391A6B6E.AD568A04@dnrc.bell-labs.com>
Date: Thu, 11 May 2000 01:12:30 -0700
From: Grenville Armitage <gja@dnrc.bell-labs.com>
Organization: Bell Laboratories, Lucent Technologies
X-Mailer: Mozilla 4.61 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Mahalingam Mani <Mani@vpnet.com>
CC: diffserv@ietf.org, l2tp@ipsec.org
Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based tunneling  
 protocols
References: <D899E9E27BE9D211842200805FA67B43923551@mail.vpnet.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit



Mahalingam Mani wrote:
> 
> What if the desired PHB is as required on a per-microflow basis - even over
> the
> tunnel? Especially the case with IKE tunnels where one may encounter
> multiple
> DS router hops in the tunnel?

I'm not really sure how to parse your question. In general
tunnels provide isolation - conventionally just topological
isolation, but increasingly we're having to consider temporal
isolation (aka QoS). So consider the tunnel as becoming more
like a link substitute in that one provisions the tunnel's
temporal characteristics (QoS) to, in the aggregate, support
the mix of traffic flowing through it. In the same way that
we might mix EF and AF traffic on a single T1 line, so we
might mix all sorts of traffic onto a tunnel having a fixed
DSCP of its own. The tunnel's QoS is determined by the outer
DSCP (and the provisioning of the DS routers along the path
followed by the tunnel).

cheers,
gja


From owner-l2tp@ipsec.org  Thu May 11 10:08:11 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id KAA28449
	for <l2tpext-archive@odin.ietf.org>; Thu, 11 May 2000 10:08:10 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id DAA07868; Thu, 11 May 2000 03:41:26 -0700 (PDT)
Received: from phoenix.ficon-tech.com (phoenix.ficon-tech.com [209.125.90.2]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id DAA07864 for <l2tp@ipsec.org>; Thu, 11 May 2000 03:41:02 -0700 (PDT)
Received: from globespan.net (192.168.137.202) by phoenix.ficon-tech.com (Worldmail 1.3.167); 11 May 2000 09:14:44 -0400
Message-ID: <391AAFC5.3610EA8D@globespan.net>
Date: Thu, 11 May 2000 09:04:05 -0400
From: Rohit Chhapolia <rohit@globespan.net>
Organization: Globespan, Inc.
X-Mailer: Mozilla 4.72 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Grenville Armitage <gja@dnrc.bell-labs.com>
CC: Mahalingam Mani <Mani@vpnet.com>, diffserv@ietf.org, l2tp@ipsec.org
Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based tunneling  
 protocols
References: <D899E9E27BE9D211842200805FA67B43923551@mail.vpnet.com> <391A6B6E.AD568A04@dnrc.bell-labs.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit

Grenville Armitage wrote:
> 
> Mahalingam Mani wrote:
> >
> > What if the desired PHB is as required on a per-microflow basis - even over
> > the
> > tunnel? Especially the case with IKE tunnels where one may encounter
> > multiple
> > DS router hops in the tunnel?
> 
> I'm not really sure how to parse your question. In general
> tunnels provide isolation - conventionally just topological
> isolation, but increasingly we're having to consider temporal
> isolation (aka QoS). So consider the tunnel as becoming more
> like a link substitute in that one provisions the tunnel's
> temporal characteristics (QoS) to, in the aggregate, support
> the mix of traffic flowing through it. In the same way that
> we might mix EF and AF traffic on a single T1 line, so we
> might mix all sorts of traffic onto a tunnel having a fixed
> DSCP of its own. The tunnel's QoS is determined by the outer
> DSCP (and the provisioning of the DS routers along the path
> followed by the tunnel).

What criteria should be used at the tunnel's ingress node to
decide the traffic that can go through this tunnel? If the
tunnel's DSCP is AF, can one send both EF and AF traffic
through it? It may lead to degradation of service for the EF class.

Rohit

-- 
Rohit Chhapolia
Globespan, Inc.
mailto:rohit@globespan.net
http://www.globespan.net


From owner-l2tp@ipsec.org  Thu May 11 13:17:39 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA03486
	for <l2tpext-archive@odin.ietf.org>; Thu, 11 May 2000 13:17:37 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id GAA08086; Thu, 11 May 2000 06:48:44 -0700 (PDT)
Received: from dirty.research.bell-labs.com (dirty.research.bell-labs.com [204.178.16.6]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id GAA08082 for <l2tp@ipsec.org>; Thu, 11 May 2000 06:48:27 -0700 (PDT)
Received: from zubin.dnrc.bell-labs.com ([135.180.130.56]) by dirty; Thu May 11 12:11:02 EDT 2000
Received: from dnrc.bell-labs.com (ex-vpn64.pa.bell-labs.com [135.250.1.64])
	by zubin.dnrc.bell-labs.com (8.9.3/8.9.3) with ESMTP id MAA15421;
	Thu, 11 May 2000 12:10:55 -0400 (EDT)
Message-ID: <391ADC66.90872B8E@dnrc.bell-labs.com>
Date: Thu, 11 May 2000 09:14:30 -0700
From: Grenville Armitage <gja@dnrc.bell-labs.com>
Organization: Bell Laboratories, Lucent Technologies
X-Mailer: Mozilla 4.61 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Rohit Chhapolia <rohit@globespan.net>
CC: diffserv@ietf.org, l2tp@ipsec.org
Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based tunneling  
 protocols
References: <D899E9E27BE9D211842200805FA67B43923551@mail.vpnet.com> <391A6B6E.AD568A04@dnrc.bell-labs.com> <391AAFC5.3610EA8D@globespan.net>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit



Rohit Chhapolia wrote:
	[..]
> What criteria should be used at the tunnel's ingress node to
> decide the traffic that can go through this tunnel? If the
> tunnel's DSCP is AF, can one send both EF and AF traffic
> through it? It may lead to degradation of service for the EF class.

In general the answer is more like 'it depends'. Since the
tunnel becomes more like a link, we need to understand its
ingress->egress QoS characteristics before we can decide
how well it would support any given mix of tunneled traffic.
Unfortunately knowledge of the tunnel's PHB alone doesn't
tell us all there is to know - it will depend on network
topology and router provisioning along the tunnel's path.
This is the nature of Diffserv (to date, although people are
working on defining e2e behaviors built from per hop behaviors)

cheers,
gja
________________________________________________________________________
Grenville Armitage                    http://members.home.net/garmitage/
Bell Labs Research Silicon Valley


From owner-l2tp@ipsec.org  Thu May 11 17:10:36 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA08280
	for <l2tpext-archive@odin.ietf.org>; Thu, 11 May 2000 17:10:35 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id KAA08509; Thu, 11 May 2000 10:46:48 -0700 (PDT)
Received: from mail-gw.hursley.ibm.com (mail-gw.hursley.ibm.com [194.196.110.15]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id KAA08505 for <l2tp@ipsec.org>; Thu, 11 May 2000 10:46:19 -0700 (PDT)
Received: from sp3at21.hursley.ibm.com (sp3at21.hursley.ibm.com [9.20.45.21]) by mail-gw.hursley.ibm.com (AIX4.3/UCB 8.8.8/8.8.8) with ESMTP id VAA50066; Thu, 11 May 2000 21:09:12 +0100
Received: from hursley.ibm.com (gsine03.us.sine.ibm.com [9.14.6.43]) by sp3at21.hursley.ibm.com (AIX4.2/UCB 8.7/8.7.3) with ESMTP id VAA16378; Thu, 11 May 2000 21:09:09 +0100 (BST)
Message-ID: <391B135A.4A221D8A@hursley.ibm.com>
Date: Thu, 11 May 2000 15:08:58 -0500
From: Brian E Carpenter <brian@hursley.ibm.com>
Organization: IBM
X-Mailer: Mozilla 4.61 [en] (Win98; I)
X-Accept-Language: en,fr
MIME-Version: 1.0
To: Grenville Armitage <gja@dnrc.bell-labs.com>
CC: Rohit Chhapolia <rohit@globespan.net>, diffserv@ietf.org, l2tp@ipsec.org,
        David Black <Black_David@emc.com>
Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based tunneling  
 protocols
References: <D899E9E27BE9D211842200805FA67B43923551@mail.vpnet.com> <391A6B6E.AD568A04@dnrc.bell-labs.com> <391AAFC5.3610EA8D@globespan.net> <391ADC66.90872B8E@dnrc.bell-labs.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit

Three comments (on the whole thread):

1. The correct place to write up the conclusion of this, when we
reach one, is in draft-ietf-diffserv-tunnels-00.txt as revised.

2. Remember that the diffserv rule that microflows must not
be reordered is met by not reordering the whole BA, and is
also met by not reordering the whole tunnel. It is completely
conformant with diffserv to perform no reordering whatever.
Viewed that way there is no problem, just an implementation
constraint.

3. We might need to add some words to RFC 2598 section 2.5,
but I see no problem in enclosing an EF flow and other flows
such as BE or AF within an EF tunnel, as long as the tunnel
has adequate bandwidth and admission control. That's why we
call it virtual wire.

   Brian

Grenville Armitage wrote:
> 
> Rohit Chhapolia wrote:
>         [..]
> > What criteria should be used at the tunnel's ingress node to
> > decide the traffic that can go through this tunnel? If the
> > tunnel's DSCP is AF, can one send both EF and AF traffic
> > through it? It may lead to degradation of service for the EF class.
> 
> In general the answer is more like 'it depends'. Since the
> tunnel becomes more like a link, we need to understand its
> ingress->egress QoS characteristics before we can decide
> how well it would support any given mix of tunneled traffic.
> Unfortunately knowledge of the tunnel's PHB alone doesn't
> tell us all there is to know - it will depend on network
> topology and router provisioning along the tunnel's path.
> This is the nature of Diffserv (to date, although people are
> working on defining e2e behaviors built from per hop behaviors)
> 
> cheers,
> gja
> ________________________________________________________________________
> Grenville Armitage                    http://members.home.net/garmitage/
> Bell Labs Research Silicon Valley
> 
> _______________________________________________
> diffserv mailing list
> diffserv@ietf.org
> http://www1.ietf.org/mailman/listinfo/diffserv
> Archive: http://www-nrg.ee.lbl.gov/diff-serv-arch/


From owner-l2tp@ipsec.org  Thu May 11 19:57:35 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id TAA11257
	for <l2tpext-archive@odin.ietf.org>; Thu, 11 May 2000 19:57:34 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id NAA08664; Thu, 11 May 2000 13:27:23 -0700 (PDT)
Received: from mxbh4.isus.emc.com (mxbh4.isus.emc.com [168.159.208.52]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id NAA08660 for <l2tp@ipsec.org>; Thu, 11 May 2000 13:26:40 -0700 (PDT)
From: Black_David@emc.com
Received: by mxbh4.isus.emc.com with Internet Mail Service (5.5.2448.0)
	id <KTZ8AX4M>; Thu, 11 May 2000 18:49:43 -0400
Message-ID: <0F31E5C394DAD311B60C00E029101A070148F8AF@CORPMX9>
To: rajeshsu@Exchange.Microsoft.com, diffserv@ietf.org, l2tp@ipsec.org
Cc: Black_David@emc.com
Subject: RE: [Diffserv] Incompatibility between Diffserv and PPP based tun
	neling protocols
Date: Thu, 11 May 2000 18:49:42 -0400
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2448.0)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

Back to the original problem scenario:

> Let's assume that there are two micro-flows of traffic within a single
L2TP tunnel, marked for different PHBs.
> The DSCP mark has to be propagated from the original packet to the
tunneled header at tunnel ingress
> (as per section 2.5 of EF PHB). The restrictions on packet re-order apply
only to a packets belonging to
> a micro flow within a particular PHB class. So, it is possible for packets
belonging to different PHBs to get
> re-ordered by DS aware nodes en-route to tunnel egress. 
>
> This re-ordering breaks PPTP and L2TP.

This seems to be a classic "Doctor it hurts when I do *this*.
-- Don't do that." scenario.  The easiest solution is to mark
EF on all the outer headers.  Beyond that, upon reflection,
the "must" in Section 2.5 of the EF RFC appears to be overstated
- it probably ought to be a "should", perhaps accompanied
by language like that in Section 5 of the AF RFC.  There
are circumstances under which EF can be run through a
non-EF tunnel without damage (e.g., overprovisioned AF),
but it is necessary to understand what one is doing.
This leads to a solution with AF or a non-zero class
selector in the outer header of the tunneled packets.

> One way of solving this problem is to create multiple L2TP or PPTP
sessions and by
> dedicating one PHB for each session. Since both L2TP and PPTP allow
re-ordering
> across sessions, the packet re-order problem is not as nasty.

This looks like the right solution - in essence to run
multiple BAs among which reordering is desired through
a tunneling protocol that cannot tolerate reordering, one
needs to use multiple tunnels (e.g., one per BA).  For
the examples being discussed, this should "just work",
as each L2TP or PPTP session will appear to be a single
microflow to the diffserv-aware nodes and hence be governed
by the diffserv injunctions against reordering packets in a
microflow -- this seems like something that could be
productively clarified in the next version of the tunnels
draft, as it's not immediately obvious from the existing
documents.

A couple of other notes on points raised in the discussion:

> The suggestion to split the two microflows into 2 different tunnels will
work,
> however with thousands of tunnels to manage in an enterprise/carrier-class
VPN
> setup this will become an administrative nightmare.

If the tunnels are per-BA rather than per-microflow, the
resulting administrative consequences should be manageable.
If not, there's always the option of choosing not to make
diffserv forwarding distinctions within the tunnel -
there's an instance of the "no free lunch" principle here.

> What criteria should be used at the tunnel's ingress node to
> decide the traffic that can go through this tunnel? If the
> tunnel's DSCP is AF, can one send both EF and AF traffic
> through it? It may lead to degradation of service for the EF class.

While it may seem like hair-splitting, this is really a
traffic and network engineering problem.  Diffserv can
say that "degradation of service for the EF class"
is wrong, but leaves it to network and equipment
designers/implementers to figure out how to make sure
it doesn't happen.  Diffserv is not a substitute for
competent network/traffic design/engineering.  I think
Grenville's already made basically this point.

As Brian noted, this is a great opportunity to figure
out what else ought to be in the tunnels draft - as
the author of that draft, I'm open to any and all
suggestions that will help avoid future confusion.

Thanks,
--David

-------------------------------------------------
David L. Black, Senior Technologist
EMC Corporation, 42 South St., Hopkinton, MA  01748
+1 (508) 435-1000 x75140, FAX: +1 (508) 497-6909
black_david@emc.com  Cellular: +1 (978) 394-7754
---------------------------------------------------



From owner-l2tp@ipsec.org  Fri May 12 06:39:48 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id GAA00256
	for <l2tpext-archive@odin.ietf.org>; Fri, 12 May 2000 06:39:47 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id XAA09319; Thu, 11 May 2000 23:49:55 -0700 (PDT)
Received: from relay1.marconi.it (relay1.marconi.it [192.106.52.1]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id XAA09315 for <l2tp@ipsec.org>; Thu, 11 May 2000 23:49:25 -0700 (PDT)
From: Gianluca.Rolandelli@Marconicomms.com
Received: from marconicomms.com (localhost [127.0.0.1])
	by relay1.marconi.it (8.9.1a/8.9.1) with SMTP id LAA24162
	for <l2tp@ipsec.org>; Fri, 12 May 2000 11:08:32 +0200 (MET DST)
Received: by marconicomms.com(Lotus SMTP MTA v4.6.3  (733.2 10-16-1998))  id C12568DD.0031E5B9 ; Fri, 12 May 2000 11:05:00 +0200
X-Lotus-FromDomain: MCMAIN@MCEXT
To: l2tp@ipsec.org
Message-ID: <C12568DD.003102F7.00@marconicomms.com>
Date: Fri, 12 May 2000 10:09:13 +0200
Subject: PPP and L2TP
Mime-Version: 1.0
Content-type: text/plain; charset=iso-8859-1
Content-Disposition: inline
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 8bit
X-MIME-Autoconverted: from quoted-printable to 8bit by ietf.org id GAA00256



Hi all,
we are reading various RFCs regarding PPP and L2TP in order to try to understand
the relationships between PPP and L2TP in the LNS system.
It seems to me that this argument isn?t treat of in none documents.

In particular, I do not have clear the behaviour of the LNS system when an
incoming call is received from the LAC.

We have supposed that a PPP peer (a remote dial-up user) starts to send LCP and
authentication packets to a LAC.
After receiving an authentication request message, with the FQDN "user@domain"
in order to understand the destination LNS, the LAC sets up a new L2TP session
(we supposed the tunnel has already been set up).
After the tunnel/session set up, the PPP remote user and the LNS will exchange
PPP packets directly (the LAC forwards them transparently).
The PPP state machine of the LNS is not started yet (it?s still in its ?PPP dead
phase?), while the PPP user is waiting for authentication ack/nack message.

I supposed (and this is what don?t understand) that the LNS, after
tunnel/session establishment, "must" send an LCP configure request to its PPP
peer (the user) in order to set up a PPP connection (the PPP connection of the
user will be restarted).
Is this true?
Are there other methods?

We argue that if the LNS does not send any LCP messages immediately after the
new L2TP session set up, the PPP remote user will wait for authentication
ack/nack, until the PPP connection will be teared down.
I do not have found this assumption on RFCs, so my question is: ?Is this a
requirement for an LNS? Or is only an implementative issue??

Moreover, even if the remote user is going to retry to send an authentication
request message again, this message will arrive directly to the LNS (after the
tunnel/session establishment, the LAC is transparently) that will silent discard
it (the LNS is still in its ?PPP dead phase?).

Thanks to all in advance for possible answers.




From owner-l2tp@ipsec.org  Fri May 12 16:52:48 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA14710
	for <l2tpext-archive@odin.ietf.org>; Fri, 12 May 2000 16:52:47 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id KAA11167; Fri, 12 May 2000 10:27:17 -0700 (PDT)
Received: from sj-msg-core-1.cisco.com (sj-msg-core-1.cisco.com [171.71.163.11]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id KAA11163 for <l2tp@ipsec.org>; Fri, 12 May 2000 10:27:04 -0700 (PDT)
Received: from airborne.cisco.com (airborne.cisco.com [171.71.154.32])
	by sj-msg-core-1.cisco.com (8.9.3/8.9.1) with ESMTP id MAA27080;
	Fri, 12 May 2000 12:50:15 -0700 (PDT)
Received: from cisco.com (sjck-dial-gw5-151.cisco.com [10.19.238.152])
	by airborne.cisco.com (Mirapoint)
	with ESMTP id AAO14384;
	Fri, 12 May 2000 12:50:06 -0700 (PDT)
Message-ID: <391C364E.FD4D839F@cisco.com>
Date: Fri, 12 May 2000 09:50:22 -0700
From: Scott Brim <sbrim@cisco.com>
X-Mailer: Mozilla 4.72 [en] (WinNT; I)
X-Accept-Language: en
MIME-Version: 1.0
To: Mahalingam Mani <Mani@vpnet.com>
CC: Grenville Armitage <gja@dnrc.bell-labs.com>, diffserv@ietf.org,
        l2tp@ipsec.org
Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based tunneling  
 protocols
References: <D899E9E27BE9D211842200805FA67B43923551@mail.vpnet.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit

If you want different treatment you should use different tunnels (until we
figure out something better).

Mahalingam Mani wrote:

> What if the desired PHB is as required on a per-microflow basis - even over
> the
> tunnel? Especially the case with IKE tunnels where one may encounter
> multiple
> DS router hops in the tunnel?




From owner-l2tp@ipsec.org  Fri May 12 18:58:48 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id SAA17641
	for <l2tpext-archive@odin.ietf.org>; Fri, 12 May 2000 18:58:47 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id MAA11339; Fri, 12 May 2000 12:22:52 -0700 (PDT)
Received: from dfssl.exchange.microsoft.com ([131.107.88.59]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id MAA11329 for <l2tp@ipsec.org>; Fri, 12 May 2000 12:19:19 -0700 (PDT)
Received: from 172.30.236.11 by dfssl.exchange.microsoft.com (InterScan E-Mail VirusWall NT); Fri, 12 May 2000 14:33:40 -0700 (Pacific Daylight Time)
Received: from SIT.platinum.corp.microsoft.com ([172.30.236.233]) by yuri.dns.microsoft.com with Microsoft SMTPSVC(5.0.2195.2023);
	 Fri, 12 May 2000 14:40:50 -0700
content-class: urn:content-classes:message
Subject: RE: [Diffserv] Incompatibility between Diffserv and PPP based tunneling   protocols
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01BFBC59.E2EDA6C0"
Date: Fri, 12 May 2000 14:34:43 -0700
X-MimeOLE: Produced By Microsoft Exchange V6.0.4366.0
Message-ID: <19398D273324D3118A2B0008C7E9A5690681E0DB@SIT.platinum.corp.microsoft.com>
Thread-Topic: [Diffserv] Incompatibility between Diffserv and PPP based tunneling   protocols
Thread-Index: Ab+7hlgdiQdbZLGOQZSL7VpK8iMeQwAvEZGA
From: "Rajesh Sundaram" <rajeshsu@Exchange.Microsoft.com>
To: "Brian E Carpenter" <brian@hursley.ibm.com>,
        "Grenville Armitage" <gja@dnrc.bell-labs.com>,
        "David Black" <Black_David@emc.com>
Cc: <diffserv@ietf.org>, <l2tp@ipsec.org>
X-OriginalArrivalTime: 12 May 2000 21:40:50.0242 (UTC) FILETIME=[BD860A20:01BFBC5A]
Sender: owner-l2tp@ipsec.org
Precedence: bulk

This is a multi-part message in MIME format.

------_=_NextPart_001_01BFBC59.E2EDA6C0
Content-Type: text/plain;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

So far, there are two proposed solutions to this problem.=20

a) Mark the outer header packets of all tunneled packets with a uniform
codepoint.=20

As Brian pointed out, this treats the tunnel like a virtual wire and
works well as long as there is sufficient bandwidth. But, there could be
a problem if EF bandwidth was limited. By multiplexing data and EF onto
an EF tunnel, we are increasing the probability of getting policed at a
DS boundary, and thereby breaking the guarantees of inner EF traffic.

b) Create a (small) set of multiple tunnels, each tunnel getting a fixed
codepoint, that is distinct from other tunnels. All packets on a
particular tunnel carry the same uniform codepoint. Before tunnel
encapsulation, inspect the inner codepoint of the packet and route the
packet to the appropriate tunnel.

I had not necessarily suggested creating a different tunnel for each PHB
(or for each micro-flow). By creating a smaller set of tunnels, we are
reducing the management overhead. This approach would need a mapping of
inner DSCPs --> outer DSCPs, which is probably controlled by the local
tunnel administrator.

The first approach seems to be a special case of the second one (where
there all inner DSCPs map to a uniform outer DSCP). Also, the second one
is more flexible. I'd appreciate comments that help us make a decision
on this.=20

Thanks,
Rajesh

-----Original Message-----
From: Brian E Carpenter [mailto:brian@hursley.ibm.com]
Sent: Thursday, May 11, 2000 1:09 PM
To: Grenville Armitage
Cc: Rohit Chhapolia; diffserv@ietf.org; l2tp@ipsec.org; David Black
Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based
tunneling protocols


Three comments (on the whole thread):

1. The correct place to write up the conclusion of this, when we
reach one, is in draft-ietf-diffserv-tunnels-00.txt as revised.

2. Remember that the diffserv rule that microflows must not
be reordered is met by not reordering the whole BA, and is
also met by not reordering the whole tunnel. It is completely
conformant with diffserv to perform no reordering whatever.
Viewed that way there is no problem, just an implementation
constraint.

3. We might need to add some words to RFC 2598 section 2.5,
but I see no problem in enclosing an EF flow and other flows
such as BE or AF within an EF tunnel, as long as the tunnel
has adequate bandwidth and admission control. That's why we
call it virtual wire.

   Brian

Grenville Armitage wrote:
>=20
> Rohit Chhapolia wrote:
>         [..]
> > What criteria should be used at the tunnel's ingress node to
> > decide the traffic that can go through this tunnel? If the
> > tunnel's DSCP is AF, can one send both EF and AF traffic
> > through it? It may lead to degradation of service for the EF class.
>=20
> In general the answer is more like 'it depends'. Since the
> tunnel becomes more like a link, we need to understand its
> ingress->egress QoS characteristics before we can decide
> how well it would support any given mix of tunneled traffic.
> Unfortunately knowledge of the tunnel's PHB alone doesn't
> tell us all there is to know - it will depend on network
> topology and router provisioning along the tunnel's path.
> This is the nature of Diffserv (to date, although people are
> working on defining e2e behaviors built from per hop behaviors)
>=20
> cheers,
> gja
>
________________________________________________________________________
> Grenville Armitage
http://members.home.net/garmitage/
> Bell Labs Research Silicon Valley
>=20
> _______________________________________________
> diffserv mailing list
> diffserv@ietf.org
> http://www1.ietf.org/mailman/listinfo/diffserv
> Archive: http://www-nrg.ee.lbl.gov/diff-serv-arch/

_______________________________________________
diffserv mailing list
diffserv@ietf.org
http://www1.ietf.org/mailman/listinfo/diffserv
Archive: http://www-nrg.ee.lbl.gov/diff-serv-arch/

------_=_NextPart_001_01BFBC59.E2EDA6C0
Content-Type: text/html;
	charset="us-ascii"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Dus-ascii">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
6.0.4366.0">
<TITLE>RE: [Diffserv] Incompatibility between Diffserv and PPP based =
tunneling   protocols</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->

<P><FONT SIZE=3D2>So far, there are two proposed solutions to this =
problem. </FONT>
</P>

<P><FONT SIZE=3D2>a) Mark the outer header packets of all tunneled =
packets with a uniform codepoint. </FONT>
</P>

<P><FONT SIZE=3D2>As Brian pointed out, this treats the tunnel like a =
virtual wire and works well as long as there is sufficient bandwidth. =
But, there could be a problem if EF bandwidth was limited. By =
multiplexing data and EF onto an EF tunnel, we are increasing the =
probability of getting policed at a DS boundary, and thereby breaking =
the guarantees of inner EF traffic.</FONT></P>

<P><FONT SIZE=3D2>b) Create a (small) set of multiple tunnels, each =
tunnel getting a fixed codepoint, that is distinct from other tunnels. =
All packets on a particular tunnel carry the same uniform codepoint. =
Before tunnel encapsulation, inspect the inner codepoint of the packet =
and route the packet to the appropriate tunnel.</FONT></P>

<P><FONT SIZE=3D2>I had not necessarily suggested creating a different =
tunnel for each PHB (or for each micro-flow). By creating a smaller set =
of tunnels, we are reducing the management overhead. This approach would =
need a mapping of inner DSCPs --&gt; outer DSCPs, which is probably =
controlled by the local tunnel administrator.</FONT></P>

<P><FONT SIZE=3D2>The first approach seems to be a special case of the =
second one (where there all inner DSCPs map to a uniform outer DSCP). =
Also, the second one is more flexible. I'd appreciate comments that help =
us make a decision on this. </FONT></P>

<P><FONT SIZE=3D2>Thanks,</FONT>

<BR><FONT SIZE=3D2>Rajesh</FONT>
</P>

<P><FONT SIZE=3D2>-----Original Message-----</FONT>

<BR><FONT SIZE=3D2>From: Brian E Carpenter [<A =
HREF=3D"mailto:brian@hursley.ibm.com">mailto:brian@hursley.ibm.com</A>]</=
FONT>

<BR><FONT SIZE=3D2>Sent: Thursday, May 11, 2000 1:09 PM</FONT>

<BR><FONT SIZE=3D2>To: Grenville Armitage</FONT>

<BR><FONT SIZE=3D2>Cc: Rohit Chhapolia; diffserv@ietf.org; =
l2tp@ipsec.org; David Black</FONT>

<BR><FONT SIZE=3D2>Subject: Re: [Diffserv] Incompatibility between =
Diffserv and PPP based</FONT>

<BR><FONT SIZE=3D2>tunneling protocols</FONT>
</P>
<BR>

<P><FONT SIZE=3D2>Three comments (on the whole thread):</FONT>
</P>

<P><FONT SIZE=3D2>1. The correct place to write up the conclusion of =
this, when we</FONT>

<BR><FONT SIZE=3D2>reach one, is in draft-ietf-diffserv-tunnels-00.txt =
as revised.</FONT>
</P>

<P><FONT SIZE=3D2>2. Remember that the diffserv rule that microflows =
must not</FONT>

<BR><FONT SIZE=3D2>be reordered is met by not reordering the whole BA, =
and is</FONT>

<BR><FONT SIZE=3D2>also met by not reordering the whole tunnel. It is =
completely</FONT>

<BR><FONT SIZE=3D2>conformant with diffserv to perform no reordering =
whatever.</FONT>

<BR><FONT SIZE=3D2>Viewed that way there is no problem, just an =
implementation</FONT>

<BR><FONT SIZE=3D2>constraint.</FONT>
</P>

<P><FONT SIZE=3D2>3. We might need to add some words to RFC 2598 section =
2.5,</FONT>

<BR><FONT SIZE=3D2>but I see no problem in enclosing an EF flow and =
other flows</FONT>

<BR><FONT SIZE=3D2>such as BE or AF within an EF tunnel, as long as the =
tunnel</FONT>

<BR><FONT SIZE=3D2>has adequate bandwidth and admission control. That's =
why we</FONT>

<BR><FONT SIZE=3D2>call it virtual wire.</FONT>
</P>

<P><FONT SIZE=3D2>&nbsp;&nbsp; Brian</FONT>
</P>

<P><FONT SIZE=3D2>Grenville Armitage wrote:</FONT>

<BR><FONT SIZE=3D2>&gt; </FONT>

<BR><FONT SIZE=3D2>&gt; Rohit Chhapolia wrote:</FONT>

<BR><FONT SIZE=3D2>&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
[..]</FONT>

<BR><FONT SIZE=3D2>&gt; &gt; What criteria should be used at the =
tunnel's ingress node to</FONT>

<BR><FONT SIZE=3D2>&gt; &gt; decide the traffic that can go through this =
tunnel? If the</FONT>

<BR><FONT SIZE=3D2>&gt; &gt; tunnel's DSCP is AF, can one send both EF =
and AF traffic</FONT>

<BR><FONT SIZE=3D2>&gt; &gt; through it? It may lead to degradation of =
service for the EF class.</FONT>

<BR><FONT SIZE=3D2>&gt; </FONT>

<BR><FONT SIZE=3D2>&gt; In general the answer is more like 'it depends'. =
Since the</FONT>

<BR><FONT SIZE=3D2>&gt; tunnel becomes more like a link, we need to =
understand its</FONT>

<BR><FONT SIZE=3D2>&gt; ingress-&gt;egress QoS characteristics before we =
can decide</FONT>

<BR><FONT SIZE=3D2>&gt; how well it would support any given mix of =
tunneled traffic.</FONT>

<BR><FONT SIZE=3D2>&gt; Unfortunately knowledge of the tunnel's PHB =
alone doesn't</FONT>

<BR><FONT SIZE=3D2>&gt; tell us all there is to know - it will depend on =
network</FONT>

<BR><FONT SIZE=3D2>&gt; topology and router provisioning along the =
tunnel's path.</FONT>

<BR><FONT SIZE=3D2>&gt; This is the nature of Diffserv (to date, =
although people are</FONT>

<BR><FONT SIZE=3D2>&gt; working on defining e2e behaviors built from per =
hop behaviors)</FONT>

<BR><FONT SIZE=3D2>&gt; </FONT>

<BR><FONT SIZE=3D2>&gt; cheers,</FONT>

<BR><FONT SIZE=3D2>&gt; gja</FONT>

<BR><FONT SIZE=3D2>&gt; =
________________________________________________________________________<=
/FONT>

<BR><FONT SIZE=3D2>&gt; Grenville =
Armitage&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp=
;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; <A =
HREF=3D"http://members.home.net/garmitage/">http://members.home.net/garmi=
tage/</A></FONT>

<BR><FONT SIZE=3D2>&gt; Bell Labs Research Silicon Valley</FONT>

<BR><FONT SIZE=3D2>&gt; </FONT>

<BR><FONT SIZE=3D2>&gt; =
_______________________________________________</FONT>

<BR><FONT SIZE=3D2>&gt; diffserv mailing list</FONT>

<BR><FONT SIZE=3D2>&gt; diffserv@ietf.org</FONT>

<BR><FONT SIZE=3D2>&gt; <A =
HREF=3D"http://www1.ietf.org/mailman/listinfo/diffserv">http://www1.ietf.=
org/mailman/listinfo/diffserv</A></FONT>

<BR><FONT SIZE=3D2>&gt; Archive: <A =
HREF=3D"http://www-nrg.ee.lbl.gov/diff-serv-arch/">http://www-nrg.ee.lbl.=
gov/diff-serv-arch/</A></FONT>
</P>

<P><FONT SIZE=3D2>_______________________________________________</FONT>

<BR><FONT SIZE=3D2>diffserv mailing list</FONT>

<BR><FONT SIZE=3D2>diffserv@ietf.org</FONT>

<BR><FONT SIZE=3D2><A =
HREF=3D"http://www1.ietf.org/mailman/listinfo/diffserv">http://www1.ietf.=
org/mailman/listinfo/diffserv</A></FONT>

<BR><FONT SIZE=3D2>Archive: <A =
HREF=3D"http://www-nrg.ee.lbl.gov/diff-serv-arch/">http://www-nrg.ee.lbl.=
gov/diff-serv-arch/</A></FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01BFBC59.E2EDA6C0--


From owner-l2tp@ipsec.org  Fri May 12 20:26:02 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA18962
	for <l2tpext-archive@odin.ietf.org>; Fri, 12 May 2000 20:26:02 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id OAA11476; Fri, 12 May 2000 14:08:49 -0700 (PDT)
Received: from dirty.research.bell-labs.com (dirty.research.bell-labs.com [204.178.16.6]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id OAA11471 for <l2tp@ipsec.org>; Fri, 12 May 2000 14:08:27 -0700 (PDT)
Received: from zubin.dnrc.bell-labs.com ([135.180.130.56]) by dirty; Fri May 12 19:30:30 EDT 2000
Received: from dnrc.bell-labs.com (dhcp-6-168.pa.bell-labs.com [135.250.6.168])
	by zubin.dnrc.bell-labs.com (8.9.3/8.9.3) with ESMTP id TAA00758;
	Fri, 12 May 2000 19:30:26 -0400 (EDT)
Message-ID: <391C94E5.EF560BF0@dnrc.bell-labs.com>
Date: Fri, 12 May 2000 16:33:57 -0700
From: Grenville Armitage <gja@dnrc.bell-labs.com>
Organization: Bell Laboratories, Lucent Technologies
X-Mailer: Mozilla 4.7 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Rajesh Sundaram <rajeshsu@Exchange.Microsoft.com>
CC: diffserv@ietf.org, l2tp@ipsec.org
Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based 
 tunneling   protocols
References: <19398D273324D3118A2B0008C7E9A5690681E0DB@SIT.platinum.corp.microsoft.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit



> Rajesh Sundaram wrote:
	[..]
> The first approach seems to be a special case of the second one (where there
> all inner DSCPs map to a uniform outer DSCP). Also, the second one is more
> flexible.

Naturally - by construction, since the "use a single tunnel like
a wire" is going to be a subset of any other scheme that involves
more than one tunnel. Certainly when I suggested simply using a
fixed DSCP for each L2TP tunnel I did not intend that to be
interpreted as "you can only have one tunnel".

>  I'd appreciate comments that help us make a decision on this.

Implement something that allows flexible mapping rules.

Perhaps it would be easier to codify rules if we keep in mind that
the DSCP for the outer (tunneling) IP header is really only providing
service to the PPP session being tunneled. The PPP session(s) are
then (in the abstract at least) providing the 'wire' for the (many)
IP flows being mapped onto each PPP session. Now, as I understand
it, the premise of this thread is that you dont want reordering of
PPP frames within a single session. Simple conclusion - when an L2TP
tunnel supports a distinct PPP session, that tunnel's outer IP header
ought to have a fixed DSCP (or at least a DSCP from a PHB Group that
enforces non-reordering). Beyond that, how you choose to map
particular classes of tunneled IP traffic onto particular PPP
sessions is another discussion entirely....

cheers,
gja
________________________________________________________________________
Grenville Armitage                    http://members.home.net/garmitage/
Bell Labs Research Silicon Valley


From owner-l2tp@ipsec.org  Fri May 12 20:37:50 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA19281
	for <l2tpext-archive@odin.ietf.org>; Fri, 12 May 2000 20:37:49 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id OAA11511; Fri, 12 May 2000 14:22:08 -0700 (PDT)
Received: from amberex01.ambernetworks.com (outgate01.ambernetworks.com [207.20.187.199]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id OAA11504 for <l2tp@ipsec.org>; Fri, 12 May 2000 14:18:35 -0700 (PDT)
Received: by amberex01.ambernetworks.com with Internet Mail Service (5.5.2448.0)
	id <KJ1S5GTC>; Fri, 12 May 2000 16:43:11 -0700
Message-ID: <405CAE450410D311B26C0050047326DC6B7A51@amberex01.ambernetworks.com>
From: Andy Koscinski <ANDYK@ambernetworks.com>
To: Rajesh Sundaram <rajeshsu@Exchange.Microsoft.com>,
        Brian E Carpenter
	 <brian@hursley.ibm.com>,
        Grenville Armitage <gja@dnrc.bell-labs.com>,
        David Black <Black_David@emc.com>
Cc: diffserv@ietf.org, l2tp@ipsec.org
Subject: RE: [Diffserv] Incompatibility between Diffserv and PPP based tun
	neling   protocols
Date: Fri, 12 May 2000 16:43:10 -0700
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2448.0)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01BFBC6B.D4B4CD20"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

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_01BFBC6B.D4B4CD20
Content-Type: text/plain;
	charset="windows-1252"

 
 

-----Original Message-----
From: Rajesh Sundaram [mailto:rajeshsu@Exchange.Microsoft.com]
Sent: Friday, May 12, 2000 2:35 PM
To: Brian E Carpenter; Grenville Armitage; David Black
Cc: diffserv@ietf.org; l2tp@ipsec.org
Subject: RE: [Diffserv] Incompatibility between Diffserv and PPP based
tunneling protocols



So far, there are two proposed solutions to this problem. 

a) Mark the outer header packets of all tunneled packets with a uniform
codepoint. 

As Brian pointed out, this treats the tunnel like a virtual wire and works
well as long as there is sufficient bandwidth. But, there could be a problem
if EF bandwidth was limited. By multiplexing data and EF onto an EF tunnel,
we are increasing the probability of getting policed at a DS boundary, and
thereby breaking the guarantees of inner EF traffic.

  b) Create a (small)  

andyk> yes, but why small? should cover all required DSCP's.

> set of multiple tunnels,  

 each tunnel getting a fixed codepoint, that is distinct from other tunnels.
All packets on a particular tunnel carry the same uniform codepoint. Before
tunnel encapsulation, inspect the inner codepoint of the packet and route
the packet to the appropriate tunnel. 

andyk> I would replace the word "tunnel" by a "l2tp session" in the above
sentence. There is a "l2tp-DS-draft-extension" which suggests a negotiable
outer DSCP per session,

not per tunnel. Each microflow with an unique DSCP requirement maps into one
l2tp session with and individual outer DSCP, which migt be different from
the outer tunnel DSCP

and also might be different from the inner DSCP (or TOS) of the microflow at
tunnel ingress.

The packet sequence in every session (microflow) is preserved, if packet
sequencing is enabled. In addition the micro flow

might have a packet sequencing of it's own, on the higher layer protocol.

 

 I think the outer DSCP per l2tp session should be unidirectional,

and might be diferent for the l2tp payload flowing in two different
directions.

thx

andy 

 

 I had not necessarily suggested creating a different tunnel for each PHB
(or for each micro-flow). By creating a smaller set of tunnels, we are
reducing the management overhead. This approach would need a mapping of
inner DSCPs --> outer DSCPs, which is probably controlled by the local
tunnel administrator.

The first approach seems to be a special case of the second one (where there
all inner DSCPs map to a uniform outer DSCP). Also, the second one is more
flexible. I'd appreciate comments that help us make a decision on this. 

Thanks, 
Rajesh 

-----Original Message----- 
From: Brian E Carpenter [ mailto:brian@hursley.ibm.com
<mailto:brian@hursley.ibm.com> ] 
Sent: Thursday, May 11, 2000 1:09 PM 
To: Grenville Armitage 
Cc: Rohit Chhapolia; diffserv@ietf.org; l2tp@ipsec.org; David Black 
Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based 
tunneling protocols 


Three comments (on the whole thread): 

1. The correct place to write up the conclusion of this, when we 
reach one, is in draft-ietf-diffserv-tunnels-00.txt as revised. 

2. Remember that the diffserv rule that microflows must not 
be reordered is met by not reordering the whole BA, and is 
also met by not reordering the whole tunnel. It is completely 
conformant with diffserv to perform no reordering whatever. 
Viewed that way there is no problem, just an implementation 
constraint. 

3. We might need to add some words to RFC 2598 section 2.5, 
but I see no problem in enclosing an EF flow and other flows 
such as BE or AF within an EF tunnel, as long as the tunnel 
has adequate bandwidth and admission control. That's why we 
call it virtual wire. 

   Brian 

Grenville Armitage wrote: 
> 
> Rohit Chhapolia wrote: 
>         [..] 
> > What criteria should be used at the tunnel's ingress node to 
> > decide the traffic that can go through this tunnel? If the 
> > tunnel's DSCP is AF, can one send both EF and AF traffic 
> > through it? It may lead to degradation of service for the EF class. 
> 
> In general the answer is more like 'it depends'. Since the 
> tunnel becomes more like a link, we need to understand its 
> ingress->egress QoS characteristics before we can decide 
> how well it would support any given mix of tunneled traffic. 
> Unfortunately knowledge of the tunnel's PHB alone doesn't 
> tell us all there is to know - it will depend on network 
> topology and router provisioning along the tunnel's path. 
> This is the nature of Diffserv (to date, although people are 
> working on defining e2e behaviors built from per hop behaviors) 
> 
> cheers, 
> gja 
> ________________________________________________________________________ 
> Grenville Armitage                    http://members.home.net/garmitage/
<http://members.home.net/garmitage/>  
> Bell Labs Research Silicon Valley 
> 
> _______________________________________________ 
> diffserv mailing list 
> diffserv@ietf.org 
> http://www1.ietf.org/mailman/listinfo/diffserv
<http://www1.ietf.org/mailman/listinfo/diffserv>  
> Archive: http://www-nrg.ee.lbl.gov/diff-serv-arch/
<http://www-nrg.ee.lbl.gov/diff-serv-arch/>  

_______________________________________________ 
diffserv mailing list 
diffserv@ietf.org 
http://www1.ietf.org/mailman/listinfo/diffserv
<http://www1.ietf.org/mailman/listinfo/diffserv>  
Archive: http://www-nrg.ee.lbl.gov/diff-serv-arch/
<http://www-nrg.ee.lbl.gov/diff-serv-arch/>  


------_=_NextPart_001_01BFBC6B.D4B4CD20
Content-Type: text/html;
	charset="windows-1252"

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV="Content-Type" CONTENT="text/html; charset=windows-1252">
<TITLE>RE: [Diffserv] Incompatibility between Diffserv and PPP based tunneling protocols</TITLE>

<META content="MSHTML 5.00.2314.1000" name=GENERATOR></HEAD>
<BODY>
<DIV>&nbsp;</DIV>
<DIV>&nbsp;</DIV>
<BLOCKQUOTE 
style="BORDER-LEFT: #0000ff 2px solid; MARGIN-LEFT: 5px; MARGIN-RIGHT: 0px; PADDING-LEFT: 5px">
  <DIV align=left class=OutlookMessageHeader dir=ltr><FONT face=Tahoma 
  size=2>-----Original Message-----<BR><B>From:</B> Rajesh Sundaram 
  [mailto:rajeshsu@Exchange.Microsoft.com]<BR><B>Sent:</B> Friday, May 12, 2000 
  2:35 PM<BR><B>To:</B> Brian E Carpenter; Grenville Armitage; David 
  Black<BR><B>Cc:</B> diffserv@ietf.org; l2tp@ipsec.org<BR><B>Subject:</B> RE: 
  [Diffserv] Incompatibility between Diffserv and PPP based tunneling 
  protocols<BR><BR></DIV></FONT><!-- Converted from text/plain format -->
  <P><FONT size=2>So far, there are two proposed solutions to this problem. 
  </FONT></P>
  <P><FONT size=2>a) Mark the outer header packets of all tunneled packets with 
  a uniform codepoint. </FONT></P>
  <P><FONT size=2>As Brian pointed out, this treats the tunnel like a virtual 
  wire and works well as long as there is sufficient bandwidth. But, there could 
  be a problem if EF bandwidth was limited. By multiplexing data and EF onto an 
  EF tunnel, we are increasing the probability of getting policed at a DS 
  boundary, and thereby breaking the guarantees of inner EF traffic.</FONT></P>
  <P><FONT size=2><SPAN class=684222323-12052000><FONT color=#0000ff 
  face=Arial>&nbsp;</FONT></SPAN><SPAN class=684222323-12052000>&nbsp;</SPAN>b) 
  Create a (small)&nbsp;<SPAN class=684222323-12052000><FONT color=#0000ff 
  face=Arial>&nbsp;</FONT></SPAN></FONT></P>
  <P><FONT color=#0000ff face=Arial size=2><SPAN 
  class=684222323-12052000>andyk&gt; yes, but why small? should cover all 
  required DSCP's.</SPAN></FONT></P>
  <P><FONT size=2><SPAN class=684222323-12052000><FONT color=#0000ff 
  face=Arial>&gt;</FONT>&nbsp;</SPAN>set of multiple tunnels,&nbsp;<SPAN 
  class=684222323-12052000><FONT color=#0000ff 
  face=Arial>&nbsp;</FONT></SPAN></FONT></P>
  <P><FONT size=2><SPAN class=684222323-12052000>&nbsp;</SPAN>each tunnel 
  getting a fixed codepoint, that is distinct from other tunnels. All packets on 
  a particular tunnel carry the same uniform codepoint. Before tunnel 
  encapsulation, inspect the inner codepoint of the packet and route the packet 
  to the appropriate tunnel.<FONT color=#0000ff face=Arial><SPAN 
  class=684222323-12052000>&nbsp;</SPAN></FONT></FONT></P>
  <P><FONT size=2><FONT color=#0000ff face=Arial><SPAN 
  class=684222323-12052000>andyk&gt; I would replace&nbsp;the word "tunnel" by a 
  "l2tp session" in the above sentence.&nbsp;There is 
  a&nbsp;"l2tp-DS-draft-extension" which suggests a negotiable outer DSCP per 
  session,</SPAN></FONT></FONT></P>
  <P><FONT size=2><FONT color=#0000ff face=Arial><SPAN 
  class=684222323-12052000>not per tunnel. Each microflow with an unique DSCP 
  requirement maps into one l2tp session with and individual outer DSCP, which 
  migt be different from the outer tunnel DSCP</SPAN></FONT></FONT></P>
  <P><FONT size=2><FONT color=#0000ff face=Arial><SPAN 
  class=684222323-12052000>and also might be different from the inner DSCP (or 
  TOS) of the microflow at tunnel ingress.</SPAN></FONT></FONT></P>
  <P><FONT size=2><FONT color=#0000ff face=Arial><SPAN 
  class=684222323-12052000>The packet sequence in every session (microflow) is 
  preserved, if packet sequencing is enabled. In addition the micro 
  flow</SPAN></FONT></FONT></P>
  <P><FONT size=2><FONT color=#0000ff face=Arial><SPAN 
  class=684222323-12052000>might have a packet sequencing of it's own, on the 
  higher layer protocol.</SPAN></FONT></FONT></P>
  <P><FONT size=2><FONT color=#0000ff face=Arial><SPAN 
  class=684222323-12052000></SPAN></FONT></FONT>&nbsp;</P>
  <P><FONT size=2><FONT color=#0000ff face=Arial><SPAN 
  class=684222323-12052000>&nbsp;I think the outer DSCP per l2tp 
  session&nbsp;should be unidirectional,</SPAN></FONT></FONT></P>
  <P><FONT size=2><FONT color=#0000ff face=Arial><SPAN 
  class=684222323-12052000>and might be diferent for the l2tp payload flowing in 
  two different directions.</SPAN></FONT></FONT></P>
  <P><FONT size=2><SPAN class=684222323-12052000><FONT color=#0000ff 
  face=Arial>thx</FONT></SPAN></FONT></P>
  <P><FONT size=2><SPAN class=684222323-12052000><FONT color=#0000ff 
  face=Arial>andy</FONT>&nbsp;</SPAN></FONT></P>
  <P><FONT size=2><SPAN class=684222323-12052000></SPAN></FONT>&nbsp;</P>
  <P><FONT size=2><SPAN class=684222323-12052000>&nbsp;</SPAN>I had not 
  necessarily suggested creating a different tunnel for each PHB (or for each 
  micro-flow). By creating a smaller set of tunnels, we are reducing the 
  management overhead. This approach would need a mapping of inner DSCPs --&gt; 
  outer DSCPs, which is probably controlled by the local tunnel 
  administrator.</FONT></P>
  <P><FONT size=2>The first approach seems to be a special case of the second 
  one (where there all inner DSCPs map to a uniform outer DSCP). Also, the 
  second one is more flexible. I'd appreciate comments that help us make a 
  decision on this. </FONT></P>
  <P><FONT size=2>Thanks,</FONT> <BR><FONT size=2>Rajesh</FONT> </P>
  <P><FONT size=2>-----Original Message-----</FONT> <BR><FONT size=2>From: Brian 
  E Carpenter [<A 
  href="mailto:brian@hursley.ibm.com">mailto:brian@hursley.ibm.com</A>]</FONT> 
  <BR><FONT size=2>Sent: Thursday, May 11, 2000 1:09 PM</FONT> <BR><FONT 
  size=2>To: Grenville Armitage</FONT> <BR><FONT size=2>Cc: Rohit Chhapolia; 
  diffserv@ietf.org; l2tp@ipsec.org; David Black</FONT> <BR><FONT 
  size=2>Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP 
  based</FONT> <BR><FONT size=2>tunneling protocols</FONT> </P><BR>
  <P><FONT size=2>Three comments (on the whole thread):</FONT> </P>
  <P><FONT size=2>1. The correct place to write up the conclusion of this, when 
  we</FONT> <BR><FONT size=2>reach one, is in draft-ietf-diffserv-tunnels-00.txt 
  as revised.</FONT> </P>
  <P><FONT size=2>2. Remember that the diffserv rule that microflows must 
  not</FONT> <BR><FONT size=2>be reordered is met by not reordering the whole 
  BA, and is</FONT> <BR><FONT size=2>also met by not reordering the whole 
  tunnel. It is completely</FONT> <BR><FONT size=2>conformant with diffserv to 
  perform no reordering whatever.</FONT> <BR><FONT size=2>Viewed that way there 
  is no problem, just an implementation</FONT> <BR><FONT 
  size=2>constraint.</FONT> </P>
  <P><FONT size=2>3. We might need to add some words to RFC 2598 section 
  2.5,</FONT> <BR><FONT size=2>but I see no problem in enclosing an EF flow and 
  other flows</FONT> <BR><FONT size=2>such as BE or AF within an EF tunnel, as 
  long as the tunnel</FONT> <BR><FONT size=2>has adequate bandwidth and 
  admission control. That's why we</FONT> <BR><FONT size=2>call it virtual 
  wire.</FONT> </P>
  <P><FONT size=2>&nbsp;&nbsp; Brian</FONT> </P>
  <P><FONT size=2>Grenville Armitage wrote:</FONT> <BR><FONT size=2>&gt; 
  </FONT><BR><FONT size=2>&gt; Rohit Chhapolia wrote:</FONT> <BR><FONT 
  size=2>&gt;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; [..]</FONT> 
  <BR><FONT size=2>&gt; &gt; What criteria should be used at the tunnel's 
  ingress node to</FONT> <BR><FONT size=2>&gt; &gt; decide the traffic that can 
  go through this tunnel? If the</FONT> <BR><FONT size=2>&gt; &gt; tunnel's DSCP 
  is AF, can one send both EF and AF traffic</FONT> <BR><FONT size=2>&gt; &gt; 
  through it? It may lead to degradation of service for the EF class.</FONT> 
  <BR><FONT size=2>&gt; </FONT><BR><FONT size=2>&gt; In general the answer is 
  more like 'it depends'. Since the</FONT> <BR><FONT size=2>&gt; tunnel becomes 
  more like a link, we need to understand its</FONT> <BR><FONT size=2>&gt; 
  ingress-&gt;egress QoS characteristics before we can decide</FONT> <BR><FONT 
  size=2>&gt; how well it would support any given mix of tunneled 
  traffic.</FONT> <BR><FONT size=2>&gt; Unfortunately knowledge of the tunnel's 
  PHB alone doesn't</FONT> <BR><FONT size=2>&gt; tell us all there is to know - 
  it will depend on network</FONT> <BR><FONT size=2>&gt; topology and router 
  provisioning along the tunnel's path.</FONT> <BR><FONT size=2>&gt; This is the 
  nature of Diffserv (to date, although people are</FONT> <BR><FONT size=2>&gt; 
  working on defining e2e behaviors built from per hop behaviors)</FONT> 
  <BR><FONT size=2>&gt; </FONT><BR><FONT size=2>&gt; cheers,</FONT> <BR><FONT 
  size=2>&gt; gja</FONT> <BR><FONT size=2>&gt; 
  ________________________________________________________________________</FONT> 
  <BR><FONT size=2>&gt; Grenville 
  Armitage&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; 
  <A 
  href="http://members.home.net/garmitage/">http://members.home.net/garmitage/</A></FONT> 
  <BR><FONT size=2>&gt; Bell Labs Research Silicon Valley</FONT> <BR><FONT 
  size=2>&gt; </FONT><BR><FONT size=2>&gt; 
  _______________________________________________</FONT> <BR><FONT size=2>&gt; 
  diffserv mailing list</FONT> <BR><FONT size=2>&gt; diffserv@ietf.org</FONT> 
  <BR><FONT size=2>&gt; <A 
  href="http://www1.ietf.org/mailman/listinfo/diffserv">http://www1.ietf.org/mailman/listinfo/diffserv</A></FONT> 
  <BR><FONT size=2>&gt; Archive: <A 
  href="http://www-nrg.ee.lbl.gov/diff-serv-arch/">http://www-nrg.ee.lbl.gov/diff-serv-arch/</A></FONT> 
  </P>
  <P><FONT size=2>_______________________________________________</FONT> 
  <BR><FONT size=2>diffserv mailing list</FONT> <BR><FONT 
  size=2>diffserv@ietf.org</FONT> <BR><FONT size=2><A 
  href="http://www1.ietf.org/mailman/listinfo/diffserv">http://www1.ietf.org/mailman/listinfo/diffserv</A></FONT> 
  <BR><FONT size=2>Archive: <A 
  href="http://www-nrg.ee.lbl.gov/diff-serv-arch/">http://www-nrg.ee.lbl.gov/diff-serv-arch/</A></FONT> 
  </P></BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01BFBC6B.D4B4CD20--


From owner-l2tp@ipsec.org  Fri May 12 20:45:38 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id UAA19336
	for <l2tpext-archive@odin.ietf.org>; Fri, 12 May 2000 20:45:38 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id OAA11543; Fri, 12 May 2000 14:32:27 -0700 (PDT)
Received: from mailsrv.acc.com (mailsrv.acc.com [129.192.64.128]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id OAA11537 for <l2tp@ipsec.org>; Fri, 12 May 2000 14:31:27 -0700 (PDT)
Received: from ericsson.com (unk105122.acc.com [129.192.105.122])
	by mailsrv.acc.com (8.9.3/8.9.1) with ESMTP id QAA06169;
	Fri, 12 May 2000 16:57:23 -0700 (PDT)
Message-ID: <391C9A79.1DF83255@ericsson.com>
Date: Fri, 12 May 2000 16:57:45 -0700
From: Evan Caves <evan.caves@ericsson.com>
X-Mailer: Mozilla 4.61 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To: Rajesh Sundaram <rajeshsu@Exchange.Microsoft.com>
CC: diffserv@ietf.org, l2tp@ipsec.org
Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based 
 tunneling   protocols
References: <19398D273324D3118A2B0008C7E9A5690681E0DB@SIT.platinum.corp.microsoft.com>
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit



> Rajesh Sundaram wrote:
> 
> So far, there are two proposed solutions to this problem.
> 
> a) Mark the outer header packets of all tunneled packets with a
> uniform codepoint.
> 
> As Brian pointed out, this treats the tunnel like a virtual wire and
> works well as long as there is sufficient bandwidth. But, there could
> be a problem if EF bandwidth was limited. By multiplexing data and EF
> onto an EF tunnel, we are increasing the probability of getting
> policed at a DS boundary, and thereby breaking the guarantees of inner
> EF traffic.

For L2 tunneling protocols this seems to be the only option. For example
an L2TP session would be created within a tunnel that provides the per
hop
behavior for a particular dialin user based on some pre-defined policy.
There
may be multiple tunnels between the same two endpoints, each providing a
different service level. However there could be microflows within the
session
which themselves are marked differently but are given the outer mark
PHB.
The microflows within the session cannot be 're-routed' down a different
tunnel that maps to the appropriate PHB. As an aside L2TP
implementations (at
least at the LAC ingress to the tunnel) may 'snoop' the PPP control
frames
in order to manage payload sequencing but usually do not goes as far as
'snooping' the IP header.

evan
-

> 
> b) Create a (small) set of multiple tunnels, each tunnel getting a
> fixed codepoint, that is distinct from other tunnels. All packets on a
> particular tunnel carry the same uniform codepoint. Before tunnel
> encapsulation, inspect the inner codepoint of the packet and route the
> packet to the appropriate tunnel.
> 
> I had not necessarily suggested creating a different tunnel for each
> PHB (or for each micro-flow). By creating a smaller set of tunnels, we
> are reducing the management overhead. This approach would need a
> mapping of inner DSCPs --> outer DSCPs, which is probably controlled
> by the local tunnel administrator.
> 
> The first approach seems to be a special case of the second one (where
> there all inner DSCPs map to a uniform outer DSCP). Also, the second
> one is more flexible. I'd appreciate comments that help us make a
> decision on this.
> 
> Thanks,
> Rajesh
> 
> -----Original Message-----
> From: Brian E Carpenter [mailto:brian@hursley.ibm.com]
> Sent: Thursday, May 11, 2000 1:09 PM
> To: Grenville Armitage
> Cc: Rohit Chhapolia; diffserv@ietf.org; l2tp@ipsec.org; David Black
> Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based
> 
> tunneling protocols
> 
> Three comments (on the whole thread):
> 
> 1. The correct place to write up the conclusion of this, when we
> reach one, is in draft-ietf-diffserv-tunnels-00.txt as revised.
> 
> 2. Remember that the diffserv rule that microflows must not
> be reordered is met by not reordering the whole BA, and is
> also met by not reordering the whole tunnel. It is completely
> conformant with diffserv to perform no reordering whatever.
> Viewed that way there is no problem, just an implementation
> constraint.
> 
> 3. We might need to add some words to RFC 2598 section 2.5,
> but I see no problem in enclosing an EF flow and other flows
> such as BE or AF within an EF tunnel, as long as the tunnel
> has adequate bandwidth and admission control. That's why we
> call it virtual wire.
> 
>    Brian
> 
> Grenville Armitage wrote:
> >
> > Rohit Chhapolia wrote:
> >         [..]
> > > What criteria should be used at the tunnel's ingress node to
> > > decide the traffic that can go through this tunnel? If the
> > > tunnel's DSCP is AF, can one send both EF and AF traffic
> > > through it? It may lead to degradation of service for the EF
> class.
> >
> > In general the answer is more like 'it depends'. Since the
> > tunnel becomes more like a link, we need to understand its
> > ingress->egress QoS characteristics before we can decide
> > how well it would support any given mix of tunneled traffic.
> > Unfortunately knowledge of the tunnel's PHB alone doesn't
> > tell us all there is to know - it will depend on network
> > topology and router provisioning along the tunnel's path.
> > This is the nature of Diffserv (to date, although people are
> > working on defining e2e behaviors built from per hop behaviors)
> >
> > cheers,
> > gja
> >
> ________________________________________________________________________
> 
> > Grenville Armitage
> http://members.home.net/garmitage/
> > Bell Labs Research Silicon Valley
> >
> > _______________________________________________
> > diffserv mailing list
> > diffserv@ietf.org
> > http://www1.ietf.org/mailman/listinfo/diffserv
> > Archive: http://www-nrg.ee.lbl.gov/diff-serv-arch/
> 
> _______________________________________________
> diffserv mailing list
> diffserv@ietf.org
> http://www1.ietf.org/mailman/listinfo/diffserv
> Archive: http://www-nrg.ee.lbl.gov/diff-serv-arch/


From owner-l2tp@ipsec.org  Fri May 12 21:28:24 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA19651
	for <l2tpext-archive@odin.ietf.org>; Fri, 12 May 2000 21:28:24 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id PAA11597; Fri, 12 May 2000 15:10:48 -0700 (PDT)
Received: from sj-msg-core-1.cisco.com (sj-msg-core-1.cisco.com [171.71.163.11]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id PAA11590 for <l2tp@ipsec.org>; Fri, 12 May 2000 15:10:33 -0700 (PDT)
Received: from airborne.cisco.com (airborne.cisco.com [171.71.154.32])
	by sj-msg-core-1.cisco.com (8.9.3/8.9.1) with ESMTP id RAA24639;
	Fri, 12 May 2000 17:33:44 -0700 (PDT)
Received: from sbrim-nt.cisco.com (sjck-dial-gw5-174.cisco.com [10.19.238.175])
	by airborne.cisco.com (Mirapoint)
	with ESMTP id AAQ03033;
	Fri, 12 May 2000 17:33:18 -0700 (PDT)
Message-Id: <4.3.2.3.2.20000512163340.00b44290@airborne.cisco.com>
X-Sender: sbrim@airborne.cisco.com
X-Mailer: QUALCOMM Windows Eudora Version 4.3.2.3 (Beta)
Date: Fri, 12 May 2000 17:32:46 -0700
To: "Rajesh Sundaram" <rajeshsu@Exchange.Microsoft.com>
From: Scott W Brim <sbrim@cisco.com>
Subject: RE: [Diffserv] Incompatibility between Diffserv and PPP based
  tunneling   protocols
Cc: <diffserv@ietf.org>, <l2tp@ipsec.org>
In-Reply-To: <19398D273324D3118A2B0008C7E9A5690681E0DB@SIT.platinum.corp
 .microsoft.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"; format=flowed
Sender: owner-l2tp@ipsec.org
Precedence: bulk

At 14:34 05/12/2000 -0700, Rajesh Sundaram wrote:
>I had not necessarily suggested creating a different tunnel for each PHB 
>(or for each micro-flow). By creating a smaller set of tunnels, we are 
>reducing the management overhead. This approach would need a mapping of 
>inner DSCPs --> outer DSCPs, which is probably controlled by the local 
>tunnel administrator.

If you're just distinguishing according to diff-serv, you want at most a 
tunnel per per-hop scheduling class (not PHB).  There are other 
multipliers which might come into play, though.  For example, if you have 
VPNs you might need to a way to distinguish them (e.g. by using different 
tunnels) in order to do fair sharing for your customers.  The complexity 
of what you need to do depends directly on the complexity of the service 
differentiation you want to deliver.  



From owner-l2tp@ipsec.org  Fri May 12 23:22:10 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA22077
	for <l2tpext-archive@odin.ietf.org>; Fri, 12 May 2000 23:22:10 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id QAA11701; Fri, 12 May 2000 16:02:34 -0700 (PDT)
Received: from drawbridge.ascend.com (drawbridge.ascend.com [198.4.92.1]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id QAA11697 for <l2tp@ipsec.org>; Fri, 12 May 2000 16:02:19 -0700 (PDT)
Received: from fw-ext.ascend.com (fw-ext [198.4.92.5])
	by drawbridge.ascend.com (8.9.1a/8.9.1) with SMTP id SAA19384;
	Fri, 12 May 2000 18:19:00 -0700 (PDT)
Received: from russet.ascend.com by fw-ext.ascend.com
          via smtpd (for drawbridge.ascend.com [198.4.92.1]) with SMTP; 13 May 2000 01:25:53 UT
Received: from wopr.eng.ascend.com (wopr.eng.ascend.com [206.65.212.178])
	by russet.ascend.com (8.9.1a/8.9.1) with ESMTP id SAA28523;
	Fri, 12 May 2000 18:25:52 -0700 (PDT)
Received: from scamp.eng.ascend.com (scamp.eng.ascend.com [135.140.53.42])
	by wopr.eng.ascend.com (8.9.1/8.9.1) with ESMTP id SAA11141;
	Fri, 12 May 2000 18:25:53 -0700 (PDT)
Received: from igoyret-pc.eng.ascend.com by scamp.eng.ascend.com (8.8.8+Sun/SMI-SVR4)
	id SAA00322; Fri, 12 May 2000 18:25:51 -0700 (PDT)
Message-Id: <3.0.5.32.20000512182622.00b079a0@scamp.eng.ascend.com>
X-Sender: igoyret@scamp.eng.ascend.com
X-Mailer: QUALCOMM Windows Eudora Pro Version 3.0.5 (32)
Date: Fri, 12 May 2000 18:26:22 -0700
To: Evan Caves <evan.caves@ericsson.com>
From: Ignacio Goyret <igoyret@lucent.com>
Subject: Re: [Diffserv] Incompatibility between Diffserv and PPP based 
  tunneling   protocols
Cc: Rajesh Sundaram <rajeshsu@Exchange.Microsoft.com>, diffserv@ietf.org,
        l2tp@ipsec.org
In-Reply-To: <391C9A79.1DF83255@ericsson.com>
References: <19398D273324D3118A2B0008C7E9A5690681E0DB@SIT.platinum.corp.microsoft.com>
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

At 04:57 PM 5/12/00 -0700, Evan Caves wrote:
>As an aside L2TP implementations (at least at the LAC ingress to the
>tunnel) may 'snoop' the PPP control frames in order to manage payload
>sequencing but usually do not goes as far as 'snooping' the IP header.

And even if they attempted to, they will fail miserably sooner or later
because of things like PPP compression, encryption, and MP (you may be
looking at just one link out of many and you only look at one fragment
of the whole PPP packet).

-Ignacio

-----------------------------------------------------------------------
Ignacio Goyret, Sr. Software Engineer             Voice: (510) 747-2823
Lucent Technologies                               FAX:   (510) 747-2859
1701 Harbor Bay Parkway, Alameda, CA 94502


From owner-l2tp@ipsec.org  Thu May 18 21:54:33 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA16393
	for <l2tpext-archive@odin.ietf.org>; Thu, 18 May 2000 21:54:32 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id OAA27672; Thu, 18 May 2000 14:54:25 -0700 (PDT)
Received: from dfssl.exchange.microsoft.com ([131.107.88.59]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id OAA27668 for <l2tp@ipsec.org>; Thu, 18 May 2000 14:52:07 -0700 (PDT)
Received: from 172.30.236.11 by dfssl.exchange.microsoft.com (InterScan E-Mail VirusWall NT); Thu, 18 May 2000 17:08:20 -0700 (Pacific Daylight Time)
Received: from fifi.platinum.corp.microsoft.com ([172.30.236.94]) by yuri.dns.microsoft.com with Microsoft SMTPSVC(5.0.2195.2023);
	 Thu, 18 May 2000 17:13:49 -0700
content-class: urn:content-classes:message
Subject: L2TP-IPSec features list
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01BFC126.CDB4A7F0"
Date: Thu, 18 May 2000 17:11:39 -0700
Message-ID: <6A05D00595BE644E9F435BE5947423F2A21E28@fifi.platinum.corp.microsoft.com>
X-MimeOLE: Produced By Microsoft Exchange V6.0.4368.4
Thread-Topic: L2TP-IPSec features list
Thread-Index: Ab/BJv4Z2y1xvh2+Ti2+eZWeNfOxSA==
From: "William Dixon" <wdixon@Exchange.Microsoft.com>
To: <l2tp@ipsec.org>
X-OriginalArrivalTime: 19 May 2000 00:13:49.0226 (UTC) FILETIME=[1B1A58A0:01BFC127]
Sender: owner-l2tp@ipsec.org
Precedence: bulk

This is a multi-part message in MIME format.

------_=_NextPart_001_01BFC126.CDB4A7F0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

L2TP list fyi.

Here is an expansion of the features Skip noted earlier in a deep thread
for L2TP/IPSec from a protocol design view, not what any particular one
vendor implements.  No particular order.  Not a comparison with IPSec
tunnel mode except for a couple of points. (bcc'ed to ipsec and ipsra WG
lists to avoid reply-all)

L2TP benefits ---------
1) L2TP supports 4 VPN scenarios
  a.) legacy client PPP dial to NAS - NAS PPP user auth, NAS forwards
PPP through L2TP tunnel to L2TP gateway=20
  b.) L2TP/IPSec capable client PPP dial to NAS - voluntary L2TP/IPSec
to L2TP gateway
  c.) L2TP/IPSec capable client directly on IP network - user voluntary
L2TP/IPSec tunnel to L2TP gateway
  d.) L2TP/IPSec gateway to gateway
     (2) PPP options can be varied depending on capabilities of IPSec
SA, since the SA is established first - e.g. encryption & IPCOMP
compression on SA means no PPP compression needed
     (1) idea of a user for tunnel helps scalable management of remote
tunnel end-points via radius.
2) can have multiple L2TP tunnel sessions inside 1 IPSec SA pair for
efficiency
3) can have single L2TP tunnel session (choose UDP source port) mapping
to single IPSec SA pair - preserves binding of SA to tunnel to user - to
the extent you can control what user traffic goes into tunnel to start
with.
4) keep alives for L2TP
5) don't have to use L2TP sequence numbers (can't with L2TP header
compression) because IPSec seq numbers can detect packet loss & estimate
path latency
6) new spec for L2TP MUX of PPP packets inside a single L2TP UDP
datagram - popular to reduce overhead for tunneling small IP packets,
like VoIP
7) new L2TP header compression draft gets [UDP][L2TP] header overhead
down to 1 or 2 bytes for majority of the tunnel.  This is 1-2 bytes per
packet more than pure IPSec tunnel mode - using PPPMUX mode of L2TP
means several PPP frames per tunnel IP packet
8) extensibility of L2TP with Attrib Value (AV) pairs, avoid extending,
complicating IKE key management protocol
  a.) AV delivered dynamic redirection of gateway dest IP for global
load balancing
  b.) AV delivered QOS values
9) can use IPSec tunnel and transport filters which apply to assigned
tunnel IP address above L2TP/PPP interface
  a.) enables tunnel within a tunnel with full interface support
  b.) enables static filtering & routes applied via Radius policy to
tunnel interfaces
10) basic L2TP is most interoperable - even before IKE & IPSec, RFC
solution designed specifically for remote access, dial remote access in
particular - which estimates show to be 60-70% of the worldwide Internet
connection method (PPP in general by all platforms, not L2TP/IPSec) in
2003.
11) more vendors testing L2TP/IPSec interop (14) at last bakeoff than
testing XAUTH/IKECFG (7)
12) use of MM/QM in transport mode makes IPSec policy consistently
transport - just an ease of maintenance thing.  So IKE negotiation for
L2TP (UDP port 1701) protections are just another port QM, as are QMs
for other things like connections to application proxies that run on the
same box.  Your transport packet filters for permit & block & secure are
easy to see and evaluate.  You don't have to worry about a tunnel policy
for access from any Client IP to an internal subnet in conflict with
transport filters to block a specific type of traffic regardless of
addresses.=20
13) Some vendors implement IPSec tunnels as routable interfaces, some
implement as strictly filter driven - don't have to worry about this for
L2TP - they are all interfaces.
  a.) can run routing protocols across tunnel interoperable with other
vendors
  b.) can carry standard, routable IP across tunnel, e.g. multicast,
broadcast, unicast IP
14) L2TP can be used in clear text mode without IPSec and get all of the
tunneling benefits when security is not a problem.
15) Without IPSec, L2TP over IP can go through NATs because it uses UDP
1701.
16) L2TP/IPSec=20
17) L2TP/IPSec VPN clients are becoming widely available by their
inclusion in Windows 2000 and future releases.  Their availability for
VPN has been announced since August 1998 when the L2TP/IPSec client was
included in Beta2 of Windows 2000.
18) Most remote access implementations support both machine auth first
with either preshared key, or much stronger certificate auth, then user
auth
19) L2TP/IPSec fully replaces and is a superset of PPTP functionality
and is much more secure
20) L2TP/IPSec today offers more features as RFC standard than IPSec
tunnel mode does currently
21) L2TP is defined to run over ATM, frame relay and X.25 non-IP
transports
22) IPSec encryption & integrity protection and IPCOMP can be hardware
accelerated


PPP Benefits ---------- all RFC standard methods, I believe
1) IP/DNS/WINs assignment through IPCP
2) User authentication
  a.) userid/password
  b.) EAP and EAP/TLS - for certificate, smartcard, SecurID, token card,
biometric
3) AAA integration - with existing AAA servers, RADIUS TACACS
4) multi-protocol support
5) Link layer negotiation for MRU (this may help in the fragmentation
arena)
6) Keepalives for both PPP
7) Multilink PPP for bundling bandwidth
8) Routable interface
9) PPP compression
10) supports idea of demand dial, so you can have demand dial filters
that kick off the PPP dial activity, and different filters which apply
above the interface=20
11) supports idea of persistent and autostart connections, so links can
come up automatically and be maintained

Wm

------_=_NextPart_001_01BFC126.CDB4A7F0
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
6.0.4368.3">
<TITLE>L2TP-IPSec features list</TITLE>
</HEAD>
<BODY>
<!-- Converted from text/plain format -->

<P><FONT SIZE=3D2>L2TP list fyi.</FONT>
</P>

<P><FONT SIZE=3D2>Here is an expansion of the features Skip noted =
earlier in a deep thread for L2TP/IPSec from a protocol design view, not =
what any particular one vendor implements.&nbsp; No particular =
order.&nbsp; Not a comparison with IPSec tunnel mode except for a couple =
of points. (bcc'ed to ipsec and ipsra WG lists to avoid =
reply-all)</FONT></P>

<P><FONT SIZE=3D2>L2TP benefits ---------</FONT>

<BR><FONT SIZE=3D2>1) L2TP supports 4 VPN scenarios</FONT>

<BR><FONT SIZE=3D2>&nbsp; a.) legacy client PPP dial to NAS - NAS PPP =
user auth, NAS forwards PPP through L2TP tunnel to L2TP gateway </FONT>

<BR><FONT SIZE=3D2>&nbsp; b.) L2TP/IPSec capable client PPP dial to NAS =
- voluntary L2TP/IPSec to L2TP gateway</FONT>

<BR><FONT SIZE=3D2>&nbsp; c.) L2TP/IPSec capable client directly on IP =
network - user voluntary L2TP/IPSec tunnel to L2TP gateway</FONT>

<BR><FONT SIZE=3D2>&nbsp; d.) L2TP/IPSec gateway to gateway</FONT>

<BR><FONT SIZE=3D2>&nbsp;&nbsp;&nbsp;&nbsp; (2) PPP options can be =
varied depending on capabilities of IPSec SA, since the SA is =
established first - e.g. encryption &amp; IPCOMP compression on SA means =
no PPP compression needed</FONT></P>

<P><FONT SIZE=3D2>&nbsp;&nbsp;&nbsp;&nbsp; (1) idea of a user for tunnel =
helps scalable management of remote tunnel end-points via radius.</FONT>

<BR><FONT SIZE=3D2>2) can have multiple L2TP tunnel sessions inside 1 =
IPSec SA pair for efficiency</FONT>

<BR><FONT SIZE=3D2>3) can have single L2TP tunnel session (choose UDP =
source port) mapping to single IPSec SA pair - preserves binding of SA =
to tunnel to user - to the extent you can control what user traffic goes =
into tunnel to start with.</FONT></P>

<P><FONT SIZE=3D2>4) keep alives for L2TP</FONT>

<BR><FONT SIZE=3D2>5) don't have to use L2TP sequence numbers (can't =
with L2TP header compression) because IPSec seq numbers can detect =
packet loss &amp; estimate path latency</FONT></P>

<P><FONT SIZE=3D2>6) new spec for L2TP MUX of PPP packets inside a =
single L2TP UDP datagram - popular to reduce overhead for tunneling =
small IP packets, like VoIP</FONT></P>

<P><FONT SIZE=3D2>7) new L2TP header compression draft gets [UDP][L2TP] =
header overhead down to 1 or 2 bytes for majority of the tunnel.&nbsp; =
This is 1-2 bytes per packet more than pure IPSec tunnel mode - using =
PPPMUX mode of L2TP means several PPP frames per tunnel IP =
packet</FONT></P>

<P><FONT SIZE=3D2>8) extensibility of L2TP with Attrib Value (AV) pairs, =
avoid extending, complicating IKE key management protocol</FONT>

<BR><FONT SIZE=3D2>&nbsp; a.) AV delivered dynamic redirection of =
gateway dest IP for global load balancing</FONT>

<BR><FONT SIZE=3D2>&nbsp; b.) AV delivered QOS values</FONT>

<BR><FONT SIZE=3D2>9) can use IPSec tunnel and transport filters which =
apply to assigned tunnel IP address above L2TP/PPP interface</FONT>

<BR><FONT SIZE=3D2>&nbsp; a.) enables tunnel within a tunnel with full =
interface support</FONT>

<BR><FONT SIZE=3D2>&nbsp; b.) enables static filtering &amp; routes =
applied via Radius policy to tunnel interfaces</FONT>

<BR><FONT SIZE=3D2>10) basic L2TP is most interoperable - even before =
IKE &amp; IPSec, RFC solution designed specifically for remote access, =
dial remote access in particular - which estimates show to be 60-70% of =
the worldwide Internet connection method (PPP in general by all =
platforms, not L2TP/IPSec) in 2003.</FONT></P>

<P><FONT SIZE=3D2>11) more vendors testing L2TP/IPSec interop (14) at =
last bakeoff than testing XAUTH/IKECFG (7)</FONT>

<BR><FONT SIZE=3D2>12) use of MM/QM in transport mode makes IPSec policy =
consistently transport - just an ease of maintenance thing.&nbsp; So IKE =
negotiation for L2TP (UDP port 1701) protections are just another port =
QM, as are QMs for other things like connections to application proxies =
that run on the same box.&nbsp; Your transport packet filters for permit =
&amp; block &amp; secure are easy to see and evaluate.&nbsp; You don't =
have to worry about a tunnel policy for access from any Client IP to an =
internal subnet in conflict with transport filters to block a specific =
type of traffic regardless of addresses. </FONT></P>

<P><FONT SIZE=3D2>13) Some vendors implement IPSec tunnels as routable =
interfaces, some implement as strictly filter driven - don't have to =
worry about this for L2TP - they are all interfaces.</FONT></P>

<P><FONT SIZE=3D2>&nbsp; a.) can run routing protocols across tunnel =
interoperable with other vendors</FONT>

<BR><FONT SIZE=3D2>&nbsp; b.) can carry standard, routable IP across =
tunnel, e.g. multicast, broadcast, unicast IP</FONT>

<BR><FONT SIZE=3D2>14) L2TP can be used in clear text mode without IPSec =
and get all of the tunneling benefits when security is not a =
problem.</FONT></P>

<P><FONT SIZE=3D2>15) Without IPSec, L2TP over IP can go through NATs =
because it uses UDP 1701.</FONT>

<BR><FONT SIZE=3D2>16) L2TP/IPSec </FONT>

<BR><FONT SIZE=3D2>17) L2TP/IPSec VPN clients are becoming widely =
available by their inclusion in Windows 2000 and future releases.&nbsp; =
Their availability for VPN has been announced since August 1998 when the =
L2TP/IPSec client was included in Beta2 of Windows 2000.</FONT></P>

<P><FONT SIZE=3D2>18) Most remote access implementations support both =
machine auth first with either preshared key, or much stronger =
certificate auth, then user auth</FONT></P>

<P><FONT SIZE=3D2>19) L2TP/IPSec fully replaces and is a superset of =
PPTP functionality and is much more secure</FONT>

<BR><FONT SIZE=3D2>20) L2TP/IPSec today offers more features as RFC =
standard than IPSec tunnel mode does currently</FONT>

<BR><FONT SIZE=3D2>21) L2TP is defined to run over ATM, frame relay and =
X.25 non-IP transports</FONT>

<BR><FONT SIZE=3D2>22) IPSec encryption &amp; integrity protection and =
IPCOMP can be hardware accelerated</FONT>
</P>
<BR>

<P><FONT SIZE=3D2>PPP Benefits ---------- all RFC standard methods, I =
believe</FONT>

<BR><FONT SIZE=3D2>1) IP/DNS/WINs assignment through IPCP</FONT>

<BR><FONT SIZE=3D2>2) User authentication</FONT>

<BR><FONT SIZE=3D2>&nbsp; a.) userid/password</FONT>

<BR><FONT SIZE=3D2>&nbsp; b.) EAP and EAP/TLS - for certificate, =
smartcard, SecurID, token card, biometric</FONT>

<BR><FONT SIZE=3D2>3) AAA integration - with existing AAA servers, =
RADIUS TACACS</FONT>

<BR><FONT SIZE=3D2>4) multi-protocol support</FONT>

<BR><FONT SIZE=3D2>5) Link layer negotiation for MRU (this may help in =
the fragmentation arena)</FONT>

<BR><FONT SIZE=3D2>6) Keepalives for both PPP</FONT>

<BR><FONT SIZE=3D2>7) Multilink PPP for bundling bandwidth</FONT>

<BR><FONT SIZE=3D2>8) Routable interface</FONT>

<BR><FONT SIZE=3D2>9) PPP compression</FONT>

<BR><FONT SIZE=3D2>10) supports idea of demand dial, so you can have =
demand dial filters that kick off the PPP dial activity, and different =
filters which apply above the interface </FONT></P>

<P><FONT SIZE=3D2>11) supports idea of persistent and autostart =
connections, so links can come up automatically and be maintained</FONT>
</P>

<P><FONT SIZE=3D2>Wm</FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01BFC126.CDB4A7F0--


From owner-l2tp@ipsec.org  Wed May 24 12:19:38 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA08283
	for <l2tpext-archive@odin.ietf.org>; Wed, 24 May 2000 12:19:36 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id EAA13140; Wed, 24 May 2000 04:41:12 -0700 (PDT)
Received: from monitor.internaut.com (mg136-128.ricochet.net [204.179.136.128]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id EAA13133 for <l2tp@ipsec.org>; Wed, 24 May 2000 04:38:12 -0700 (PDT)
Received: from vaiobean (vaiobean.ntdev.microsoft.com [204.57.137.66] (may be forged))
	by monitor.internaut.com (8.9.2/8.8.8) with SMTP id GAA31048;
	Wed, 24 May 2000 06:44:39 -0700 (PDT)
Reply-To: <aboba@internaut.com>
From: "Bernard Aboba" <aboba@internaut.com>
To: <lll@shastanets.com>, <l2tp@ipsec.org>
Cc: <gwz@cisco.com>, <ebooth@cisco.com>, <wdixon@MICROSOFT.com>,
        <townsley@cisco.com>
Subject: RE: Comments/questions to l2tp over ipsec draft
Date: Wed, 24 May 2000 07:07:03 -0700
Message-ID: <006e01bfc589$57b849f0$428939cc@ntdev.microsoft.com>
MIME-Version: 1.0
Content-Type: multipart/alternative;
	boundary="----=_NextPart_000_006F_01BFC54E.AB5C7F30"
X-Priority: 3 (Normal)
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook CWS, Build 9.0.2416 (9.0.2910.0)
In-Reply-To: <078292D50C98D2118D090008C7E9C6A608B0BBF5@STAY.platinum.corp.microsoft.com>
X-Mimeole: Produced By Microsoft MimeOLE V5.50.4131.1600
Importance: Normal
Sender: owner-l2tp@ipsec.org
Precedence: bulk

This is a multi-part message in MIME format.

------=_NextPart_000_006F_01BFC54E.AB5C7F30
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit

FW: Comments/questions to l2tp over ipsec draft > 1. I think sometimes L2TP
is unnecessarily tightly coupled with IPSEC.

  We discussed this during the L2TPEXT WG meeting in Adelaide and

  I believe we came to the same conclusion.

   > If the L2TP tunnel goes down, there wouldn't be anymore L2TP traffic on
that
   > IPSEC SA. IPSEC will detect such idleness through some idle timer and
bring
   > down that SA or its SA lifetime will eventually expire (the idle timer
is of
   > course more efficient).

  The concern was about leaving lots of zombie IPSEC SAs on the LNS.

  I think that the idle timer idea handles this though.

   > In other words, we don't need L2TP and IPSEC to communicate with one
another
   > when either one goes down.

  I agree that this is not a requirement. Should we change

  the language in the draft to MAY?

   > Can't the initiator have "src Any-Port" in its outbound-1 filter?

  This would mean that the client would *always* speak IPSEC

  to the LNS, on any port. It is simpler, as you state. Can anyone

  think of a scenario in which this would cause problems?

   > 2. Sections 9.2.x. Could you rename the word "filter" to "security
policy"
   > or "IPSEC policy"?

  Yes. This language would be clearer.

   > 3. Definition of "Voluntary Tunneling".
   > Why is the "gateway to gateway scenario" considered voluntary
tunneling?
   > Shouldn't it be compulsory tunneling?

  I think it should be considered compulsory tunneling, yes.

   > 4. Does L2TP really allow a responder to select any IP addr?

  Yes.

  > Do you know of any vendors doing it?

  Not at the moment.

   > Also, what would their reasons be?

  To support load balancing between LNSes?

   > I guess not using 1701 as the port is also unusual, isn't it?

  There are vendors who float the LNS port, though I think it is

  fair to say they are in the minority. Floating the client source

  port is quite common (though Windows 2000 doesn't do this).

   > For load balancing, if the IP addresses are different, you would end
   > up with more than one IKE phase 1, won't you?

  Yes.

  > So I don't see why the L2TP IP addr changing is related to load
balancing.

  Perhaps someone who is using this or intends to do so could explain the
utility?

   > 6. How come the draft doesn't require that IPSEC in transport mode be
used
   > (unless i missed it)? It's probably worth mentioning for
interoperability.

  This is indeed required, though the words "transport mode" are not used in

  the draft. I agree that it is probably worth mentioning.

   > 7. Section 9.2.3. Why not disallow the responder from modifying an IP
   > address when running L2TP over IPSEC? How about treating this as a
   > misconfiguration? If we rarely run into such situation, it may not be
worth
   > supporting some IP addr change detection, dynamically manipulating the
IPSEC
   > filters, etc.

   I'll leave it to those who feel they need this feature to respond.


------=_NextPart_000_006F_01BFC54E.AB5C7F30
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.0 Transitional//EN">
<HTML><HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<TITLE>FW: Comments/questions to l2tp over ipsec draft</TITLE>

<META content=3D"MSHTML 5.00.3103.1000" name=3DGENERATOR></HEAD>
<BODY>
<BLOCKQUOTE style=3D"MARGIN-RIGHT: 0px">
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>1. I think sometimes L2TP =
is=20
  unnecessarily tightly coupled with IPSEC.&nbsp;<FONT color=3D#0000ff=20
  face=3DArial><SPAN =
class=3D498444613-24052000>&nbsp;</SPAN></FONT></FONT></P>
  <P><FONT size=3D2><FONT color=3D#0000ff face=3DArial><SPAN=20
  class=3D498444613-24052000>We discussed this during the L2TPEXT WG =
meeting in=20
  Adelaide and</SPAN></FONT></FONT></P>
  <P><FONT size=3D2><FONT color=3D#0000ff face=3DArial><SPAN=20
  class=3D498444613-24052000>I believe we came to the same=20
  conclusion.&nbsp;</SPAN></FONT></FONT><FONT size=3D2><FONT =
color=3D#0000ff=20
  face=3DArial><SPAN =
class=3D498444613-24052000>&nbsp;</SPAN></FONT></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>If the L2TP tunnel goes =
down, there=20
  wouldn't be anymore L2TP traffic on that</FONT>&nbsp;<BR><FONT =
size=3D2><SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff =
face=3DArial>&nbsp;<FONT=20
  color=3D#000000 face=3D"Times New Roman"=20
  size=3D3>&gt;</FONT>&nbsp;</FONT></SPAN>IPSEC SA. IPSEC will detect =
such=20
  idleness through some idle timer and bring</FONT>&nbsp;<BR><FONT =
size=3D2><SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff =
face=3DArial>&nbsp;<FONT=20
  color=3D#000000 face=3D"Times New Roman"=20
  size=3D3>&gt;</FONT>&nbsp;</FONT></SPAN>down that SA or its SA =
lifetime will=20
  eventually expire (the idle timer is of</FONT>&nbsp;<BR><FONT =
size=3D2><SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff =
face=3DArial>&nbsp;<FONT=20
  color=3D#000000 face=3D"Times New =
Roman">&gt;</FONT>&nbsp;</FONT></SPAN>course=20
  more efficient).<SPAN class=3D498444613-24052000><FONT color=3D#0000ff =

  face=3DArial>&nbsp;</FONT></SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>The=20
  concern was about leaving lots of zombie IPSEC SAs on the LNS.=20
  </SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>I=20
  think that the idle timer idea handles this though. </SPAN></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>In other words, we don't =
need L2TP=20
  and IPSEC to communicate with one another</FONT>&nbsp;<BR><FONT =
size=3D2><SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff =
face=3DArial>&nbsp;<FONT=20
  color=3D#000000 face=3D"Times New =
Roman">&gt;</FONT>&nbsp;</FONT></SPAN>when=20
  either one goes down.&nbsp;<SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;</FONT></SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>I=20
  agree that this is not a requirement. Should we =
change</SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>the=20
  language in the draft to MAY?</SPAN></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>Can't the initiator have =
"src</FONT>=20
  <FONT size=3D2>Any-Port" in its outbound-1 filter?&nbsp;<SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff=20
  face=3DArial>&nbsp;</FONT></SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>This=20
  would mean that the client would *always* speak =
IPSEC</SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>to the=20
  LNS, on any port. It is simpler, as you state. Can =
anyone</SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>think=20
  of a scenario in which this would cause problems? </SPAN></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>2. Sections 9.2.x. Could =
you rename=20
  the word "filter" to "security policy"</FONT>&nbsp;<BR><FONT =
size=3D2><SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff =
face=3DArial>&nbsp;<FONT=20
  color=3D#000000 face=3D"Times New Roman" =
size=3D3>&gt;</FONT>&nbsp;</FONT></SPAN>or=20
  "IPSEC policy"?&nbsp;<FONT color=3D#0000ff face=3DArial><SPAN=20
  class=3D498444613-24052000>&nbsp;</SPAN></FONT></FONT></P>
  <P><FONT size=3D2><FONT color=3D#0000ff face=3DArial><SPAN=20
  class=3D498444613-24052000>Yes. This language would be clearer.=20
  </SPAN></FONT></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>3. Definition of "Voluntary =

  Tunneling".</FONT>&nbsp;<BR><FONT size=3D2><SPAN =
class=3D498444613-24052000><FONT=20
  color=3D#0000ff face=3DArial>&nbsp;<FONT color=3D#000000 face=3D"Times =
New Roman"=20
  size=3D3>&gt;</FONT>&nbsp;</FONT></SPAN>Why is the "gateway to gateway =
scenario"=20
  considered voluntary tunneling?</FONT>&nbsp;<BR><FONT size=3D2><SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff =
face=3DArial>&nbsp;<FONT=20
  color=3D#000000 face=3D"Times New Roman"=20
  size=3D3>&gt;</FONT>&nbsp;</FONT></SPAN>Shouldn't it be compulsory=20
  tunneling?</FONT>&nbsp;<FONT color=3D#0000ff face=3DArial =
size=3D2><SPAN=20
  class=3D498444613-24052000>&nbsp;</SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>I=20
  think it should be considered compulsory tunneling, yes.=20
  &nbsp;</SPAN></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>4. Does L2TP really allow a =
responder=20
  to select any IP addr?&nbsp;<SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;</FONT></SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>Yes.=20
  </SPAN></FONT><FONT size=3D2></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&gt;&nbsp;Do&nbsp;</FONT></SPAN>you know of any =
vendors</FONT>=20
  <FONT size=3D2>doing it?&nbsp;<SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;</FONT></SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>Not at=20
  the moment. &nbsp;</SPAN></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>Also, what would their =
reasons=20
  be?&nbsp;<SPAN class=3D498444613-24052000><FONT color=3D#0000ff=20
  face=3DArial>&nbsp;</FONT></SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>To=20
  support&nbsp;load balancing between LNSes?</SPAN></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>I guess not using 1701 as =
the</FONT>=20
  <FONT size=3D2>port is also unusual, isn't it?<SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff=20
  face=3DArial>&nbsp;</FONT></SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>There=20
  are vendors who float the LNS port, though I think it =
is</SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>fair=20
  to say they are in the minority. Floating the client =
source</SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>port=20
  is quite common (though Windows 2000 doesn't do this). =
</SPAN></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>For load balancing, if the =
IP=20
  addresses are different, you would end</FONT>&nbsp;<BR><FONT =
size=3D2><SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff =
face=3DArial>&nbsp;<FONT=20
  color=3D#000000 face=3D"Times New =
Roman">&gt;</FONT>&nbsp;</FONT></SPAN>up with=20
  more than one IKE phase 1, won't you?&nbsp;<SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff=20
  face=3DArial>&nbsp;</FONT></SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>Yes.=20
  </SPAN></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000>&gt;&nbsp;</SPAN>So =
I don't see=20
  why the L2TP IP</FONT>&nbsp;<FONT size=3D2>addr changing is related to =
load=20
  balancing.<SPAN class=3D498444613-24052000><FONT color=3D#0000ff=20
  face=3DArial>&nbsp;</FONT></SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN=20
  class=3D498444613-24052000>Perhaps someone who is using this or =
intends to do so=20
  could explain the utility?</SPAN></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>6. How come the draft =
doesn't require=20
  that IPSEC in transport mode be used</FONT>&nbsp;<BR><FONT =
size=3D2><SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff =
face=3DArial>&nbsp;<FONT=20
  color=3D#000000 face=3D"Times New Roman"=20
  size=3D3>&gt;</FONT>&nbsp;</FONT></SPAN>(unless i missed it)? It's =
probably=20
  worth mentioning for interoperability.</FONT>&nbsp;<FONT =
color=3D#0000ff=20
  face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>&nbsp;</SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>This=20
  is indeed required, though the words "transport mode" are not used=20
  in&nbsp;</SPAN></FONT></P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN =
class=3D498444613-24052000>the=20
  draft. I agree that it is probably worth mentioning. =
&nbsp;</SPAN></FONT></P>
  <P><FONT size=3D2><SPAN class=3D498444613-24052000><FONT =
color=3D#0000ff=20
  face=3DArial>&nbsp;&gt;&nbsp;</FONT></SPAN>7. Section 9.2.3. Why not =
disallow=20
  the responder from modifying an IP</FONT>&nbsp;<BR><FONT =
size=3D2><SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff =
face=3DArial>&nbsp;<FONT=20
  color=3D#000000 face=3D"Times New Roman"=20
  size=3D3>&gt;</FONT>&nbsp;</FONT></SPAN>address when running L2TP over =
IPSEC?=20
  How about treating this as a</FONT>&nbsp;<BR><FONT size=3D2><SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff =
face=3DArial>&nbsp;<FONT=20
  color=3D#000000 face=3D"Times New Roman"=20
  size=3D3>&gt;</FONT>&nbsp;</FONT></SPAN>misconfiguration? If we rarely =
run into=20
  such situation, it may not be worth</FONT>&nbsp;<BR><FONT =
size=3D2><SPAN=20
  class=3D498444613-24052000><FONT color=3D#0000ff =
face=3DArial>&nbsp;<FONT=20
  color=3D#000000 face=3D"Times New Roman"=20
  size=3D3>&gt;</FONT>&nbsp;</FONT></SPAN>supporting some IP addr change =

  detection, dynamically manipulating the IPSEC</FONT>&nbsp;<BR><FONT=20
  size=3D2><SPAN class=3D498444613-24052000><FONT color=3D#0000ff=20
  face=3DArial>&nbsp;<FONT color=3D#000000 face=3D"Times New Roman"=20
  size=3D3>&gt;</FONT>&nbsp;</FONT></SPAN>filters, etc.</FONT> </P>
  <P><FONT color=3D#0000ff face=3DArial size=3D2><SPAN=20
  class=3D498444613-24052000>&nbsp;I'll leave it to those who feel they =
need this=20
  feature to respond. =
&nbsp;</SPAN></FONT></P></BLOCKQUOTE></BODY></HTML>

------=_NextPart_000_006F_01BFC54E.AB5C7F30--



From owner-l2tp@ipsec.org  Thu May 25 05:05:06 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id FAA02198
	for <l2tpext-archive@odin.ietf.org>; Thu, 25 May 2000 05:05:05 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id VAA14360; Wed, 24 May 2000 21:53:48 -0700 (PDT)
Received: from p-voyageur.issy.cnet.fr (p-voyageur.issy.cnet.fr [139.100.0.39]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id VAA14356 for <l2tp@ipsec.org>; Wed, 24 May 2000 21:53:32 -0700 (PDT)
Received: by p-voyageur.issy.cnet.fr with Internet Mail Service (5.5.2448.0)
	id <KV3FNAP4>; Thu, 25 May 2000 09:16:48 +0200
Message-ID: <91A311FF6A85D3118DDF0060080C3D820E312C@lat3721.lannion.cnet.fr>
From: HOULLIER Francis FTRD/DMI/CAE <francis.houllier@rd.francetelecom.fr>
To: "'l2tp@ipsec.org'" <l2tp@ipsec.org>
Subject: Question about incoming calls.
Date: Thu, 25 May 2000 09:16:04 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2448.0)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

On a LNS, I receive a first ICRQ with an assigned session ID of 1. My LNS
sends an ICRP with an assigned session ID of 1 too. The session works and
stay open.
Then, I receive another ICRQ with an assigned session ID of 1. What must I
do ?
For me, this ICRQ is not acceptable and I send a CDN. Am I right ?

Another case, the LNS receiving that ICRQ with the assigned session ID of 1,
sends an ICRP with an assigned session ID of 2.What happens ?

Thank you for your help.


Francis HOULLIER
FTR&D / DMI / SIR
42 rue des Coutures BP 6243 
14066 CAEN CEDEX

E-mail : francis.houllier@francetelecom.fr
Tel : 02 31 75 91 65
Fax : 02 31 73 56 26



From owner-l2tp@ipsec.org  Thu May 25 09:29:07 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA07714
	for <l2tpext-archive@odin.ietf.org>; Thu, 25 May 2000 09:29:06 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id CAA14584; Thu, 25 May 2000 02:36:31 -0700 (PDT)
Received: from relay1.marconi.it (relay1.marconi.it [192.106.52.1]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id CAA14580 for <l2tp@ipsec.org>; Thu, 25 May 2000 02:35:48 -0700 (PDT)
Received: from marconicomms.com (localhost [127.0.0.1])
	by relay1.marconi.it (8.9.1a/8.9.1) with SMTP id NAA13420;
	Thu, 25 May 2000 13:54:29 +0200 (MET DST)
Received: by marconicomms.com(Lotus SMTP MTA v4.6.3  (733.2 10-16-1998))  id C12568EA.0041BC8C ; Thu, 25 May 2000 13:58:01 +0200
X-Lotus-FromDomain: MCMAIN@MCEXT
From: "Barbara Nichel" <Barbara.Nichel@Marconicomms.com>
To: "HOULLIER Francis FTRD/DMI/CAE" <francis.houllier@rd.francetelecom.fr>,
        l2tp@ipsec.org
Message-ID: <C12568EA.0041BB5A.00@marconicomms.com>
Date: Thu, 25 May 2000 13:58:18 +0200
Subject: Re: Question about incoming calls.
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline
Sender: owner-l2tp@ipsec.org
Precedence: bulk



Hi Francis,

for me, if LAC sends an ICRQ with Assigned Session ID = 1(Session ID in message
header
MUST be 0), LNS can respond with an ICRP containing the same Assigned Session ID
of 1,
because this identifier exchanged by AVP during session setup has local
significance only.
The Assigned Session ID within an ICRQ and a ICRP message is used to inform the
peer (the receiver) on the identifier used by the sender to identify the
session.
The same is for Assigned Tunnel ID AVP.
I think you must avoid to assign the same Session ID to differents sessions on
the same tunnel,
but you can use the same identifier for a session on both the ends of the tunnel
and so use
the same value for the Assigned Session AVP in ICRQ and ICRP.


Regards,
          Barbara









"HOULLIER Francis FTRD/DMI/CAE" <francis.houllier@rd.francetelecom.fr> on
05/25/2000 09:16:04 AM
                                                                                
                                                                                
                                                                                


                                                              
                                                              
                                                              
 To:      "'l2tp@ipsec.org'" <l2tp@ipsec.org>                 
                                                              
 cc:      (bcc: Barbara Nichel/MAIN/MC1)                      
                                                              
                                                              
                                                              
 Subject: Question about incoming calls.                      
                                                              








On a LNS, I receive a first ICRQ with an assigned session ID of 1. My LNS
sends an ICRP with an assigned session ID of 1 too. The session works and
stay open.
Then, I receive another ICRQ with an assigned session ID of 1. What must I
do ?
For me, this ICRQ is not acceptable and I send a CDN. Am I right ?

Another case, the LNS receiving that ICRQ with the assigned session ID of 1,
sends an ICRP with an assigned session ID of 2.What happens ?

Thank you for your help.


Francis HOULLIER
FTR&D / DMI / SIR
42 rue des Coutures BP 6243
14066 CAEN CEDEX

E-mail : francis.houllier@francetelecom.fr
Tel : 02 31 75 91 65
Fax : 02 31 73 56 26






From owner-l2tp@ipsec.org  Thu May 25 09:56:32 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id JAA08401
	for <l2tpext-archive@odin.ietf.org>; Thu, 25 May 2000 09:56:31 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id DAA14629; Thu, 25 May 2000 03:13:50 -0700 (PDT)
Received: from p-voyageur.issy.cnet.fr (p-voyageur.issy.cnet.fr [139.100.0.39]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id DAA14625 for <l2tp@ipsec.org>; Thu, 25 May 2000 03:13:32 -0700 (PDT)
Received: by p-voyageur.issy.cnet.fr with Internet Mail Service (5.5.2448.0)
	id <KV3FNGFD>; Thu, 25 May 2000 14:36:30 +0200
Message-ID: <91A311FF6A85D3118DDF0060080C3D820E3132@lat3721.lannion.cnet.fr>
From: HOULLIER Francis FTRD/DMI/CAE <francis.houllier@rd.francetelecom.fr>
To: "'Barbara Nichel'" <Barbara.Nichel@Marconicomms.com>,
        HOULLIER Francis FTRD/DMI/CAE <francis.houllier@rd.francetelecom.fr>,
        l2tp@ipsec.org
Subject: RE: Question about incoming calls, number 2.
Date: Thu, 25 May 2000 14:35:49 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2448.0)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

In fact, the real question is :
When a LNS receives an ICRP containing an assigned session Id  which is
already in use in a working session (in the same tunnel coming from the same
LAC), what MUST (or SHOULD, or MAY) this LNS do ?

Francis HOULLIER
FTR&D / DMI / SIR
42 rue des Coutures BP 6243 
14066 CAEN CEDEX

E-mail : francis.houllier@francetelecom.fr
Tel : 02 31 75 91 65
Fax : 02 31 73 56 26


From owner-l2tp@ipsec.org  Thu May 25 12:43:08 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA13403
	for <l2tpext-archive@odin.ietf.org>; Thu, 25 May 2000 12:43:07 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id FAA16112; Thu, 25 May 2000 05:51:43 -0700 (PDT)
Received: from showboat.teradyne.com ([198.51.251.10]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id FAA16108 for <l2tp@ipsec.org>; Thu, 25 May 2000 05:51:18 -0700 (PDT)
Received: from chorus.teradyne.com (chorus.teradyne.com [131.101.1.195])
	by showboat.teradyne.com (8.8.8+Sun/8.8.8) with ESMTP id LAA24489;
	Thu, 25 May 2000 11:13:34 -0400 (EDT)
Received: from notes.teradyne.com (jaypeak.corp.teradyne.com [131.101.17.23]) by chorus.teradyne.com (8.8.8+Sun/8.7.1) with SMTP id LAA06135; Thu, 25 May 2000 11:14:40 -0400 (EDT)
Received: by notes.teradyne.com(Lotus SMTP MTA v4.6.3  (733.2 10-16-1998))  id 852568EA.0053F0DC ; Thu, 25 May 2000 11:16:51 -0400
X-Lotus-FromDomain: TERADYNE
From: "David Wong" <David_Wong@notes.teradyne.com>
To: "HOULLIER Francis FTRD/DMI/CAE" <francis.houllier@rd.francetelecom.fr>
cc: "'Barbara Nichel'" <Barbara.Nichel@Marconicomms.com>, l2tp@ipsec.org
Message-ID: <852568EA.0053BC35.00@notes.teradyne.com>
Date: Thu, 25 May 2000 11:12:18 -0400
Subject: RE: Question about incoming calls, number 2.
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline
Sender: owner-l2tp@ipsec.org
Precedence: bulk

Hi, Francis. Did you mean an ICRQ instead of an ICRP?

On page 62 of the RFC (section 7.4.2), there is a state transition table. The
last line of the table applies to the established state. If an ICRQ, ICRP, or
ICCN is received by the LNS, then it must send a CDN and clean up and go back to
the idle state.





"HOULLIER Francis FTRD/DMI/CAE" <francis.houllier@rd.francetelecom.fr> on
05/25/2000 08:35:49 AM

To:   "'Barbara Nichel'" <Barbara.Nichel@Marconicomms.com>, "HOULLIER Francis
      FTRD/DMI/CAE" <francis.houllier@rd.francetelecom.fr>, l2tp@ipsec.org
cc:    (bcc: David Wong/Bos/Teradyne)

Subject:  RE: Question about incoming calls, number 2.




In fact, the real question is :
When a LNS receives an ICRP containing an assigned session Id  which is
already in use in a working session (in the same tunnel coming from the same
LAC), what MUST (or SHOULD, or MAY) this LNS do ?

Francis HOULLIER
FTR&D / DMI / SIR
42 rue des Coutures BP 6243
14066 CAEN CEDEX

E-mail : francis.houllier@francetelecom.fr
Tel : 02 31 75 91 65
Fax : 02 31 73 56 26






From owner-l2tp@ipsec.org  Thu May 25 23:00:30 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA26880
	for <l2tpext-archive@odin.ietf.org>; Thu, 25 May 2000 23:00:30 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id PAA17117; Thu, 25 May 2000 15:47:26 -0700 (PDT)
Received: from baynet.baynetworks.com (ns1.BayNetworks.COM [134.177.3.20]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id PAA17113 for <l2tp@ipsec.org>; Thu, 25 May 2000 15:44:47 -0700 (PDT)
Received: from mailhost.BayNetworks.COM (h016b.s86b1.BayNetworks.COM [134.177.1.107])
	by baynet.baynetworks.com (8.9.1/8.9.1) with ESMTP id SAA12463;
	Thu, 25 May 2000 18:06:33 -0700 (PDT)
Received: from shasta-exch.shastanets.com (mailserver.shastanets.com [47.82.16.150])
	by mailhost.BayNetworks.COM (8.9.1/8.8.8) with ESMTP id SAA08610;
	Thu, 25 May 2000 18:06:31 -0700 (PDT)
Received: by mailserver.shastanets.com with Internet Mail Service (5.5.2650.21)
	id <LBZ4J01J>; Thu, 25 May 2000 18:07:53 -0700
Message-ID: <4DEC824F6F1FD3119BEF0000E86CEA0101E49772@mailserver.shastanets.com>
From: Ly Loi <LLL@shastanets.com>
To: "'aboba@internaut.com'" <aboba@internaut.com>,
        Ly Loi
	 <LLL@shastanets.com>, l2tp@ipsec.org
Cc: gwz@cisco.com, ebooth@cisco.com, wdixon@MICROSOFT.com, townsley@cisco.com
Subject: RE: Comments/questions to l2tp over ipsec draft
Date: Thu, 25 May 2000 18:07:52 -0700
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01BFC6AE.D1956020"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

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_01BFC6AE.D1956020
Content-Type: text/plain;
	charset="iso-8859-1"

Bernard, thanks for the reply.
Actually, ipsec in transport mode would apply only to voluntary tunneling.
For compulsory tunneling, section 4.1 of RFC 2401 mandates that ipsec in
tunnel mode be used since the L2TP pkts will be considered 'transiting
traffic'. 
 
A note aside, compulsory tunneling would be another reason why L2TP and
IPSEC shouldn't be tightly coupled. In compulsory tunneling, the L2TP
initiator isn't on the same box as the IPSEC initiator.
 
Section 7.1 of the l2tp over ipsec draft states that one MUST use IPSEC ESP
for securing L2TP control pkts and SHOULD use it for L2TP data pkts. IPSEC
policies cannot distinguish L2TP control pkts from L2TP data pkts, therefore
the negotiated IPSEC policies are shared by both control and data pkts. So
how can we have a MUST for the control pkts but not a MUST for the data
pkts? Why even suggest a distinction between control pkts and data pkts? 
 
Thanks,
 
- Ly

 -----Original Message-----
From: Bernard Aboba [mailto:aboba@internaut.com]
Sent: Wednesday, May 24, 2000 7:07 AM
To: lll@shastanets.com; l2tp@ipsec.org
Cc: gwz@cisco.com; ebooth@cisco.com; wdixon@microsoft.com;
townsley@cisco.com
Subject: RE: Comments/questions to l2tp over ipsec draft



 > 1. I think sometimes L2TP is unnecessarily tightly coupled with IPSEC.  

We discussed this during the L2TPEXT WG meeting in Adelaide and

I believe we came to the same conclusion.  

 > If the L2TP tunnel goes down, there wouldn't be anymore L2TP traffic on
that 
 > IPSEC SA. IPSEC will detect such idleness through some idle timer and
bring 
 > down that SA or its SA lifetime will eventually expire (the idle timer is
of 
 > course more efficient). 

The concern was about leaving lots of zombie IPSEC SAs on the LNS. 

I think that the idle timer idea handles this though. 

 > In other words, we don't need L2TP and IPSEC to communicate with one
another 
 > when either one goes down.  

I agree that this is not a requirement. Should we change

the language in the draft to MAY?

 > Can't the initiator have "src Any-Port" in its outbound-1 filter?  

This would mean that the client would *always* speak IPSEC

to the LNS, on any port. It is simpler, as you state. Can anyone

think of a scenario in which this would cause problems? 

 > 2. Sections 9.2.x. Could you rename the word "filter" to "security
policy" 
 > or "IPSEC policy"?  

Yes. This language would be clearer. 

 > 3. Definition of "Voluntary Tunneling". 
 > Why is the "gateway to gateway scenario" considered voluntary tunneling? 
 > Shouldn't it be compulsory tunneling?  

I think it should be considered compulsory tunneling, yes.  

 > 4. Does L2TP really allow a responder to select any IP addr?  

Yes. 

> Do you know of any vendors doing it?  

Not at the moment.  

 > Also, what would their reasons be?  

To support load balancing between LNSes?

 > I guess not using 1701 as the port is also unusual, isn't it? 

There are vendors who float the LNS port, though I think it is

fair to say they are in the minority. Floating the client source

port is quite common (though Windows 2000 doesn't do this). 

 > For load balancing, if the IP addresses are different, you would end 
 > up with more than one IKE phase 1, won't you?  

Yes. 

> So I don't see why the L2TP IP addr changing is related to load balancing.


Perhaps someone who is using this or intends to do so could explain the
utility?

 > 6. How come the draft doesn't require that IPSEC in transport mode be
used 
 > (unless i missed it)? It's probably worth mentioning for
interoperability.  

This is indeed required, though the words "transport mode" are not used in 

the draft. I agree that it is probably worth mentioning.  

 > 7. Section 9.2.3. Why not disallow the responder from modifying an IP 
 > address when running L2TP over IPSEC? How about treating this as a 
 > misconfiguration? If we rarely run into such situation, it may not be
worth 
 > supporting some IP addr change detection, dynamically manipulating the
IPSEC 
 > filters, etc. 

 I'll leave it to those who feel they need this feature to respond.  


------_=_NextPart_001_01BFC6AE.D1956020
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">
<TITLE>FW: Comments/questions to l2tp over ipsec draft</TITLE>

<META content="MSHTML 5.00.2919.6307" name=GENERATOR></HEAD>
<BODY>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=682120900-26052000>Bernard, thanks for the reply.</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=682120900-26052000>Actually, ipsec in transport mode would apply only to 
voluntary tunneling. For compulsory tunneling, section 4.1 of RFC 2401 mandates 
that ipsec in tunnel mode be used since the L2TP pkts will be considered 
'transiting traffic'. </SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=682120900-26052000></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=682120900-26052000>A note 
aside, compulsory tunneling would be another reason why L2TP and IPSEC shouldn't 
be tightly coupled. In compulsory tunneling, the L2TP initiator isn't on the 
same box as the IPSEC initiator.</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=682120900-26052000></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=682120900-26052000>Section 7.1 of the l2tp over ipsec draft states that 
one MUST use IPSEC ESP for securing L2TP control pkts and SHOULD use it for L2TP 
data pkts. IPSEC policies cannot distinguish L2TP control pkts from L2TP data 
pkts, therefore the negotiated IPSEC policies are shared by both control and 
data pkts. So how can we have a MUST for the control pkts but not a MUST for the 
data pkts? Why even suggest a distinction between control pkts and data pkts? 
</SPAN></FONT></DIV>
<DIV>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=682120900-26052000>Thanks,</SPAN></FONT></DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN 
class=682120900-26052000></SPAN></FONT>&nbsp;</DIV>
<DIV><FONT color=#0000ff face=Arial size=2><SPAN class=682120900-26052000>- 
Ly</SPAN></FONT></DIV>
<DIV><SPAN class=682120900-26052000></SPAN><FONT face=Tahoma><BR><FONT 
size=2><SPAN class=682120900-26052000><FONT color=#0000ff 
face=Arial>&nbsp;</FONT></SPAN>-----Original Message-----<BR><B>From:</B> 
Bernard Aboba [mailto:aboba@internaut.com]<BR><B>Sent:</B> Wednesday, May 24, 
2000 7:07 AM<BR><B>To:</B> lll@shastanets.com; l2tp@ipsec.org<BR><B>Cc:</B> 
gwz@cisco.com; ebooth@cisco.com; wdixon@microsoft.com; 
townsley@cisco.com<BR><B>Subject:</B> RE: Comments/questions to l2tp over ipsec 
draft<BR><BR></FONT></DIV>
<BLOCKQUOTE style="MARGIN-RIGHT: 0px"></FONT>
  <BLOCKQUOTE style="MARGIN-RIGHT: 0px">
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>1. I think sometimes L2TP is 
    unnecessarily tightly coupled with IPSEC.&nbsp;<FONT color=#0000ff 
    face=Arial><SPAN class=498444613-24052000>&nbsp;</SPAN></FONT></FONT></P>
    <P><FONT size=2><FONT color=#0000ff face=Arial><SPAN 
    class=498444613-24052000>We discussed this during the L2TPEXT WG meeting in 
    Adelaide and</SPAN></FONT></FONT></P>
    <P><FONT size=2><FONT color=#0000ff face=Arial><SPAN 
    class=498444613-24052000>I believe we came to the same 
    conclusion.&nbsp;</SPAN></FONT></FONT><FONT size=2><FONT color=#0000ff 
    face=Arial><SPAN class=498444613-24052000>&nbsp;</SPAN></FONT></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>If the L2TP tunnel goes down, there 
    wouldn't be anymore L2TP traffic on that</FONT>&nbsp;<BR><FONT size=2><SPAN 
    class=498444613-24052000><FONT color=#0000ff face=Arial>&nbsp;<FONT 
    color=#000000 face="Times New Roman" 
    size=3>&gt;</FONT>&nbsp;</FONT></SPAN>IPSEC SA. IPSEC will detect such 
    idleness through some idle timer and bring</FONT>&nbsp;<BR><FONT 
    size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;<FONT color=#000000 face="Times New Roman" 
    size=3>&gt;</FONT>&nbsp;</FONT></SPAN>down that SA or its SA lifetime will 
    eventually expire (the idle timer is of</FONT>&nbsp;<BR><FONT size=2><SPAN 
    class=498444613-24052000><FONT color=#0000ff face=Arial>&nbsp;<FONT 
    color=#000000 face="Times New Roman">&gt;</FONT>&nbsp;</FONT></SPAN>course 
    more efficient).<SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;</FONT></SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>The 
    concern was about leaving lots of zombie IPSEC SAs on the LNS. 
    </SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>I 
    think that the idle timer idea handles this though. </SPAN></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>In other words, we don't need L2TP 
    and IPSEC to communicate with one another</FONT>&nbsp;<BR><FONT size=2><SPAN 
    class=498444613-24052000><FONT color=#0000ff face=Arial>&nbsp;<FONT 
    color=#000000 face="Times New Roman">&gt;</FONT>&nbsp;</FONT></SPAN>when 
    either one goes down.&nbsp;<SPAN class=498444613-24052000><FONT 
    color=#0000ff face=Arial>&nbsp;</FONT></SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>I 
    agree that this is not a requirement. Should we change</SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>the 
    language in the draft to MAY?</SPAN></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>Can't the initiator have 
    "src</FONT> <FONT size=2>Any-Port" in its outbound-1 filter?&nbsp;<SPAN 
    class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;</FONT></SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>This 
    would mean that the client would *always* speak IPSEC</SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>to 
    the LNS, on any port. It is simpler, as you state. Can 
    anyone</SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN 
    class=498444613-24052000>think of a scenario in which this would cause 
    problems? </SPAN></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>2. Sections 9.2.x. Could you rename 
    the word "filter" to "security policy"</FONT>&nbsp;<BR><FONT size=2><SPAN 
    class=498444613-24052000><FONT color=#0000ff face=Arial>&nbsp;<FONT 
    color=#000000 face="Times New Roman" 
    size=3>&gt;</FONT>&nbsp;</FONT></SPAN>or "IPSEC policy"?&nbsp;<FONT 
    color=#0000ff face=Arial><SPAN 
    class=498444613-24052000>&nbsp;</SPAN></FONT></FONT></P>
    <P><FONT size=2><FONT color=#0000ff face=Arial><SPAN 
    class=498444613-24052000>Yes. This language would be clearer. 
    </SPAN></FONT></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>3. Definition of "Voluntary 
    Tunneling".</FONT>&nbsp;<BR><FONT size=2><SPAN 
    class=498444613-24052000><FONT color=#0000ff face=Arial>&nbsp;<FONT 
    color=#000000 face="Times New Roman" 
    size=3>&gt;</FONT>&nbsp;</FONT></SPAN>Why is the "gateway to gateway 
    scenario" considered voluntary tunneling?</FONT>&nbsp;<BR><FONT size=2><SPAN 
    class=498444613-24052000><FONT color=#0000ff face=Arial>&nbsp;<FONT 
    color=#000000 face="Times New Roman" 
    size=3>&gt;</FONT>&nbsp;</FONT></SPAN>Shouldn't it be compulsory 
    tunneling?</FONT>&nbsp;<FONT color=#0000ff face=Arial size=2><SPAN 
    class=498444613-24052000>&nbsp;</SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>I 
    think it should be considered compulsory tunneling, yes. 
    &nbsp;</SPAN></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>4. Does L2TP really allow a 
    responder to select any IP addr?&nbsp;<SPAN class=498444613-24052000><FONT 
    color=#0000ff face=Arial>&nbsp;</FONT></SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>Yes. 
    </SPAN></FONT><FONT size=2></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&gt;&nbsp;Do&nbsp;</FONT></SPAN>you know of any vendors</FONT> 
    <FONT size=2>doing it?&nbsp;<SPAN class=498444613-24052000><FONT 
    color=#0000ff face=Arial>&nbsp;</FONT></SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>Not 
    at the moment. &nbsp;</SPAN></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>Also, what would their reasons 
    be?&nbsp;<SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;</FONT></SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>To 
    support&nbsp;load balancing between LNSes?</SPAN></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>I guess not using 1701 as 
    the</FONT> <FONT size=2>port is also unusual, isn't it?<SPAN 
    class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;</FONT></SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN 
    class=498444613-24052000>There are vendors who float the LNS port, though I 
    think it is</SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>fair 
    to say they are in the minority. Floating the client 
source</SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>port 
    is quite common (though Windows 2000 doesn't do this). </SPAN></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>For load balancing, if the IP 
    addresses are different, you would end</FONT>&nbsp;<BR><FONT size=2><SPAN 
    class=498444613-24052000><FONT color=#0000ff face=Arial>&nbsp;<FONT 
    color=#000000 face="Times New Roman">&gt;</FONT>&nbsp;</FONT></SPAN>up with 
    more than one IKE phase 1, won't you?&nbsp;<SPAN 
    class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;</FONT></SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>Yes. 
    </SPAN></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000>&gt;&nbsp;</SPAN>So I don't 
    see why the L2TP IP</FONT>&nbsp;<FONT size=2>addr changing is related to 
    load balancing.<SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;</FONT></SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN 
    class=498444613-24052000>Perhaps someone who is using this or intends to do 
    so could explain the utility?</SPAN></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>6. How come the draft doesn't 
    require that IPSEC in transport mode be used</FONT>&nbsp;<BR><FONT 
    size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;<FONT color=#000000 face="Times New Roman" 
    size=3>&gt;</FONT>&nbsp;</FONT></SPAN>(unless i missed it)? It's probably 
    worth mentioning for interoperability.</FONT>&nbsp;<FONT color=#0000ff 
    face=Arial size=2><SPAN class=498444613-24052000>&nbsp;</SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>This 
    is indeed required, though the words "transport mode" are not used 
    in&nbsp;</SPAN></FONT></P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN class=498444613-24052000>the 
    draft. I agree that it is probably worth mentioning. 
&nbsp;</SPAN></FONT></P>
    <P><FONT size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;&gt;&nbsp;</FONT></SPAN>7. Section 9.2.3. Why not disallow 
    the responder from modifying an IP</FONT>&nbsp;<BR><FONT size=2><SPAN 
    class=498444613-24052000><FONT color=#0000ff face=Arial>&nbsp;<FONT 
    color=#000000 face="Times New Roman" 
    size=3>&gt;</FONT>&nbsp;</FONT></SPAN>address when running L2TP over IPSEC? 
    How about treating this as a</FONT>&nbsp;<BR><FONT size=2><SPAN 
    class=498444613-24052000><FONT color=#0000ff face=Arial>&nbsp;<FONT 
    color=#000000 face="Times New Roman" 
    size=3>&gt;</FONT>&nbsp;</FONT></SPAN>misconfiguration? If we rarely run 
    into such situation, it may not be worth</FONT>&nbsp;<BR><FONT size=2><SPAN 
    class=498444613-24052000><FONT color=#0000ff face=Arial>&nbsp;<FONT 
    color=#000000 face="Times New Roman" 
    size=3>&gt;</FONT>&nbsp;</FONT></SPAN>supporting some IP addr change 
    detection, dynamically manipulating the IPSEC</FONT>&nbsp;<BR><FONT 
    size=2><SPAN class=498444613-24052000><FONT color=#0000ff 
    face=Arial>&nbsp;<FONT color=#000000 face="Times New Roman" 
    size=3>&gt;</FONT>&nbsp;</FONT></SPAN>filters, etc.</FONT> </P>
    <P><FONT color=#0000ff face=Arial size=2><SPAN 
    class=498444613-24052000>&nbsp;I'll leave it to those who feel they need 
    this feature to respond. 
&nbsp;</SPAN></FONT></P></BLOCKQUOTE></BLOCKQUOTE></BODY></HTML>

------_=_NextPart_001_01BFC6AE.D1956020--


From owner-l2tp@ipsec.org  Fri May 26 01:06:28 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA28691
	for <l2tpext-archive@odin.ietf.org>; Fri, 26 May 2000 01:06:27 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id SAA17279; Thu, 25 May 2000 18:22:21 -0700 (PDT)
Received: from sj-msg-core-2.cisco.com (sj-msg-core-2.cisco.com [171.69.43.88]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id SAA17275 for <l2tp@ipsec.org>; Thu, 25 May 2000 18:21:09 -0700 (PDT)
Received: from ebooth-u10.cisco.com (ebooth-u10.cisco.com [161.44.58.20])
	by sj-msg-core-2.cisco.com (8.9.3/8.9.1) with ESMTP id UAA24745;
	Thu, 25 May 2000 20:44:32 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by ebooth-u10.cisco.com (8.8.8-Cisco List Logging/CISCO.WS.1.2) with ESMTP id XAA23047; Thu, 25 May 2000 23:44:23 -0400 (EDT)
Date: Thu, 25 May 2000 23:44:23 -0400 (EDT)
From: Skip Booth <ebooth@cisco.com>
To: Ly Loi <LLL@shastanets.com>
cc: "'aboba@internaut.com'" <aboba@internaut.com>, l2tp@ipsec.org,
        gwz@cisco.com, wdixon@MICROSOFT.com, townsley@cisco.com
Subject: RE: Comments/questions to l2tp over ipsec draft
In-Reply-To: <4DEC824F6F1FD3119BEF0000E86CEA0101E49772@mailserver.shastanets.com>
Message-ID: <Pine.GSO.4.10.10005252330400.22985-100000@ebooth-u10.cisco.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-l2tp@ipsec.org
Precedence: bulk



On Thu, 25 May 2000, Ly Loi wrote:

> Bernard, thanks for the reply.
> Actually, ipsec in transport mode would apply only to voluntary tunneling.
> For compulsory tunneling, section 4.1 of RFC 2401 mandates that ipsec in
> tunnel mode be used since the L2TP pkts will be considered 'transiting
> traffic'. 

Not sure what you mean here.  L2TP packets, regardless of compulsory or
voluntary tunneling, are always locally originated/destined packets.  Thus if
IPsec resides in the same device as the L2TP stack then it makes perfect sense
to protect these packets using transport mode (and very little sense to use
tunnel mode, other than some of the NAT issues).

>  
> A note aside, compulsory tunneling would be another reason why L2TP and
> IPSEC shouldn't be tightly coupled. In compulsory tunneling, the L2TP
> initiator isn't on the same box as the IPSEC initiator.

When we are talking about L2TP security, yes it is.  The draft describes how to
protect L2TP using IPSEC.  So in the compulsory case, IPsec protects all L2TP
traffic between the LAC and the LNS.  L2TP is completely unaware of whether the
PPP client being tunneld is using IPsec to protect the IP traffic.

>  
> Section 7.1 of the l2tp over ipsec draft states that one MUST use IPSEC ESP
> for securing L2TP control pkts and SHOULD use it for L2TP data pkts. IPSEC
> policies cannot distinguish L2TP control pkts from L2TP data pkts, therefore
> the negotiated IPSEC policies are shared by both control and data pkts. So
> how can we have a MUST for the control pkts but not a MUST for the data
> pkts? Why even suggest a distinction between control pkts and data pkts?

In the compulsory tunneling case, it might be very useful to protect just the
control channel and not the data channel (especially in the event that the
client is already protecting its IP traffic using IPsec).  However, as you
mention IPsec has no mechanism for distinguishing between these traffic types.
I think these statements may have been added with the foresight of either L2TP
ADC or L2TP HC becoming a reality, and in that case, you could protect the
control channel differently than the data channel.

-Skip

>  
> Thanks,
>  
> - Ly
> 
>  -----Original Message-----
> From: Bernard Aboba [mailto:aboba@internaut.com]
> Sent: Wednesday, May 24, 2000 7:07 AM
> To: lll@shastanets.com; l2tp@ipsec.org
> Cc: gwz@cisco.com; ebooth@cisco.com; wdixon@microsoft.com;
> townsley@cisco.com
> Subject: RE: Comments/questions to l2tp over ipsec draft
> 
> 
> 
>  > 1. I think sometimes L2TP is unnecessarily tightly coupled with IPSEC.  
> 
> We discussed this during the L2TPEXT WG meeting in Adelaide and
> 
> I believe we came to the same conclusion.  
> 
>  > If the L2TP tunnel goes down, there wouldn't be anymore L2TP traffic on
> that 
>  > IPSEC SA. IPSEC will detect such idleness through some idle timer and
> bring 
>  > down that SA or its SA lifetime will eventually expire (the idle timer is
> of 
>  > course more efficient). 
> 
> The concern was about leaving lots of zombie IPSEC SAs on the LNS. 
> 
> I think that the idle timer idea handles this though. 
> 
>  > In other words, we don't need L2TP and IPSEC to communicate with one
> another 
>  > when either one goes down.  
> 
> I agree that this is not a requirement. Should we change
> 
> the language in the draft to MAY?
> 
>  > Can't the initiator have "src Any-Port" in its outbound-1 filter?  
> 
> This would mean that the client would *always* speak IPSEC
> 
> to the LNS, on any port. It is simpler, as you state. Can anyone
> 
> think of a scenario in which this would cause problems? 
> 
>  > 2. Sections 9.2.x. Could you rename the word "filter" to "security
> policy" 
>  > or "IPSEC policy"?  
> 
> Yes. This language would be clearer. 
> 
>  > 3. Definition of "Voluntary Tunneling". 
>  > Why is the "gateway to gateway scenario" considered voluntary tunneling? 
>  > Shouldn't it be compulsory tunneling?  
> 
> I think it should be considered compulsory tunneling, yes.  
> 
>  > 4. Does L2TP really allow a responder to select any IP addr?  
> 
> Yes. 
> 
> > Do you know of any vendors doing it?  
> 
> Not at the moment.  
> 
>  > Also, what would their reasons be?  
> 
> To support load balancing between LNSes?
> 
>  > I guess not using 1701 as the port is also unusual, isn't it? 
> 
> There are vendors who float the LNS port, though I think it is
> 
> fair to say they are in the minority. Floating the client source
> 
> port is quite common (though Windows 2000 doesn't do this). 
> 
>  > For load balancing, if the IP addresses are different, you would end 
>  > up with more than one IKE phase 1, won't you?  
> 
> Yes. 
> 
> > So I don't see why the L2TP IP addr changing is related to load balancing.
> 
> 
> Perhaps someone who is using this or intends to do so could explain the
> utility?
> 
>  > 6. How come the draft doesn't require that IPSEC in transport mode be
> used 
>  > (unless i missed it)? It's probably worth mentioning for
> interoperability.  
> 
> This is indeed required, though the words "transport mode" are not used in 
> 
> the draft. I agree that it is probably worth mentioning.  
> 
>  > 7. Section 9.2.3. Why not disallow the responder from modifying an IP 
>  > address when running L2TP over IPSEC? How about treating this as a 
>  > misconfiguration? If we rarely run into such situation, it may not be
> worth 
>  > supporting some IP addr change detection, dynamically manipulating the
> IPSEC 
>  > filters, etc. 
> 
>  I'll leave it to those who feel they need this feature to respond.  
> 
> 



From owner-l2tp@ipsec.org  Fri May 26 01:30:23 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id BAA01886
	for <l2tpext-archive@odin.ietf.org>; Fri, 26 May 2000 01:30:22 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id SAA17324; Thu, 25 May 2000 18:48:24 -0700 (PDT)
Received: from sj-msg-core-1.cisco.com (sj-msg-core-1.cisco.com [171.71.163.11]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id SAA17320 for <l2tp@ipsec.org>; Thu, 25 May 2000 18:47:12 -0700 (PDT)
Received: from ebooth-u10.cisco.com (ebooth-u10.cisco.com [161.44.58.20])
	by sj-msg-core-1.cisco.com (8.9.3/8.9.1) with ESMTP id VAA27893;
	Thu, 25 May 2000 21:10:35 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1]) by ebooth-u10.cisco.com (8.8.8-Cisco List Logging/CISCO.WS.1.2) with ESMTP id AAA23108; Fri, 26 May 2000 00:10:26 -0400 (EDT)
Date: Fri, 26 May 2000 00:10:26 -0400 (EDT)
From: Skip Booth <ebooth@cisco.com>
To: lll@shastanets.com, Bernard Aboba <aboba@internaut.com>
cc: l2tp@ipsec.org, gwz@cisco.com, wdixon@MICROSOFT.com,
        Mark Townsley <townsley@cisco.com>
Subject: RE: Comments/questions to l2tp over ipsec draft
In-Reply-To: <006e01bfc589$57b849f0$428939cc@ntdev.microsoft.com>
Message-ID: <Pine.GSO.4.10.10005252345440.22985-100000@ebooth-u10.cisco.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-l2tp@ipsec.org
Precedence: bulk


Responses inline.

On Wed, 24 May 2000, Bernard Aboba wrote:

> FW: Comments/questions to l2tp over ipsec draft > 1. I think sometimes L2TP
> is unnecessarily tightly coupled with IPSEC.
> 
>   We discussed this during the L2TPEXT WG meeting in Adelaide and
> 
>   I believe we came to the same conclusion.
> 
>    > If the L2TP tunnel goes down, there wouldn't be anymore L2TP traffic on
> that
>    > IPSEC SA. IPSEC will detect such idleness through some idle timer and
> bring
>    > down that SA or its SA lifetime will eventually expire (the idle timer
> is of
>    > course more efficient).
> 
>   The concern was about leaving lots of zombie IPSEC SAs on the LNS.
> 
>   I think that the idle timer idea handles this though.

The idle timer is not anything which is standardized and it should not be
assumed that an implementation has such a mechanism within IPsec.  In the remote
access space where you are terminating lots of voluntary tunnels, I believe it
is very useful to teardown the phase 2 SA after receiving notification from
either L2TP that the L2TP tunnel no longer exists.  If you don't you this
mechanism than you leave dangling SA's until you try to perform a refresh.  If
your refresh time is long than the amount of Phase 2 SA's sitting around could
conceivably become very large, especially when the traditional model for remote
access has been relatively short connect times.  In my mind, this is clearly an
area where L2TP/PPP add value to the current IPsec and IKE specifications.

> 
>    > In other words, we don't need L2TP and IPSEC to communicate with one
> another
>    > when either one goes down.
> 
>   I agree that this is not a requirement. Should we change
> 
>   the language in the draft to MAY?

I could live with a MAY, however I really prefer a SHOULD.  Is there a
convincing argument why we wouldn't want to make this a SHOULD?

> 
>    > Can't the initiator have "src Any-Port" in its outbound-1 filter?
> 
>   This would mean that the client would *always* speak IPSEC
> 
>   to the LNS, on any port. It is simpler, as you state. Can anyone
> 
>   think of a scenario in which this would cause problems?

The reason this was a specific port is to handle the case where the responder
(i.e. LNS), choose to float it's port.  In this case, a new phase 2 will start
from the LNS to the LAC and the LAC must be prepared to accept the quick mode
id's proposed.  In this case, I don't think it's wise for the LAC to have a
responder filter of ANY for it's dest port.  This should be the specific port
that was chosen by the LAC as it's source port for the L2TP Tunnel.

> 
>    > 2. Sections 9.2.x. Could you rename the word "filter" to "security
> policy"
>    > or "IPSEC policy"?
> 
>   Yes. This language would be clearer.
> 
>    > 3. Definition of "Voluntary Tunneling".
>    > Why is the "gateway to gateway scenario" considered voluntary
> tunneling?
>    > Shouldn't it be compulsory tunneling?
> 
>   I think it should be considered compulsory tunneling, yes.

I always referred to L2TP gateway to gateway tunneling as voluntary tunneling
because the PPP stack and the L2TP stack reside in the same device.  This seems
to me almost the same paradigm as the client to gateway case.  Why would you use
the term compulsory to describe this scenario.

> 
>    > 4. Does L2TP really allow a responder to select any IP addr?
> 
>   Yes.
> 
>   > Do you know of any vendors doing it?
> 
>   Not at the moment.
> 
>    > Also, what would their reasons be?
> 
>   To support load balancing between LNSes?
> 
>    > I guess not using 1701 as the port is also unusual, isn't it?
> 
>   There are vendors who float the LNS port, though I think it is
> 
>   fair to say they are in the minority. Floating the client source
> 
>   port is quite common (though Windows 2000 doesn't do this).
> 
>    > For load balancing, if the IP addresses are different, you would end
>    > up with more than one IKE phase 1, won't you?
> 
>   Yes.
> 
>   > So I don't see why the L2TP IP addr changing is related to load
> balancing.
> 
>   Perhaps someone who is using this or intends to do so could explain the
> utility?

For instance, I may want a different LNS to terminate the tunnel.  In the
STOPCCN, the LNS has the ability to propose an alternate address (in the result
and error code AVP, "try another").  This alternate address could be the address
of a less-utilized LNS.

> 
>    > 6. How come the draft doesn't require that IPSEC in transport mode be
> used
>    > (unless i missed it)? It's probably worth mentioning for
> interoperability.
> 
>   This is indeed required, though the words "transport mode" are not used in
> 
>   the draft. I agree that it is probably worth mentioning.

Yes.  We also need to add a section on what are the requirements to run
L2TP/IPsec through a NAT.  This all should make it into the next rev of the
draft.

> 
>    > 7. Section 9.2.3. Why not disallow the responder from modifying an IP
>    > address when running L2TP over IPSEC? How about treating this as a
>    > misconfiguration? If we rarely run into such situation, it may not be
> worth
>    > supporting some IP addr change detection, dynamically manipulating the
> IPSEC
>    > filters, etc.
> 
>    I'll leave it to those who feel they need this feature to respond.

It can be removed at the expense of having to solve this problem at some future
point.  I think this could be a very nice feature for L2TP, and I feel we should
build in support for this now and not later.  This is a MAY to start with, so if
you don't support it, that's fine.  But at least there is a documented mechanism
for how this should be done.  If this never gets implemented then assuming we
make it to proposed standard we can remove this before moving to draft standard.

-Skip

> 
> 



From owner-l2tp@ipsec.org  Fri May 26 04:26:16 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id EAA11304
	for <l2tpext-archive@odin.ietf.org>; Fri, 26 May 2000 04:26:15 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id VAA17625; Thu, 25 May 2000 21:37:15 -0700 (PDT)
Received: from relay1.marconi.it (relay1.marconi.it [192.106.52.1]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id VAA17621 for <l2tp@ipsec.org>; Thu, 25 May 2000 21:36:56 -0700 (PDT)
Received: from marconicomms.com (localhost [127.0.0.1])
	by relay1.marconi.it (8.9.1a/8.9.1) with SMTP id IAA23822;
	Fri, 26 May 2000 08:56:03 +0200 (MET DST)
Received: by marconicomms.com(Lotus SMTP MTA v4.6.3  (733.2 10-16-1998))  id C12568EB.002668F2 ; Fri, 26 May 2000 08:59:32 +0200
X-Lotus-FromDomain: MCMAIN@MCEXT
From: "Barbara Nichel" <Barbara.Nichel@Marconicomms.com>
To: "HOULLIER Francis FTRD/DMI/CAE" <francis.houllier@rd.francetelecom.fr>,
        l2tp@ipsec.org
Message-ID: <C12568EB.002666AD.00@marconicomms.com>
Date: Fri, 26 May 2000 08:59:50 +0200
Subject: RE: Question about incoming calls, number 2.
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline
Sender: owner-l2tp@ipsec.org
Precedence: bulk



Hi Francis,

looking the state transition table (7.4.2 LNS Incoming Call States), I think
that the case you are referring to,
could match this one:

LNS state:     IDLE
Event:         Receive ICRQ, not acceptable

=> next LNS state:  IDLE
     Action:        Send CDN, Clean up.

The result is that also the existing (and working) session with the same Session
ID offending
you will be closed.
Personally I think that such a situation is a symptom o malfunction, so I would
clean up the entire tunnel.

Regards,
          Barbara




From owner-l2tp@ipsec.org  Fri May 26 11:37:41 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA18380
	for <l2tpext-archive@odin.ietf.org>; Fri, 26 May 2000 11:37:40 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id EAA19355; Fri, 26 May 2000 04:56:26 -0700 (PDT)
Received: from databus.databus.com (databus.databus.com [198.186.154.34]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id EAA19351 for <l2tp@ipsec.org>; Fri, 26 May 2000 04:56:15 -0700 (PDT)
From: Barney Wolff <barney@databus.com>
To: l2tp@ipsec.org
Date: Fri, 26 May 2000 10:14 EDT
Subject: RE: Question about incoming calls, number 2.
Content-Type: text/plain
Message-ID: <392e880c0.689b@databus.databus.com>
Sender: owner-l2tp@ipsec.org
Precedence: bulk

This whole discussion seems to have ignored the l2tp reliable delivery
mechanism, which reasonably ensures that a mere retransmission will
not even be seen by the state machine.  Getting a packet from the
right addr/port, to the right addr/port, with the right tunnel id,
and a sequence that will be accepted, and yet is not from your real
peer - all that implies an active attacker who can both sniff and
inject packets.  But that's exactly the threat that only IPsec or
an equivalent can counter.

Barney Wolff  <barney@databus.com>


From owner-l2tp@ipsec.org  Fri May 26 12:08:09 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA19069
	for <l2tpext-archive@odin.ietf.org>; Fri, 26 May 2000 12:08:08 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id EAA19342; Fri, 26 May 2000 04:41:47 -0700 (PDT)
Received: from ans2-pub.dttus.com (ans2-pub.dttus.com [204.151.11.3]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id EAA19338 for <l2tp@ipsec.org>; Fri, 26 May 2000 04:40:37 -0700 (PDT)
Received: by ans2-pub.dttus.com id JAA27890
  (InterLock SMTP Gateway 3.0 for l2tp@ipsec.org);
  Fri, 26 May 2000 09:04:16 -0500
Received: by ans2-pub.dttus.com (Internal Mail Agent-2);
  Fri, 26 May 2000 09:04:16 -0500
Received: by ans2-pub.dttus.com (Internal Mail Agent-1);
  Fri, 26 May 2000 09:04:16 -0500
Message-Id: <2F4A3F6B7BF5D3119E2500508B94E656FC8607@uscntnt405.us.deloitte.com>
From: "Verma, Rohit (US - Chicago)" <rverma@dc.com>
To: Barbara Nichel <Barbara.Nichel@Marconicomms.com>,
        HOULLIER Francis FTRD/DMI/CAE <francis.houllier@rd.francetelecom.fr>,
        l2tp@ipsec.org
Subject: RE: Question about incoming calls, number 2.
Date: Fri, 26 May 2000 10:01:38 -0400
Mime-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: multipart/alternative;
	boundary="----_=_NextPart_001_01BFC71A.E9375D20"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

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_01BFC71A.E9375D20
Content-Type: text/plain;
	charset="iso-8859-1"

I don't think you should do a blind cleanup because you could get a packet
from an LNS that put a bad Tunnel Id in the packet and you will end up
losing all connection with the good LNS.
My suggestion will be to discard a packet that is out of state.

Regards

Rohit Verma
Deloitte Consulting
180, N. Stetson Avenue
Chicago, Illinois 60601
Phone: 312-374-2475


-----Original Message-----
From: Barbara Nichel [mailto:Barbara.Nichel@Marconicomms.com]
Sent: Friday, May 26, 2000 8:00 AM
To: HOULLIER Francis FTRD/DMI/CAE; l2tp@ipsec.org
Subject: RE: Question about incoming calls, number 2.




Hi Francis,

looking the state transition table (7.4.2 LNS Incoming Call States), I think
that the case you are referring to,
could match this one:

LNS state:     IDLE
Event:         Receive ICRQ, not acceptable

=> next LNS state:  IDLE
     Action:        Send CDN, Clean up.

The result is that also the existing (and working) session with the same
Session
ID offending
you will be closed.
Personally I think that such a situation is a symptom o malfunction, so I
would
clean up the entire tunnel.

Regards,
          Barbara


------_=_NextPart_001_01BFC71A.E9375D20
Content-Type: text/html;
	charset="iso-8859-1"
Content-Transfer-Encoding: quoted-printable

<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 3.2//EN">
<HTML>
<HEAD>
<META HTTP-EQUIV=3D"Content-Type" CONTENT=3D"text/html; =
charset=3Diso-8859-1">
<META NAME=3D"Generator" CONTENT=3D"MS Exchange Server version =
5.5.2652.35">
<TITLE>RE: Question about incoming calls, number 2.</TITLE>
</HEAD>
<BODY>

<P><FONT SIZE=3D2>I don't think you should do a blind cleanup because =
you could get a packet from an LNS that put a bad Tunnel Id in the =
packet and you will end up losing all connection with the good =
LNS.</FONT></P>

<P><FONT SIZE=3D2>My suggestion will be to discard a packet that is out =
of state.</FONT>
</P>

<P><FONT SIZE=3D2>Regards</FONT>
</P>

<P><FONT SIZE=3D2>Rohit Verma</FONT>
<BR><FONT SIZE=3D2>Deloitte Consulting</FONT>
<BR><FONT SIZE=3D2>180, N. Stetson Avenue</FONT>
<BR><FONT SIZE=3D2>Chicago, Illinois 60601</FONT>
<BR><FONT SIZE=3D2>Phone: 312-374-2475</FONT>
</P>
<BR>

<P><FONT SIZE=3D2>-----Original Message-----</FONT>
<BR><FONT SIZE=3D2>From: Barbara Nichel [<A =
HREF=3D"mailto:Barbara.Nichel@Marconicomms.com">mailto:Barbara.Nichel@Ma=
rconicomms.com</A>]</FONT>
<BR><FONT SIZE=3D2>Sent: Friday, May 26, 2000 8:00 AM</FONT>
<BR><FONT SIZE=3D2>To: HOULLIER Francis FTRD/DMI/CAE; =
l2tp@ipsec.org</FONT>
<BR><FONT SIZE=3D2>Subject: RE: Question about incoming calls, number =
2.</FONT>
</P>
<BR>
<BR>
<BR>

<P><FONT SIZE=3D2>Hi Francis,</FONT>
</P>

<P><FONT SIZE=3D2>looking the state transition table (7.4.2 LNS =
Incoming Call States), I think</FONT>
<BR><FONT SIZE=3D2>that the case you are referring to,</FONT>
<BR><FONT SIZE=3D2>could match this one:</FONT>
</P>

<P><FONT SIZE=3D2>LNS state:&nbsp;&nbsp;&nbsp;&nbsp; IDLE</FONT>
<BR><FONT =
SIZE=3D2>Event:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Receive =
ICRQ, not acceptable</FONT>
</P>

<P><FONT SIZE=3D2>=3D&gt; next LNS state:&nbsp; IDLE</FONT>
<BR><FONT SIZE=3D2>&nbsp;&nbsp;&nbsp;&nbsp; =
Action:&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; Send CDN, Clean =
up.</FONT>
</P>

<P><FONT SIZE=3D2>The result is that also the existing (and working) =
session with the same Session</FONT>
<BR><FONT SIZE=3D2>ID offending</FONT>
<BR><FONT SIZE=3D2>you will be closed.</FONT>
<BR><FONT SIZE=3D2>Personally I think that such a situation is a =
symptom o malfunction, so I would</FONT>
<BR><FONT SIZE=3D2>clean up the entire tunnel.</FONT>
</P>

<P><FONT SIZE=3D2>Regards,</FONT>
<BR><FONT =
SIZE=3D2>&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp;&nbsp; =
Barbara</FONT>
</P>

</BODY>
</HTML>
------_=_NextPart_001_01BFC71A.E9375D20--


From owner-l2tp@ipsec.org  Fri May 26 13:20:42 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA20723
	for <l2tpext-archive@odin.ietf.org>; Fri, 26 May 2000 13:20:40 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id GAA19475; Fri, 26 May 2000 06:34:11 -0700 (PDT)
Received: from p-biset.issy.cnet.fr (p-biset.issy.cnet.fr [139.100.0.33]) by ns1.sailpix.com (8.8.8/8.6.9) with SMTP id GAA19471 for <l2tp@ipsec.org>; Fri, 26 May 2000 06:33:49 -0700 (PDT)
Received: by p-biset.issy.cnet.fr with Internet Mail Service (5.5.2448.0)
	id <KV3GBPXM>; Fri, 26 May 2000 17:57:12 +0200
Message-ID: <91A311FF6A85D3118DDF0060080C3D820E3134@lat3721.lannion.cnet.fr>
From: HOULLIER Francis FTRD/DMI/CAE <francis.houllier@rd.francetelecom.fr>
To: l2tp@ipsec.org
Subject: RE: Question about incoming calls, number 2.
Date: Fri, 26 May 2000 17:56:13 +0200
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2448.0)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

Thanks all for your help.
To close the discussion, I will send a CDN when I receive that wrong ICRQ.
Close the whole tunnel does not match with the Control connection states.

Francis HOULLIER
FTR&D / DMI / SIR
42 rue des Coutures BP 6243 
14066 CAEN CEDEX

E-mail : francis.houllier@francetelecom.fr
Tel : 02 31 75 91 65
Fax : 02 31 73 56 26


From owner-l2tp@ipsec.org  Fri May 26 13:35:18 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA21035
	for <l2tpext-archive@odin.ietf.org>; Fri, 26 May 2000 13:35:18 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id GAA19438; Fri, 26 May 2000 06:18:46 -0700 (PDT)
Received: from baynet.baynetworks.com (ns1.BayNetworks.COM [134.177.3.20]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id GAA19433 for <l2tp@ipsec.org>; Fri, 26 May 2000 06:17:14 -0700 (PDT)
Received: from mailhost.BayNetworks.COM (h016b.s86b1.BayNetworks.COM [134.177.1.107])
	by baynet.baynetworks.com (8.9.1/8.9.1) with ESMTP id IAA00386;
	Fri, 26 May 2000 08:39:01 -0700 (PDT)
Received: from shasta-exch.shastanets.com (mailserver.shastanets.com [47.82.16.150])
	by mailhost.BayNetworks.COM (8.9.1/8.8.8) with ESMTP id IAA29018;
	Fri, 26 May 2000 08:38:59 -0700 (PDT)
Received: by mailserver.shastanets.com with Internet Mail Service (5.5.2650.21)
	id <LBZ4KAK5>; Fri, 26 May 2000 08:40:19 -0700
Message-ID: <4DEC824F6F1FD3119BEF0000E86CEA0101E49774@mailserver.shastanets.com>
From: Ly Loi <LLL@shastanets.com>
To: "'Skip Booth'" <ebooth@cisco.com>, Ly Loi <LLL@shastanets.com>
Cc: "'aboba@internaut.com'" <aboba@internaut.com>, l2tp@ipsec.org,
        gwz@cisco.com, wdixon@MICROSOFT.com, townsley@cisco.com
Subject: RE: Comments/questions to l2tp over ipsec draft
Date: Fri, 26 May 2000 08:40:17 -0700
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

Skip, I was referring to the case where the LAC and the ipsec tunnel
initiator didn't reside on the same device:
LAC---gateway(ipsec tunnel initiator)---------------LNS(ipsec tunnel
terminator)

Is the draft supposed to cover this case too? This case would be considered
compulsory tunneling, ipsec tunneling :-)
If L2TP and IPSEC reside on the same device, then yes, we all agree IPSEC
would run in transport mode and I trust this will be stated in a future
version of the draft, thanks.

- Ly

-----Original Message-----
From: Skip Booth [mailto:ebooth@cisco.com]
Sent: Thursday, May 25, 2000 8:44 PM
To: Ly Loi
Cc: 'aboba@internaut.com'; l2tp@ipsec.org; gwz@cisco.com;
wdixon@MICROSOFT.com; townsley@cisco.com
Subject: RE: Comments/questions to l2tp over ipsec draft




On Thu, 25 May 2000, Ly Loi wrote:

> Bernard, thanks for the reply.
> Actually, ipsec in transport mode would apply only to voluntary tunneling.
> For compulsory tunneling, section 4.1 of RFC 2401 mandates that ipsec in
> tunnel mode be used since the L2TP pkts will be considered 'transiting
> traffic'. 

Not sure what you mean here.  L2TP packets, regardless of compulsory or
voluntary tunneling, are always locally originated/destined packets.  Thus
if
IPsec resides in the same device as the L2TP stack then it makes perfect
sense
to protect these packets using transport mode (and very little sense to use
tunnel mode, other than some of the NAT issues).


>  
> A note aside, compulsory tunneling would be another reason why L2TP and
> IPSEC shouldn't be tightly coupled. In compulsory tunneling, the L2TP
> initiator isn't on the same box as the IPSEC initiator.

When we are talking about L2TP security, yes it is.  The draft describes how
to
protect L2TP using IPSEC.  So in the compulsory case, IPsec protects all
L2TP
traffic between the LAC and the LNS.  L2TP is completely unaware of whether
the
PPP client being tunneld is using IPsec to protect the IP traffic.

>  
> Section 7.1 of the l2tp over ipsec draft states that one MUST use IPSEC
ESP
> for securing L2TP control pkts and SHOULD use it for L2TP data pkts. IPSEC
> policies cannot distinguish L2TP control pkts from L2TP data pkts,
therefore
> the negotiated IPSEC policies are shared by both control and data pkts. So
> how can we have a MUST for the control pkts but not a MUST for the data
> pkts? Why even suggest a distinction between control pkts and data pkts?

In the compulsory tunneling case, it might be very useful to protect just
the
control channel and not the data channel (especially in the event that the
client is already protecting its IP traffic using IPsec).  However, as you
mention IPsec has no mechanism for distinguishing between these traffic
types.
I think these statements may have been added with the foresight of either
L2TP
ADC or L2TP HC becoming a reality, and in that case, you could protect the
control channel differently than the data channel.

-Skip

>  
> Thanks,
>  
> - Ly
> 
>  -----Original Message-----
> From: Bernard Aboba [mailto:aboba@internaut.com]
> Sent: Wednesday, May 24, 2000 7:07 AM
> To: lll@shastanets.com; l2tp@ipsec.org
> Cc: gwz@cisco.com; ebooth@cisco.com; wdixon@microsoft.com;
> townsley@cisco.com
> Subject: RE: Comments/questions to l2tp over ipsec draft
> 
> 
> 
>  > 1. I think sometimes L2TP is unnecessarily tightly coupled with IPSEC.

> 
> We discussed this during the L2TPEXT WG meeting in Adelaide and
> 
> I believe we came to the same conclusion.  
> 
>  > If the L2TP tunnel goes down, there wouldn't be anymore L2TP traffic on
> that 
>  > IPSEC SA. IPSEC will detect such idleness through some idle timer and
> bring 
>  > down that SA or its SA lifetime will eventually expire (the idle timer
is
> of 
>  > course more efficient). 
> 
> The concern was about leaving lots of zombie IPSEC SAs on the LNS. 
> 
> I think that the idle timer idea handles this though. 
> 
>  > In other words, we don't need L2TP and IPSEC to communicate with one
> another 
>  > when either one goes down.  
> 
> I agree that this is not a requirement. Should we change
> 
> the language in the draft to MAY?
> 
>  > Can't the initiator have "src Any-Port" in its outbound-1 filter?  
> 
> This would mean that the client would *always* speak IPSEC
> 
> to the LNS, on any port. It is simpler, as you state. Can anyone
> 
> think of a scenario in which this would cause problems? 
> 
>  > 2. Sections 9.2.x. Could you rename the word "filter" to "security
> policy" 
>  > or "IPSEC policy"?  
> 
> Yes. This language would be clearer. 
> 
>  > 3. Definition of "Voluntary Tunneling". 
>  > Why is the "gateway to gateway scenario" considered voluntary
tunneling? 
>  > Shouldn't it be compulsory tunneling?  
> 
> I think it should be considered compulsory tunneling, yes.  
> 
>  > 4. Does L2TP really allow a responder to select any IP addr?  
> 
> Yes. 
> 
> > Do you know of any vendors doing it?  
> 
> Not at the moment.  
> 
>  > Also, what would their reasons be?  
> 
> To support load balancing between LNSes?
> 
>  > I guess not using 1701 as the port is also unusual, isn't it? 
> 
> There are vendors who float the LNS port, though I think it is
> 
> fair to say they are in the minority. Floating the client source
> 
> port is quite common (though Windows 2000 doesn't do this). 
> 
>  > For load balancing, if the IP addresses are different, you would end 
>  > up with more than one IKE phase 1, won't you?  
> 
> Yes. 
> 
> > So I don't see why the L2TP IP addr changing is related to load
balancing.
> 
> 
> Perhaps someone who is using this or intends to do so could explain the
> utility?
> 
>  > 6. How come the draft doesn't require that IPSEC in transport mode be
> used 
>  > (unless i missed it)? It's probably worth mentioning for
> interoperability.  
> 
> This is indeed required, though the words "transport mode" are not used in

> 
> the draft. I agree that it is probably worth mentioning.  
> 
>  > 7. Section 9.2.3. Why not disallow the responder from modifying an IP 
>  > address when running L2TP over IPSEC? How about treating this as a 
>  > misconfiguration? If we rarely run into such situation, it may not be
> worth 
>  > supporting some IP addr change detection, dynamically manipulating the
> IPSEC 
>  > filters, etc. 
> 
>  I'll leave it to those who feel they need this feature to respond.  
> 
> 


From owner-l2tp@ipsec.org  Fri May 26 13:54:08 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA21462
	for <l2tpext-archive@odin.ietf.org>; Fri, 26 May 2000 13:54:08 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id HAA19632; Fri, 26 May 2000 07:31:36 -0700 (PDT)
Received: from tcb.net (tcb.net [205.168.100.1]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id HAA19628 for <l2tp@ipsec.org>; Fri, 26 May 2000 07:31:21 -0700 (PDT)
Received: from sofos.tcb.net (sofos.tcb.net [127.0.0.1])
	by tcb.net (8.9.3/8.9.3) with ESMTP id KAA18216
	for <l2tp@ipsec.org>; Fri, 26 May 2000 10:56:27 -0600
Message-Id: <200005261656.KAA18216@tcb.net>
To: "'l2tp@ipsec.org'" <l2tp@ipsec.org>
From: Danny McPherson <danny@tcb.net>
Reply-To: danny@tcb.net
Subject: Re: L2TP Circuit Emulation Services Extension 
Date: Fri, 26 May 2000 10:56:27 -0600
Sender: owner-l2tp@ipsec.org
Precedence: bulk


Folks,
I'm getting ready to post a revision to draft-mcpherson-l2tp-es-00.txt 
to the I-D directory and would like to solicit additional comments 
beforehand.

I'd also like to understand concerns of anyone with objections to 
adopting the draft as a working group item.  I've already asked the
chair to request expansion of the WG charter from the AD.

Thanks to everyone that's provided feedback so far, and thanks in 
advance to those who are planning to!

-danny

P.S.  The mailing list archives are still in need of attention ..
and the 'list subscribers' access should be disabled or restircted 
to avoid future spam incidents.


From owner-l2tp@ipsec.org  Fri May 26 16:37:47 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id QAA25133
	for <l2tpext-archive@odin.ietf.org>; Fri, 26 May 2000 16:37:46 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id KAA19955; Fri, 26 May 2000 10:17:54 -0700 (PDT)
Received: from uzura.cisco.com (uzura.cisco.com [161.44.3.77]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id KAA19951 for <l2tp@ipsec.org>; Fri, 26 May 2000 10:17:32 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1])
	by uzura.cisco.com (8.8.8-Cisco List Logging/8.8.8) with ESMTP id PAA17233;
	Fri, 26 May 2000 15:36:16 -0400 (EDT)
Date: Fri, 26 May 2000 15:36:16 -0400 (EDT)
From: Skip Booth <ebooth@cisco.com>
To: Ly Loi <LLL@shastanets.com>
cc: "'aboba@internaut.com'" <aboba@internaut.com>, l2tp@ipsec.org,
        gwz@cisco.com, wdixon@MICROSOFT.com, townsley@cisco.com
Subject: RE: Comments/questions to l2tp over ipsec draft
In-Reply-To: <4DEC824F6F1FD3119BEF0000E86CEA0101E49774@mailserver.shastanets.com>
Message-ID: <Pine.GSO.4.10.10005261530370.15705-100000@uzura.cisco.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-l2tp@ipsec.org
Precedence: bulk



On Fri, 26 May 2000, Ly Loi wrote:

> Skip, I was referring to the case where the LAC and the ipsec tunnel
> initiator didn't reside on the same device:
> LAC---gateway(ipsec tunnel initiator)---------------LNS(ipsec tunnel
> terminator)
> 
> Is the draft supposed to cover this case too? This case would be considered
> compulsory tunneling, ipsec tunneling :-)

The draft discusses security concerns in the compulsory model.  Specifically, it
talks about how the link between the client and the LAC is potentially not
secure and that clients may require end to end security.  If the client is
running end to end security than potentially no encryption is required for L2TP
data packets between the LAC and LNS, but authentication may be.  As you noted
however, currently there is no way standardized method for IPsec to
differentiate between L2TP data and control packets.

Does the draft need to go into more detail in this area?  If so what more would
you like to be discussed?

> If L2TP and IPSEC reside on the same device, then yes, we all agree IPSEC
> would run in transport mode and I trust this will be stated in a future
> version of the draft, thanks.

Yes.

-Skip

> 
> - Ly
> 

[rest of text snipped]



From owner-l2tp@ipsec.org  Fri May 26 17:38:51 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA26635
	for <l2tpext-archive@odin.ietf.org>; Fri, 26 May 2000 17:38:50 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id LAA20052; Fri, 26 May 2000 11:25:31 -0700 (PDT)
Received: from calcite.rhyolite.com (calcite.rhyolite.com [38.159.140.3]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id LAA20048 for <l2tp@ipsec.org>; Fri, 26 May 2000 11:25:14 -0700 (PDT)
Received: (from vjs@localhost)
	by calcite.rhyolite.com (8.9.3/calcite) id OAA18326
	for l2tp@ipsec.org  env-from <vjs>;
	Fri, 26 May 2000 14:49:01 -0600 (MDT)
Date: Fri, 26 May 2000 14:49:01 -0600 (MDT)
From: Vernon Schryver <vjs@calcite.rhyolite.com>
Message-Id: <200005262049.OAA18326@calcite.rhyolite.com>
To: l2tp@ipsec.org
Sender: owner-l2tp@ipsec.org
Precedence: bulk

Before dealing with him, please note that Mr. John Brusa says the 
following about his spam:

] Business is very successful. I'm sorry you feel the way you do.
] Best of luck to you.


Please also note:
    traceroute to mriexec.com (207.231.93.10) 
    11  gw7-QUIKNET-DOM.scrl.eli.net (209.210.249.214)  173.689 ms
    12  rtr10.fe1-0.core.router.quiknet.com (207.183.249.244)  171.209 ms
    13  192.168.10.46 (192.168.10.46)  177.294 ms
    14  rtr2.e0.rsvl.edge.router.quiknet.com (207.183.227.130)  193.823 ms
    15  mrigate.rsvl.dsl.quiknet.com (207.231.93.10)  216.778 ms

Over the years I've found abuse@eli.net reasonably responsive.
I don't remember quiknet.com off hand, but that and the fact that
they maintain an abuse@quiknet.com mailbox are good signs.


Vernon Schryver    vjs@rhyolite.com


From owner-l2tp@ipsec.org  Fri May 26 23:15:46 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id XAA01766
	for <l2tpext-archive@odin.ietf.org>; Fri, 26 May 2000 23:15:45 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id QAA20335; Fri, 26 May 2000 16:51:09 -0700 (PDT)
Received: from southpass.baynetworks.com (ns2.BayNetworks.COM [134.177.3.16]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id QAA20331 for <l2tp@ipsec.org>; Fri, 26 May 2000 16:50:21 -0700 (PDT)
Received: from mailhost.BayNetworks.COM (h016b.s86b1.BayNetworks.COM [134.177.1.107])
	by southpass.baynetworks.com (8.9.1/8.9.1) with ESMTP id TAA00576;
	Fri, 26 May 2000 19:06:05 -0700 (PDT)
Received: from shasta-exch.shastanets.com (mailserver.shastanets.com [47.82.16.150])
	by mailhost.BayNetworks.COM (8.9.1/8.8.8) with ESMTP id TAA11723;
	Fri, 26 May 2000 19:11:57 -0700 (PDT)
Received: by mailserver.shastanets.com with Internet Mail Service (5.5.2650.21)
	id <LWSAJH29>; Fri, 26 May 2000 19:13:19 -0700
Message-ID: <4DEC824F6F1FD3119BEF0000E86CEA0101E49778@mailserver.shastanets.com>
From: Ly Loi <LLL@shastanets.com>
To: "'Skip Booth'" <ebooth@cisco.com>, Ly Loi <LLL@shastanets.com>
Cc: "'aboba@internaut.com'" <aboba@internaut.com>, l2tp@ipsec.org,
        gwz@cisco.com, wdixon@MICROSOFT.com, townsley@cisco.com
Subject: RE: Comments/questions to l2tp over ipsec draft
Date: Fri, 26 May 2000 19:13:10 -0700
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

Skip, security between the client and the LAC isn't what I was referring to.
In my scenario, the PPP and the LAC can reside on the box, or different
boxes. However, the LAC doesn't reside on the same box as the IPSEC tunnel
initiator.
To be more specific, what I have in mind is a group of LACs sharing 1 IPSEC
SA. This IPSEC SA is between the gateway and the LNS.
                     
LAC1---------gateway------------ipsec_tunnel--------------LNS
LAC2-----------|
LAC3-----------|

Unlike the cases covered by the draft where each L2TP tunnel is mapped to an
IPSEC SA, here, multiple L2TP tunnels map to one IPSEC SA. If the draft
could also cover this case, it'd be more complete. 

A few more comments:

1. Could you not require that L2TP start the retransmission timer only after
the IPSEC SA has come up (section 9.2.2)? I stated the reason in my original
email. In addition, in the above scenario, this requirement wouldn't be
applicable.

2. The draft lists the IPSEC filters as if they represent both the IPSEC
policies and the IPSEC SAs, which is rather confusing since IPSEC SAs are
derived from IPSEC policies. If you could decouple them in your examples,
it'd be great.

3. I don't know how others feel about having an external entity (e.g. L2TP)
injecting some filters into the IPSEC policy database, but this appears to
me to be a local implementation issue. Also, this wouldn't be feasible for
the above scenario. I however appreciate the problems the draft is trying to
solve.

Thanks,

- Ly





> -----Original Message-----
> From: Skip Booth [mailto:ebooth@cisco.com]
> Sent: Friday, May 26, 2000 12:36 PM
> To: Ly Loi
> Cc: 'aboba@internaut.com'; l2tp@ipsec.org; gwz@cisco.com;
> wdixon@MICROSOFT.com; townsley@cisco.com
> Subject: RE: Comments/questions to l2tp over ipsec draft
> 
> 
> 
> 
> On Fri, 26 May 2000, Ly Loi wrote:
> 
> > Skip, I was referring to the case where the LAC and the ipsec tunnel
> > initiator didn't reside on the same device:
> > LAC---gateway(ipsec tunnel initiator)---------------LNS(ipsec tunnel
> > terminator)
> > 
> > Is the draft supposed to cover this case too? This case 
> would be considered
> > compulsory tunneling, ipsec tunneling :-)
> 
> The draft discusses security concerns in the compulsory 
> model.  Specifically, it
> talks about how the link between the client and the LAC is 
> potentially not
> secure and that clients may require end to end security.  If 
> the client is
> running end to end security than potentially no encryption is 
> required for L2TP
> data packets between the LAC and LNS, but authentication may 
> be.  As you noted
> however, currently there is no way standardized method for IPsec to
> differentiate between L2TP data and control packets.
> 
> Does the draft need to go into more detail in this area?  If 
> so what more would
> you like to be discussed?
> 
> > If L2TP and IPSEC reside on the same device, then yes, we 
> all agree IPSEC
> > would run in transport mode and I trust this will be stated 
> in a future
> > version of the draft, thanks.
> 
> Yes.
> 
> -Skip
> 
> > 
> > - Ly
> > 
> 
> [rest of text snipped]
> 


From owner-l2tp@ipsec.org  Sat May 27 11:14:38 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id LAA18926
	for <l2tpext-archive@odin.ietf.org>; Sat, 27 May 2000 11:14:37 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id EAA22288; Sat, 27 May 2000 04:36:13 -0700 (PDT)
Received: from brahma.roc.com ([202.54.67.218]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id EAA22284 for <l2tp@ipsec.org>; Sat, 27 May 2000 04:35:58 -0700 (PDT)
Received: from suren.roc.com ([172.16.5.74])
	by brahma.roc.com (8.8.5/8.8.5) with SMTP id TAA03816
	for <l2tp@ipsec.org>; Sat, 27 May 2000 19:26:50 +0530 (IST)
Message-Id: <3.0.1.32.20000527191558.007771d8@172.16.1.10>
X-Sender: suren@172.16.1.10
X-Mailer: Windows Eudora Light Version 3.0.1 (32)
Date: Sat, 27 May 2000 19:15:58 +0500
To: l2tp@ipsec.org
From: Surendranath <suren@trinc.com>
Subject: Any free software for l2tp?
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: owner-l2tp@ipsec.org
Precedence: bulk


Hi 
Is there any free software available for LAC and LNS on any platform?

thanks in advance.
surendra.


From owner-l2tp@ipsec.org  Sat May 27 15:36:16 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id PAA20757
	for <l2tpext-archive@odin.ietf.org>; Sat, 27 May 2000 15:36:15 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id JAA22508; Sat, 27 May 2000 09:12:52 -0700 (PDT)
Received: from franklin.cisco.com (franklin.cisco.com [171.70.156.17]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id JAA22504 for <l2tp@ipsec.org>; Sat, 27 May 2000 09:11:57 -0700 (PDT)
Received: from gwz (atlantis-dial-1-89.cisco.com [171.68.181.90]) by franklin.cisco.com (8.8.6 (PHNE_17190)/CISCO.SERVER.1.2) with SMTP id LAA05003; Sat, 27 May 2000 11:32:48 -0700 (PDT)
Reply-To: <gwz@cisco.com>
From: "Glen Zorn" <gwz@cisco.com>
To: "Ly Loi" <LLL@shastanets.com>, "'Skip Booth'" <ebooth@cisco.com>
Cc: <aboba@internaut.com>, <l2tp@ipsec.org>, <wdixon@MICROSOFT.com>,
        <townsley@cisco.com>, "Glen Zorn" <gwz@cisco.com>
Subject: RE: Comments/questions to l2tp over ipsec draft
Date: Sat, 27 May 2000 11:16:29 -0700
Message-ID: <NDBBIHMPILAAGDHPCIOPKEKGCFAA.gwz@cisco.com>
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 (9.0.2910.0)
Importance: Normal
In-Reply-To: <4DEC824F6F1FD3119BEF0000E86CEA0101E49778@mailserver.shastanets.com>
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit

Ly Loi [mailto://LLL@shastanets.com] writes:

> Skip, security between the client and the LAC isn't what I was
> referring to.
> In my scenario, the PPP and the LAC can reside on the box, or different
> boxes. However, the LAC doesn't reside on the same box as the IPSEC tunnel
> initiator.
> To be more specific, what I have in mind is a group of LACs
> sharing 1 IPSEC
> SA. This IPSEC SA is between the gateway and the LNS.
>
> LAC1---------gateway------------ipsec_tunnel--------------LNS
> LAC2-----------|
> LAC3-----------|

It appears that you are missing a "security gateway" at the LNS end.

>
> Unlike the cases covered by the draft where each L2TP tunnel is
> mapped to an
> IPSEC SA, here, multiple L2TP tunnels map to one IPSEC SA. If the draft
> could also cover this case, it'd be more complete.

I don't think that the point of the draft is to present an exhaustive list
of every method (no  matter how lame) that can be used to "secure" L2TP
tunnels.

>
> A few more comments:
>
> 1. Could you not require that L2TP start the retransmission timer
> only after
> the IPSEC SA has come up (section 9.2.2)? I stated the reason in
> my original
> email. In addition, in the above scenario, this requirement wouldn't be
> applicable.
>
> 2. The draft lists the IPSEC filters as if they represent both the IPSEC
> policies and the IPSEC SAs, which is rather confusing since IPSEC SAs are
> derived from IPSEC policies. If you could decouple them in your examples,
> it'd be great.
>
> 3. I don't know how others feel about having an external entity
> (e.g. L2TP)
> injecting some filters into the IPSEC policy database, but this appears to
> me to be a local implementation issue. Also, this wouldn't be feasible for
> the above scenario. I however appreciate the problems the draft
> is trying to
> solve.
>
> Thanks,
>
> - Ly
>
>
>
>
>
> > -----Original Message-----
> > From: Skip Booth [mailto:ebooth@cisco.com]
> > Sent: Friday, May 26, 2000 12:36 PM
> > To: Ly Loi
> > Cc: 'aboba@internaut.com'; l2tp@ipsec.org; gwz@cisco.com;
> > wdixon@MICROSOFT.com; townsley@cisco.com
> > Subject: RE: Comments/questions to l2tp over ipsec draft
> >
> >
> >
> >
> > On Fri, 26 May 2000, Ly Loi wrote:
> >
> > > Skip, I was referring to the case where the LAC and the ipsec tunnel
> > > initiator didn't reside on the same device:
> > > LAC---gateway(ipsec tunnel initiator)---------------LNS(ipsec tunnel
> > > terminator)
> > >
> > > Is the draft supposed to cover this case too? This case
> > would be considered
> > > compulsory tunneling, ipsec tunneling :-)
> >
> > The draft discusses security concerns in the compulsory
> > model.  Specifically, it
> > talks about how the link between the client and the LAC is
> > potentially not
> > secure and that clients may require end to end security.  If
> > the client is
> > running end to end security than potentially no encryption is
> > required for L2TP
> > data packets between the LAC and LNS, but authentication may
> > be.  As you noted
> > however, currently there is no way standardized method for IPsec to
> > differentiate between L2TP data and control packets.
> >
> > Does the draft need to go into more detail in this area?  If
> > so what more would
> > you like to be discussed?
> >
> > > If L2TP and IPSEC reside on the same device, then yes, we
> > all agree IPSEC
> > > would run in transport mode and I trust this will be stated
> > in a future
> > > version of the draft, thanks.
> >
> > Yes.
> >
> > -Skip
> >
> > >
> > > - Ly
> > >
> >
> > [rest of text snipped]
> >
>
>



From owner-l2tp@ipsec.org  Sat May 27 17:29:59 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA21199
	for <l2tpext-archive@odin.ietf.org>; Sat, 27 May 2000 17:29:58 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id LAA22621; Sat, 27 May 2000 11:12:02 -0700 (PDT)
Received: from uzura.cisco.com (uzura.cisco.com [161.44.3.77]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id LAA22617 for <l2tp@ipsec.org>; Sat, 27 May 2000 11:10:53 -0700 (PDT)
Received: from localhost (localhost [127.0.0.1])
	by uzura.cisco.com (8.8.8-Cisco List Logging/8.8.8) with ESMTP id QAA01282;
	Sat, 27 May 2000 16:29:38 -0400 (EDT)
Date: Sat, 27 May 2000 16:29:38 -0400 (EDT)
From: Skip Booth <ebooth@cisco.com>
To: Ly Loi <LLL@shastanets.com>
cc: "'aboba@internaut.com'" <aboba@internaut.com>, l2tp@ipsec.org,
        gwz@cisco.com, wdixon@MICROSOFT.com, townsley@cisco.com
Subject: RE: Comments/questions to l2tp over ipsec draft
In-Reply-To: <4DEC824F6F1FD3119BEF0000E86CEA0101E49778@mailserver.shastanets.com>
Message-ID: <Pine.GSO.4.10.10005271608510.27794-100000@uzura.cisco.com>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-l2tp@ipsec.org
Precedence: bulk



On Fri, 26 May 2000, Ly Loi wrote:

> Skip, security between the client and the LAC isn't what I was referring to.
> In my scenario, the PPP and the LAC can reside on the box, or different
> boxes. However, the LAC doesn't reside on the same box as the IPSEC tunnel
> initiator.
> To be more specific, what I have in mind is a group of LACs sharing 1 IPSEC
> SA. This IPSEC SA is between the gateway and the LNS.
>                      
> LAC1---------gateway------------ipsec_tunnel--------------LNS
> LAC2-----------|
> LAC3-----------|
> 
> Unlike the cases covered by the draft where each L2TP tunnel is mapped to an
> IPSEC SA, here, multiple L2TP tunnels map to one IPSEC SA. If the draft
> could also cover this case, it'd be more complete. 

I have to agree with Glen here.  To me this is really outside the scope of the
L2TP security draft.  This seems to fall into every other IPsec tunnel mode
case here and it doesn't seem to have any specific requirements of L2TP.  If you
think otherwise, then please feel free to specify what we are missing and the
group can decide whether these need to be added to the draft.

> 
> A few more comments:
> 
> 1. Could you not require that L2TP start the retransmission timer only after
> the IPSEC SA has come up (section 9.2.2)? I stated the reason in my original
> email. In addition, in the above scenario, this requirement wouldn't be
> applicable.

Again, I think the draft specifically addresses an implementation where IPsec
and L2TP exist on the same device.  In this case, the text in section 9.2.2
seems to make sense to me.  We can change this to a MAY just to lesson the
requirements of implementations which do not want to have this type of
mechanism.  However, given the amount of time it can take to establish the phase
1/2 SAs, implementations should consider this as a design objective for their
retransmit timers.  Otherwise, with the exponential backoff that is designed
into L2TP, we could be into the 2 to 4 second retransmit period.  In some
extreme situations, even 8 seconds.  That was the reason I added the text on
when to send the SCCRQ.  I would like to hear some feedback from other on
whether this should be a SHOULD or a MAY before we make the change.

> 
> 2. The draft lists the IPSEC filters as if they represent both the IPSEC
> policies and the IPSEC SAs, which is rather confusing since IPSEC SAs are
> derived from IPSEC policies. If you could decouple them in your examples,
> it'd be great.

Is this just a semantics issue for you?  In reality they are filters and we just
like to call them policies.  Every implementation I have ever looked at
specifies this as filters and then associates the filters with a security
policy.  I would prefer to keep the semantics as filters since that allows us to
talk about inbound and outbound filters vs inbound and outbound policies which
doesn't seem to make as much sense to me.  Again I certainly am willing to make
this change however I would like to see some consensus by the group on whether
this actually makes the draft more or less readable/accurate.

> 
> 3. I don't know how others feel about having an external entity (e.g. L2TP)
> injecting some filters into the IPSEC policy database, but this appears to
> me to be a local implementation issue. Also, this wouldn't be feasible for
> the above scenario. I however appreciate the problems the draft is trying to
> solve.

If you are a LNS only and you only support connections on port 1701 then you do
not have to worry about injecting filters.  However if you are a LAC then you
MUST be able to do this to handle the ability for the LNS to float it's port
(Port floating is defined in the L2TP RFC). We went through this in much detail
and could not arrive at a solution which allowed the L2TP engine not to inject
filters without either 1) opening up security holes, 2) decreasing
functionality.  The security holes dealt with the LAC or LNS having to open up
it's port range for all traffic vs just L2TP traffic.  The decrease in
functionality was not allowing L2TP security implementations from moving ports
or addresses.  The solution in the draft accomdates both of these and allows
only L2TP traffic to be sent through the IPsec SAs.

-Skip

> 
> Thanks,
> 
> - Ly
> 
> 
> 
> 
> 
> > -----Original Message-----
> > From: Skip Booth [mailto:ebooth@cisco.com]
> > Sent: Friday, May 26, 2000 12:36 PM
> > To: Ly Loi
> > Cc: 'aboba@internaut.com'; l2tp@ipsec.org; gwz@cisco.com;
> > wdixon@MICROSOFT.com; townsley@cisco.com
> > Subject: RE: Comments/questions to l2tp over ipsec draft
> > 
> > 
> > 
> > 
> > On Fri, 26 May 2000, Ly Loi wrote:
> > 
> > > Skip, I was referring to the case where the LAC and the ipsec tunnel
> > > initiator didn't reside on the same device:
> > > LAC---gateway(ipsec tunnel initiator)---------------LNS(ipsec tunnel
> > > terminator)
> > > 
> > > Is the draft supposed to cover this case too? This case 
> > would be considered
> > > compulsory tunneling, ipsec tunneling :-)
> > 
> > The draft discusses security concerns in the compulsory 
> > model.  Specifically, it
> > talks about how the link between the client and the LAC is 
> > potentially not
> > secure and that clients may require end to end security.  If 
> > the client is
> > running end to end security than potentially no encryption is 
> > required for L2TP
> > data packets between the LAC and LNS, but authentication may 
> > be.  As you noted
> > however, currently there is no way standardized method for IPsec to
> > differentiate between L2TP data and control packets.
> > 
> > Does the draft need to go into more detail in this area?  If 
> > so what more would
> > you like to be discussed?
> > 
> > > If L2TP and IPSEC reside on the same device, then yes, we 
> > all agree IPSEC
> > > would run in transport mode and I trust this will be stated 
> > in a future
> > > version of the draft, thanks.
> > 
> > Yes.
> > 
> > -Skip
> > 
> > > 
> > > - Ly
> > > 
> > 
> > [rest of text snipped]
> > 
> 
> 



From owner-l2tp@ipsec.org  Tue May 30 08:01:48 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id IAA26059
	for <l2tpext-archive@odin.ietf.org>; Tue, 30 May 2000 08:01:47 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id AAA28683; Tue, 30 May 2000 00:54:57 -0700 (PDT)
Received: from mercury.Sun.COM (mercury.Sun.COM [192.9.25.1]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id AAA28679 for <l2tp@ipsec.org>; Tue, 30 May 2000 00:54:43 -0700 (PDT)
Received: from eastmail2.East.Sun.COM ([129.148.1.241])
	by mercury.Sun.COM (8.9.3+Sun/8.9.3) with ESMTP id DAA01065;
	Tue, 30 May 2000 03:18:29 -0700 (PDT)
Received: from dhcp-174-233.east.sun.com (dhcp-174-233.East.Sun.COM [129.148.174.233])
	by eastmail2.East.Sun.COM (8.9.3+Sun/8.9.3/ENSMAIL,v1.7) with ESMTP id GAA12106;
	Tue, 30 May 2000 06:18:29 -0400 (EDT)
Received: (from carlsonj@localhost)
	by dhcp-174-233.east.sun.com (8.9.3+Sun/8.9.3) id GAA08837;
	Tue, 30 May 2000 06:18:28 -0400 (EDT)
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Message-ID: <14643.38260.82608.901088@gargle.gargle.HOWL>
Date: Tue, 30 May 2000 06:18:28 -0400 (EDT)
From: James Carlson <james.d.carlson@east.sun.com>
To: Surendranath <suren@trinc.com>
Cc: l2tp@ipsec.org
Subject: Re: Any free software for l2tp?
In-Reply-To: Surendranath's message of 27 May 2000 19:15:58
References: <3.0.1.32.20000527191558.007771d8@172.16.1.10>
X-Mailer: VM 6.75 under Emacs 20.6.1
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit

Surendranath writes:
> Is there any free software available for LAC and LNS on any platform?

There is this one, though I haven't seen work done on it in a while:

	http://www.marko.net/l2tp/

Also be aware of this:

	http://www.ietf.org/ietf/IPR/CISCO-L2TP

-- 
James Carlson, Internet Engineering       <james.d.carlson@east.sun.com>
SUN Microsystems / 1 Network Drive         71.234W   Vox +1 781 442 2084
MS UBUR02-212 / Burlington MA 01803-2757   42.497N   Fax +1 781 442 1677
"PPP Design and Debugging" --- http://people.ne.mediaone.net/carlson/ppp


From owner-l2tp@ipsec.org  Tue May 30 12:53:10 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id MAA04327
	for <l2tpext-archive@odin.ietf.org>; Tue, 30 May 2000 12:53:09 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id GAA00409; Tue, 30 May 2000 06:16:19 -0700 (PDT)
Received: from baynet.baynetworks.com (ns1.BayNetworks.COM [134.177.3.20]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id GAA00405 for <l2tp@ipsec.org>; Tue, 30 May 2000 06:15:13 -0700 (PDT)
Received: from mailhost.BayNetworks.COM (h016b.s86b1.BayNetworks.COM [134.177.1.107])
	by baynet.baynetworks.com (8.9.1/8.9.1) with ESMTP id IAA12442;
	Tue, 30 May 2000 08:31:55 -0700 (PDT)
Received: from shasta-exch.shastanets.com (mailserver.shastanets.com [47.82.16.150])
	by mailhost.BayNetworks.COM (8.9.1/8.8.8) with ESMTP id IAA03320;
	Tue, 30 May 2000 08:31:53 -0700 (PDT)
Received: by mailserver.shastanets.com with Internet Mail Service (5.5.2650.21)
	id <LWSAJJND>; Tue, 30 May 2000 08:33:10 -0700
Message-ID: <4DEC824F6F1FD3119BEF0000E86CEA0101E49779@mailserver.shastanets.com>
From: Ly Loi <LLL@shastanets.com>
To: "'gwz@cisco.com'" <gwz@cisco.com>, Ly Loi <LLL@shastanets.com>,
        "'Skip Booth'" <ebooth@cisco.com>
Cc: aboba@internaut.com, l2tp@ipsec.org, wdixon@MICROSOFT.com,
        townsley@cisco.com
Subject: RE: Comments/questions to l2tp over ipsec draft
Date: Tue, 30 May 2000 08:33:03 -0700
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

Glen, the LNS is the tunnel terminator; e.g. the security gateway.

- Ly

> -----Original Message-----
> From: Glen Zorn [mailto:gwz@cisco.com]
> Sent: Saturday, May 27, 2000 11:16 AM
> To: Ly Loi; 'Skip Booth'
> Cc: aboba@internaut.com; l2tp@ipsec.org; wdixon@MICROSOFT.com;
> townsley@cisco.com; Glen Zorn
> Subject: RE: Comments/questions to l2tp over ipsec draft
> 
> 
> Ly Loi [mailto://LLL@shastanets.com] writes:
> 
> > Skip, security between the client and the LAC isn't what I was
> > referring to.
> > In my scenario, the PPP and the LAC can reside on the box, 
> or different
> > boxes. However, the LAC doesn't reside on the same box as 
> the IPSEC tunnel
> > initiator.
> > To be more specific, what I have in mind is a group of LACs
> > sharing 1 IPSEC
> > SA. This IPSEC SA is between the gateway and the LNS.
> >
> > LAC1---------gateway------------ipsec_tunnel--------------LNS
> > LAC2-----------|
> > LAC3-----------|
> 
> It appears that you are missing a "security gateway" at the LNS end.
> 
> >
> > Unlike the cases covered by the draft where each L2TP tunnel is
> > mapped to an
> > IPSEC SA, here, multiple L2TP tunnels map to one IPSEC SA. 
> If the draft
> > could also cover this case, it'd be more complete.
> 
> I don't think that the point of the draft is to present an 
> exhaustive list
> of every method (no  matter how lame) that can be used to 
> "secure" L2TP
> tunnels.
> 
> >
> > A few more comments:
> >
> > 1. Could you not require that L2TP start the retransmission timer
> > only after
> > the IPSEC SA has come up (section 9.2.2)? I stated the reason in
> > my original
> > email. In addition, in the above scenario, this requirement 
> wouldn't be
> > applicable.
> >
> > 2. The draft lists the IPSEC filters as if they represent 
> both the IPSEC
> > policies and the IPSEC SAs, which is rather confusing since 
> IPSEC SAs are
> > derived from IPSEC policies. If you could decouple them in 
> your examples,
> > it'd be great.
> >
> > 3. I don't know how others feel about having an external entity
> > (e.g. L2TP)
> > injecting some filters into the IPSEC policy database, but 
> this appears to
> > me to be a local implementation issue. Also, this wouldn't 
> be feasible for
> > the above scenario. I however appreciate the problems the draft
> > is trying to
> > solve.
> >
> > Thanks,
> >
> > - Ly
> >
> >
> >
> >
> >
> > > -----Original Message-----
> > > From: Skip Booth [mailto:ebooth@cisco.com]
> > > Sent: Friday, May 26, 2000 12:36 PM
> > > To: Ly Loi
> > > Cc: 'aboba@internaut.com'; l2tp@ipsec.org; gwz@cisco.com;
> > > wdixon@MICROSOFT.com; townsley@cisco.com
> > > Subject: RE: Comments/questions to l2tp over ipsec draft
> > >
> > >
> > >
> > >
> > > On Fri, 26 May 2000, Ly Loi wrote:
> > >
> > > > Skip, I was referring to the case where the LAC and the 
> ipsec tunnel
> > > > initiator didn't reside on the same device:
> > > > LAC---gateway(ipsec tunnel 
> initiator)---------------LNS(ipsec tunnel
> > > > terminator)
> > > >
> > > > Is the draft supposed to cover this case too? This case
> > > would be considered
> > > > compulsory tunneling, ipsec tunneling :-)
> > >
> > > The draft discusses security concerns in the compulsory
> > > model.  Specifically, it
> > > talks about how the link between the client and the LAC is
> > > potentially not
> > > secure and that clients may require end to end security.  If
> > > the client is
> > > running end to end security than potentially no encryption is
> > > required for L2TP
> > > data packets between the LAC and LNS, but authentication may
> > > be.  As you noted
> > > however, currently there is no way standardized method 
> for IPsec to
> > > differentiate between L2TP data and control packets.
> > >
> > > Does the draft need to go into more detail in this area?  If
> > > so what more would
> > > you like to be discussed?
> > >
> > > > If L2TP and IPSEC reside on the same device, then yes, we
> > > all agree IPSEC
> > > > would run in transport mode and I trust this will be stated
> > > in a future
> > > > version of the draft, thanks.
> > >
> > > Yes.
> > >
> > > -Skip
> > >
> > > >
> > > > - Ly
> > > >
> > >
> > > [rest of text snipped]
> > >
> >
> >
> 


From owner-l2tp@ipsec.org  Tue May 30 13:24:06 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA04914
	for <l2tpext-archive@odin.ietf.org>; Tue, 30 May 2000 13:24:05 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id GAA00456; Tue, 30 May 2000 06:58:44 -0700 (PDT)
Received: from baynet.baynetworks.com (ns1.BayNetworks.COM [134.177.3.20]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id GAA00452 for <l2tp@ipsec.org>; Tue, 30 May 2000 06:56:50 -0700 (PDT)
Received: from mailhost.BayNetworks.COM (h016b.s86b1.BayNetworks.COM [134.177.1.107])
	by baynet.baynetworks.com (8.9.1/8.9.1) with ESMTP id JAA14214;
	Tue, 30 May 2000 09:18:27 -0700 (PDT)
Received: from shasta-exch.shastanets.com (mailserver.shastanets.com [47.82.16.150])
	by mailhost.BayNetworks.COM (8.9.1/8.8.8) with ESMTP id JAA07147;
	Tue, 30 May 2000 09:18:25 -0700 (PDT)
Received: by mailserver.shastanets.com with Internet Mail Service (5.5.2650.21)
	id <LWSAJJSZ>; Tue, 30 May 2000 09:19:43 -0700
Message-ID: <4DEC824F6F1FD3119BEF0000E86CEA0101E4977A@mailserver.shastanets.com>
From: Ly Loi <LLL@shastanets.com>
To: "'Skip Booth'" <ebooth@cisco.com>, Ly Loi <LLL@shastanets.com>
Cc: "'aboba@internaut.com'" <aboba@internaut.com>, l2tp@ipsec.org,
        gwz@cisco.com, wdixon@MICROSOFT.com, townsley@cisco.com
Subject: RE: Comments/questions to l2tp over ipsec draft
Date: Tue, 30 May 2000 09:19:39 -0700
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

Skip, see comments below.

> -----Original Message-----
> From: Skip Booth [mailto:ebooth@cisco.com]
> Sent: Saturday, May 27, 2000 1:30 PM
> To: Ly Loi
> Cc: 'aboba@internaut.com'; l2tp@ipsec.org; gwz@cisco.com;
> wdixon@MICROSOFT.com; townsley@cisco.com
> Subject: RE: Comments/questions to l2tp over ipsec draft
> 
> 
> 
> 
> On Fri, 26 May 2000, Ly Loi wrote:
> 
> > Skip, security between the client and the LAC isn't what I 
> was referring to.
> > In my scenario, the PPP and the LAC can reside on the box, 
> or different
> > boxes. However, the LAC doesn't reside on the same box as 
> the IPSEC tunnel
> > initiator.
> > To be more specific, what I have in mind is a group of LACs 
> sharing 1 IPSEC
> > SA. This IPSEC SA is between the gateway and the LNS.
> >                      
> > LAC1---------gateway------------ipsec_tunnel--------------LNS
> > LAC2-----------|
> > LAC3-----------|
> > 
> > Unlike the cases covered by the draft where each L2TP 
> tunnel is mapped to an
> > IPSEC SA, here, multiple L2TP tunnels map to one IPSEC SA. 
> If the draft
> > could also cover this case, it'd be more complete. 
> 
> I have to agree with Glen here.  To me this is really outside 
> the scope of the
> L2TP security draft.  This seems to fall into every other 
> IPsec tunnel mode
> case here and it doesn't seem to have any specific 
> requirements of L2TP.  If you
> think otherwise, then please feel free to specify what we are 
> missing and the
> group can decide whether these need to be added to the draft.

Well, the title of the draft encompasses the IPSEC tunnel case, so I
expected that scenario to be included. Since the draft is rather
implementation-specific, one has to remember that all of its proposals apply
only to IPSEC in transport mode when L2TP and IPSEC are located in the same
box. May be the title should be changed to "Securing L2TP Using IPSEC in
transport mode"? :-)

> 
> > 
> > A few more comments:
> > 
> > 1. Could you not require that L2TP start the retransmission 
> timer only after
> > the IPSEC SA has come up (section 9.2.2)? I stated the 
> reason in my original
> > email. In addition, in the above scenario, this requirement 
> wouldn't be
> > applicable.
> 
> Again, I think the draft specifically addresses an 
> implementation where IPsec
> and L2TP exist on the same device.  In this case, the text in 
> section 9.2.2
> seems to make sense to me.  We can change this to a MAY just 
> to lesson the
> requirements of implementations which do not want to have this type of
> mechanism.  However, given the amount of time it can take to 
> establish the phase
> 1/2 SAs, implementations should consider this as a design 
> objective for their
> retransmit timers.  Otherwise, with the exponential backoff 
> that is designed
> into L2TP, we could be into the 2 to 4 second retransmit 
> period.  In some
> extreme situations, even 8 seconds.  That was the reason I 
> added the text on
> when to send the SCCRQ.  I would like to hear some feedback 
> from other on
> whether this should be a SHOULD or a MAY before we make the change.

I believe the L2TP exponential backoff gives IPSEC plenty of time to
establish a phase1 and phase2 SAs. If you assume a L2TP retransmission count
of 5, as recommended by the L2TP RFC, you'd have 1+2+4+8+8 = 23 secs. If
IPSEC takes longer than 23 secs to come up, L2TP should time out. But
without the L2TP retransmission timer running, how is it going to time out?
Also, I'd like L2TP to be unaware of IPSEC as much as possible. L2TP should
run its retransmission timer regardless of whether IPSEC is
running/configured. If for any reason (ipsec, infra-structure, network, etc)
L2TP can't come up, L2TP should time out.

> 
> > 
> > 2. The draft lists the IPSEC filters as if they represent 
> both the IPSEC
> > policies and the IPSEC SAs, which is rather confusing since 
> IPSEC SAs are
> > derived from IPSEC policies. If you could decouple them in 
> your examples,
> > it'd be great.
> 
> Is this just a semantics issue for you?  In reality they are 
> filters and we just
> like to call them policies.  Every implementation I have ever 
> looked at
> specifies this as filters and then associates the filters 
> with a security
> policy.  I would prefer to keep the semantics as filters 
> since that allows us to
> talk about inbound and outbound filters vs inbound and 
> outbound policies which
> doesn't seem to make as much sense to me.  Again I certainly 
> am willing to make
> this change however I would like to see some consensus by the 
> group on whether
> this actually makes the draft more or less readable/accurate.

You hit the nail on the head. IPSEC doesn't refer to policies as inbound vs
outbound, only SAs. This draft however talks about inbound/outbound filters.
My interpretation of the draft is that if it's talking about the
configuration before the L2TP/IPSEC negotiations, the filters represent the
IPSEC policies, thus the term 'inbound/outbound filters' are
misleading/confusing. If it's talking about the configuration after the
L2TP/IPSEC negotiations, the filters represent the inbound/outbound SAs,
which is also confusing since previously, they represented IPSEC policies. 

To answer your question, if my interpretations of the draft (as stated in
the above paragraph) are correct, then 'yes', it's a semantics issue.

> 
> > 
> > 3. I don't know how others feel about having an external 
> entity (e.g. L2TP)
> > injecting some filters into the IPSEC policy database, but 
> this appears to
> > me to be a local implementation issue. Also, this wouldn't 
> be feasible for
> > the above scenario. I however appreciate the problems the 
> draft is trying to
> > solve.
> 
> If you are a LNS only and you only support connections on 
> port 1701 then you do
> not have to worry about injecting filters.  However if you 
> are a LAC then you
> MUST be able to do this to handle the ability for the LNS to 
> float it's port
> (Port floating is defined in the L2TP RFC). We went through 
> this in much detail
> and could not arrive at a solution which allowed the L2TP 
> engine not to inject
> filters without either 1) opening up security holes, 2) decreasing
> functionality.  The security holes dealt with the LAC or LNS 
> having to open up
> it's port range for all traffic vs just L2TP traffic.  The decrease in
> functionality was not allowing L2TP security implementations 
> from moving ports
> or addresses.  The solution in the draft accomdates both of 
> these and allows
> only L2TP traffic to be sent through the IPsec SAs.

I understand. However, you are limiting the user to configuring an IPSEC
policy that's applicable only to L2TP. What if the user wants an IPSEC
policy that says for example "I want to encrypt all L2TP traffic and PPTP
traffic"? By implementing the draft, you would have removed the PPTP policy
or one has to be careful in not modifying the whole policy.
I'd like to suggest that the injection of new filters into the IPSEC policy
database be changed from a 'must' to a 'should' or 'may'. From an
implementation and user configuration's point of view, the vendor can always
offer a knob to the user to allow for automatic injection of policies in the
IPSEC policy database should the user desire such behavior; then internally
in the code, the injection could take effect. Otherwise, I don't think we
should make an assumption about what the user wants. What do you think?

Thanks,

- Ly

> 
> -Skip
> 
> > 
> > Thanks,
> > 
> > - Ly
> > 
> > 
> > 
> > 
> > 
> > > -----Original Message-----
> > > From: Skip Booth [mailto:ebooth@cisco.com]
> > > Sent: Friday, May 26, 2000 12:36 PM
> > > To: Ly Loi
> > > Cc: 'aboba@internaut.com'; l2tp@ipsec.org; gwz@cisco.com;
> > > wdixon@MICROSOFT.com; townsley@cisco.com
> > > Subject: RE: Comments/questions to l2tp over ipsec draft
> > > 
> > > 
> > > 
> > > 
> > > On Fri, 26 May 2000, Ly Loi wrote:
> > > 
> > > > Skip, I was referring to the case where the LAC and the 
> ipsec tunnel
> > > > initiator didn't reside on the same device:
> > > > LAC---gateway(ipsec tunnel 
> initiator)---------------LNS(ipsec tunnel
> > > > terminator)
> > > > 
> > > > Is the draft supposed to cover this case too? This case 
> > > would be considered
> > > > compulsory tunneling, ipsec tunneling :-)
> > > 
> > > The draft discusses security concerns in the compulsory 
> > > model.  Specifically, it
> > > talks about how the link between the client and the LAC is 
> > > potentially not
> > > secure and that clients may require end to end security.  If 
> > > the client is
> > > running end to end security than potentially no encryption is 
> > > required for L2TP
> > > data packets between the LAC and LNS, but authentication may 
> > > be.  As you noted
> > > however, currently there is no way standardized method 
> for IPsec to
> > > differentiate between L2TP data and control packets.
> > > 
> > > Does the draft need to go into more detail in this area?  If 
> > > so what more would
> > > you like to be discussed?
> > > 
> > > > If L2TP and IPSEC reside on the same device, then yes, we 
> > > all agree IPSEC
> > > > would run in transport mode and I trust this will be stated 
> > > in a future
> > > > version of the draft, thanks.
> > > 
> > > Yes.
> > > 
> > > -Skip
> > > 
> > > > 
> > > > - Ly
> > > > 
> > > 
> > > [rest of text snipped]
> > > 
> > 
> > 
> 


From owner-l2tp@ipsec.org  Tue May 30 14:46:50 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id OAA06862
	for <l2tpext-archive@odin.ietf.org>; Tue, 30 May 2000 14:46:49 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id IAA00585; Tue, 30 May 2000 08:00:33 -0700 (PDT)
Received: from baynet.baynetworks.com (ns1.BayNetworks.COM [134.177.3.20]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id HAA00574 for <l2tp@ipsec.org>; Tue, 30 May 2000 07:58:24 -0700 (PDT)
Received: from mailhost.BayNetworks.COM (h016b.s86b1.BayNetworks.COM [134.177.1.107])
	by baynet.baynetworks.com (8.9.1/8.9.1) with ESMTP id KAA16205;
	Tue, 30 May 2000 10:20:12 -0700 (PDT)
Received: from shasta-exch.shastanets.com (mailserver.shastanets.com [47.82.16.150])
	by mailhost.BayNetworks.COM (8.9.1/8.8.8) with ESMTP id KAA12283;
	Tue, 30 May 2000 10:20:11 -0700 (PDT)
Received: by mailserver.shastanets.com with Internet Mail Service (5.5.2650.21)
	id <LWSAJJ6G>; Tue, 30 May 2000 10:21:29 -0700
Message-ID: <4DEC824F6F1FD3119BEF0000E86CEA0101E4977F@mailserver.shastanets.com>
From: Ly Loi <LLL@shastanets.com>
To: Ly Loi <LLL@shastanets.com>, "'Skip Booth'" <ebooth@cisco.com>
Cc: "'aboba@internaut.com'" <aboba@internaut.com>,
        "'l2tp@ipsec.org'"
	 <l2tp@ipsec.org>,
        "'gwz@cisco.com'" <gwz@cisco.com>,
        "'wdixon@MICROSOFT.com'" <wdixon@MICROSOFT.com>,
        "'townsley@cisco.com'"
	 <townsley@cisco.com>
Subject: RE: Comments/questions to l2tp over ipsec draft
Date: Tue, 30 May 2000 10:21:27 -0700
MIME-Version: 1.0
X-Mailer: Internet Mail Service (5.5.2650.21)
Content-Type: text/plain;
	charset="iso-8859-1"
Sender: owner-l2tp@ipsec.org
Precedence: bulk


Skip, my example below regarding configuring "I want to encrypt all L2TP
traffic and PPTP traffic" isn't allowed by IPSEC, so never mind this
example; sorry about that. But I still would like to see the injection not
be a requirement.

- Ly

> -----Original Message-----
> From: Ly Loi 
> Sent: Tuesday, May 30, 2000 9:20 AM
> To: 'Skip Booth'; Ly Loi
> Cc: 'aboba@internaut.com'; l2tp@ipsec.org; gwz@cisco.com;
> wdixon@MICROSOFT.com; townsley@cisco.com
> Subject: RE: Comments/questions to l2tp over ipsec draft
> 
> 
> Skip, see comments below.
> 
> > -----Original Message-----
> > From: Skip Booth [mailto:ebooth@cisco.com]
> > Sent: Saturday, May 27, 2000 1:30 PM
> > To: Ly Loi
> > Cc: 'aboba@internaut.com'; l2tp@ipsec.org; gwz@cisco.com;
> > wdixon@MICROSOFT.com; townsley@cisco.com
> > Subject: RE: Comments/questions to l2tp over ipsec draft
> > 
> > 
> > 
> > 
> > On Fri, 26 May 2000, Ly Loi wrote:
> > 
> > > Skip, security between the client and the LAC isn't what I 
> > was referring to.
> > > In my scenario, the PPP and the LAC can reside on the box, 
> > or different
> > > boxes. However, the LAC doesn't reside on the same box as 
> > the IPSEC tunnel
> > > initiator.
> > > To be more specific, what I have in mind is a group of LACs 
> > sharing 1 IPSEC
> > > SA. This IPSEC SA is between the gateway and the LNS.
> > >                      
> > > LAC1---------gateway------------ipsec_tunnel--------------LNS
> > > LAC2-----------|
> > > LAC3-----------|
> > > 
> > > Unlike the cases covered by the draft where each L2TP 
> > tunnel is mapped to an
> > > IPSEC SA, here, multiple L2TP tunnels map to one IPSEC SA. 
> > If the draft
> > > could also cover this case, it'd be more complete. 
> > 
> > I have to agree with Glen here.  To me this is really outside 
> > the scope of the
> > L2TP security draft.  This seems to fall into every other 
> > IPsec tunnel mode
> > case here and it doesn't seem to have any specific 
> > requirements of L2TP.  If you
> > think otherwise, then please feel free to specify what we are 
> > missing and the
> > group can decide whether these need to be added to the draft.
> 
> Well, the title of the draft encompasses the IPSEC tunnel 
> case, so I expected that scenario to be included. Since the 
> draft is rather implementation-specific, one has to remember 
> that all of its proposals apply only to IPSEC in transport 
> mode when L2TP and IPSEC are located in the same box. May be 
> the title should be changed to "Securing L2TP Using IPSEC in 
> transport mode"? :-)
> 
> > 
> > > 
> > > A few more comments:
> > > 
> > > 1. Could you not require that L2TP start the retransmission 
> > timer only after
> > > the IPSEC SA has come up (section 9.2.2)? I stated the 
> > reason in my original
> > > email. In addition, in the above scenario, this requirement 
> > wouldn't be
> > > applicable.
> > 
> > Again, I think the draft specifically addresses an 
> > implementation where IPsec
> > and L2TP exist on the same device.  In this case, the text in 
> > section 9.2.2
> > seems to make sense to me.  We can change this to a MAY just 
> > to lesson the
> > requirements of implementations which do not want to have 
> this type of
> > mechanism.  However, given the amount of time it can take to 
> > establish the phase
> > 1/2 SAs, implementations should consider this as a design 
> > objective for their
> > retransmit timers.  Otherwise, with the exponential backoff 
> > that is designed
> > into L2TP, we could be into the 2 to 4 second retransmit 
> > period.  In some
> > extreme situations, even 8 seconds.  That was the reason I 
> > added the text on
> > when to send the SCCRQ.  I would like to hear some feedback 
> > from other on
> > whether this should be a SHOULD or a MAY before we make the change.
> 
> I believe the L2TP exponential backoff gives IPSEC plenty of 
> time to establish a phase1 and phase2 SAs. If you assume a 
> L2TP retransmission count of 5, as recommended by the L2TP 
> RFC, you'd have 1+2+4+8+8 = 23 secs. If IPSEC takes longer 
> than 23 secs to come up, L2TP should time out. But without 
> the L2TP retransmission timer running, how is it going to 
> time out? Also, I'd like L2TP to be unaware of IPSEC as much 
> as possible. L2TP should run its retransmission timer 
> regardless of whether IPSEC is running/configured. If for any 
> reason (ipsec, infra-structure, network, etc) L2TP can't come 
> up, L2TP should time out.
> 
> > 
> > > 
> > > 2. The draft lists the IPSEC filters as if they represent 
> > both the IPSEC
> > > policies and the IPSEC SAs, which is rather confusing since 
> > IPSEC SAs are
> > > derived from IPSEC policies. If you could decouple them in 
> > your examples,
> > > it'd be great.
> > 
> > Is this just a semantics issue for you?  In reality they are 
> > filters and we just
> > like to call them policies.  Every implementation I have ever 
> > looked at
> > specifies this as filters and then associates the filters 
> > with a security
> > policy.  I would prefer to keep the semantics as filters 
> > since that allows us to
> > talk about inbound and outbound filters vs inbound and 
> > outbound policies which
> > doesn't seem to make as much sense to me.  Again I certainly 
> > am willing to make
> > this change however I would like to see some consensus by the 
> > group on whether
> > this actually makes the draft more or less readable/accurate.
> 
> You hit the nail on the head. IPSEC doesn't refer to policies 
> as inbound vs outbound, only SAs. This draft however talks 
> about inbound/outbound filters. My interpretation of the 
> draft is that if it's talking about the configuration before 
> the L2TP/IPSEC negotiations, the filters represent the IPSEC 
> policies, thus the term 'inbound/outbound filters' are 
> misleading/confusing. If it's talking about the configuration 
> after the L2TP/IPSEC negotiations, the filters represent the 
> inbound/outbound SAs, which is also confusing since 
> previously, they represented IPSEC policies. 
> 
> To answer your question, if my interpretations of the draft 
> (as stated in the above paragraph) are correct, then 'yes', 
> it's a semantics issue.
> 
> > 
> > > 
> > > 3. I don't know how others feel about having an external 
> > entity (e.g. L2TP)
> > > injecting some filters into the IPSEC policy database, but 
> > this appears to
> > > me to be a local implementation issue. Also, this wouldn't 
> > be feasible for
> > > the above scenario. I however appreciate the problems the 
> > draft is trying to
> > > solve.
> > 
> > If you are a LNS only and you only support connections on 
> > port 1701 then you do
> > not have to worry about injecting filters.  However if you 
> > are a LAC then you
> > MUST be able to do this to handle the ability for the LNS to 
> > float it's port
> > (Port floating is defined in the L2TP RFC). We went through 
> > this in much detail
> > and could not arrive at a solution which allowed the L2TP 
> > engine not to inject
> > filters without either 1) opening up security holes, 2) decreasing
> > functionality.  The security holes dealt with the LAC or LNS 
> > having to open up
> > it's port range for all traffic vs just L2TP traffic.  The 
> decrease in
> > functionality was not allowing L2TP security implementations 
> > from moving ports
> > or addresses.  The solution in the draft accomdates both of 
> > these and allows
> > only L2TP traffic to be sent through the IPsec SAs.
> 
> I understand. However, you are limiting the user to 
> configuring an IPSEC policy that's applicable only to L2TP. 
> What if the user wants an IPSEC policy that says for example 
> "I want to encrypt all L2TP traffic and PPTP traffic"? By 
> implementing the draft, you would have removed the PPTP 
> policy or one has to be careful in not modifying the whole policy.
> I'd like to suggest that the injection of new filters into 
> the IPSEC policy database be changed from a 'must' to a 
> 'should' or 'may'. From an implementation and user 
> configuration's point of view, the vendor can always offer a 
> knob to the user to allow for automatic injection of policies 
> in the IPSEC policy database should the user desire such 
> behavior; then internally in the code, the injection could 
> take effect. Otherwise, I don't think we should make an 
> assumption about what the user wants. What do you think?
> 
> Thanks,
> 
> - Ly
> 
> > 
> > -Skip
> > 
> > > 
> > > Thanks,
> > > 
> > > - Ly
> > > 
> > > 
> > > 
> > > 
> > > 
> > > > -----Original Message-----
> > > > From: Skip Booth [mailto:ebooth@cisco.com]
> > > > Sent: Friday, May 26, 2000 12:36 PM
> > > > To: Ly Loi
> > > > Cc: 'aboba@internaut.com'; l2tp@ipsec.org; gwz@cisco.com;
> > > > wdixon@MICROSOFT.com; townsley@cisco.com
> > > > Subject: RE: Comments/questions to l2tp over ipsec draft
> > > > 
> > > > 
> > > > 
> > > > 
> > > > On Fri, 26 May 2000, Ly Loi wrote:
> > > > 
> > > > > Skip, I was referring to the case where the LAC and the 
> > ipsec tunnel
> > > > > initiator didn't reside on the same device:
> > > > > LAC---gateway(ipsec tunnel 
> > initiator)---------------LNS(ipsec tunnel
> > > > > terminator)
> > > > > 
> > > > > Is the draft supposed to cover this case too? This case 
> > > > would be considered
> > > > > compulsory tunneling, ipsec tunneling :-)
> > > > 
> > > > The draft discusses security concerns in the compulsory 
> > > > model.  Specifically, it
> > > > talks about how the link between the client and the LAC is 
> > > > potentially not
> > > > secure and that clients may require end to end security.  If 
> > > > the client is
> > > > running end to end security than potentially no encryption is 
> > > > required for L2TP
> > > > data packets between the LAC and LNS, but authentication may 
> > > > be.  As you noted
> > > > however, currently there is no way standardized method 
> > for IPsec to
> > > > differentiate between L2TP data and control packets.
> > > > 
> > > > Does the draft need to go into more detail in this area?  If 
> > > > so what more would
> > > > you like to be discussed?
> > > > 
> > > > > If L2TP and IPSEC reside on the same device, then yes, we 
> > > > all agree IPSEC
> > > > > would run in transport mode and I trust this will be stated 
> > > > in a future
> > > > > version of the draft, thanks.
> > > > 
> > > > Yes.
> > > > 
> > > > -Skip
> > > > 
> > > > > 
> > > > > - Ly
> > > > > 
> > > > 
> > > > [rest of text snipped]
> > > > 
> > > 
> > > 
> > 
> 


From owner-l2tp@ipsec.org  Tue May 30 17:52:19 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id RAA11301
	for <l2tpext-archive@odin.ietf.org>; Tue, 30 May 2000 17:52:18 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id LAA00973; Tue, 30 May 2000 11:31:25 -0700 (PDT)
Received: from chef.redback.com (chef.redback.com [155.53.36.21]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id LAA00969 for <l2tp@ipsec.org>; Tue, 30 May 2000 11:31:19 -0700 (PDT)
Received: (from carrel@localhost)
	by chef.redback.com (8.8.8/8.8.8) id NAA04244
	for l2tp@ipsec.org; Tue, 30 May 2000 13:54:40 -0700 (PDT)
Received: from brahma.roc.com ([202.54.67.218]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id EAA22226 for <l2tp@ipsec.org>; Sat, 27 May 2000 04:22:26 -0700 (PDT)
Received: from suren.roc.com ([172.16.5.74])
	by brahma.roc.com (8.8.5/8.8.5) with SMTP id TAA03724
	for <l2tp@ipsec.org>; Sat, 27 May 2000 19:20:30 +0530 (IST)
Message-Id: <3.0.1.32.20000527190931.007771d8@172.16.1.10>
X-Sender: suren@172.16.1.10
X-Mailer: Windows Eudora Light Version 3.0.1 (32)
Date: Sat, 27 May 2000 19:09:31 +0500
To: l2tp@ipsec.org
From: Surendranath <suren@trinc.com>
Subject: help needed regarding l2tp - nat issues
Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Sender: owner-l2tp@ipsec.org
Precedence: bulk

Hi 
My name is surendra, working with ROC, Hyderabad, India.
I have a question regarding l2tp-nat issues.

My NAT server is in between LAC and LNS.
As the Source port of a packet, which is comming from LNS, is 
may not be equal to 1701, I want to exclude the source port field
to find a match at NAT server. I am thinking, this is the only change,
I need to do in nat server to deal with l2tp packets. 
Am I missing something??

Can anybody please explain me, what is the actual problem apart from
this in l2tp-nat case? Is there any need to look into Tunnel Id/ Session Id 
in l2tp header and to do any demultiplexing?   Because, to extract these
from Assigned-Tunnel-ID/ Assigned SessionId AVPs, NAT server dont know the
shared secret. 

thanks,
surendra.



From owner-l2tp@ipsec.org  Wed May 31 03:10:43 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id DAA01117
	for <l2tpext-archive@odin.ietf.org>; Wed, 31 May 2000 03:10:43 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id UAA02090; Tue, 30 May 2000 20:27:32 -0700 (PDT)
Received: from fsnt.future.futusoft.com ([203.197.140.35]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id UAA02086 for <l2tp@ipsec.org>; Tue, 30 May 2000 20:27:14 -0700 (PDT)
Received: from kailash.future.futsoft.com (unverified) by fsnt.future.futusoft.com
 (Content Technologies SMTPRS 2.0.15) with ESMTP id <B0000588121@fsnt.future.futusoft.com>;
 Wed, 31 May 2000 11:31:18 +0530
Received: from himalaya.future.futsoft.com (himalaya.future.futsoft.com [10.0.10.15]) by kailash.future.futsoft.com (8.7.1/8.7.1) with ESMTP id LAA27600; Wed, 31 May 2000 11:06:43 +0530
Received: from laxmanb.future.futsoft.com ([10.0.12.128]) by himalaya.future.futsoft.com (8.6.11/8.6.9) with SMTP id LAA26067; Wed, 31 May 2000 11:21:49 +0530
Received: by localhost with Microsoft MAPI; Wed, 31 May 2000 11:14:07 +0530
Message-Id: <01BFCAF1.56768440.laxmanb@future.futsoft.com>
From: LAXMANA RAO <laxmanb@future.futsoft.com>
To: "'l2tp@ipsec.org'" <l2tp@ipsec.org>
Cc: "'townsley@cisco.com'" <townsley@cisco.com>
Subject: I need small clarification ......................
Date: Wed, 31 May 2000 11:14:07 +0530
X-Mailer: Microsoft Internet E-mail/MAPI - 8.0.0.4211
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit

Hi,
     Can i get the Free s/w  for LAC in any platform..?if U have any idea please let me know the site or location to download the same.

Thanks & regards
laxman 

LAXMANA RAO B
SOFTWARE ENGINEER
FUTURE SOFTWARE PVT.LTD.
CHENNAI,INDIA
PH.NO:91-44-4330550
E-MAIL:laxmanb@future.futsoft.com




From owner-l2tp@ipsec.org  Wed May 31 13:22:46 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id NAA17176
	for <l2tpext-archive@odin.ietf.org>; Wed, 31 May 2000 13:22:45 -0400 (EDT)
Received: (from majordomo@localhost) by ns1.sailpix.com (8.8.8/8.6.9) id GAA03884; Wed, 31 May 2000 06:25:15 -0700 (PDT)
Received: from sj-msg-core-1.cisco.com (sj-msg-core-1.cisco.com [171.71.163.11]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id GAA03878 for <l2tp@ipsec.org>; Wed, 31 May 2000 06:25:05 -0700 (PDT)
Received: from townsley-ultra.cisco.com (townsley-ultra.cisco.com [161.44.58.38])
	by sj-msg-core-1.cisco.com (8.9.3/8.9.1) with ESMTP id IAA05124
	for <l2tp@ipsec.org>; Wed, 31 May 2000 08:48:36 -0700 (PDT)
Received: from cisco.com (townsley-home.cisco.com [161.44.99.50]) by townsley-ultra.cisco.com (8.8.4-Cisco.1/CISCO.WS.1.2) with ESMTP id LAA23710 for <l2tp@ipsec.org>; Wed, 31 May 2000 11:48:27 -0400 (EDT)
Message-ID: <39353426.D81259FA@cisco.com>
Date: Wed, 31 May 2000 11:47:50 -0400
From: "W. Mark Townsley" <townsley@cisco.com>
Organization: cisco Systems
X-Mailer: Mozilla 4.7 [en] (WinNT; U)
X-Accept-Language: en
MIME-Version: 1.0
To: l2tp@ipsec.org
Subject: L2TP MIB - Working Group Last Call
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit


This is last call.  I will advise the area directors that we want to
take Layer Two Tunneling Protocol "L2TP" Management Information Base
<draft-ietf-l2tpext-l2tp-mib-00.txt>) to Proposed Standard on June 14,
2000 unless there is substantive comment raised on this list.

- Mark


From owner-l2tp@ipsec.org  Wed May 31 21:09:28 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA29084
	for <l2tpext-archive@odin.ietf.org>; Wed, 31 May 2000 21:09:27 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id OAA04516; Wed, 31 May 2000 14:19:07 -0700 (PDT)
Received: from mw.3com.com (intergate.usr.com [149.112.20.3]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id OAA04509 for <l2tp@ipsec.org>; Wed, 31 May 2000 14:15:59 -0700 (PDT)
Received: from mwgate02.mw.3com.com by mw.3com.com (8.8.5/3.1.090690-3Com Corporation)
	id SAA10288; Wed, 31 May 2000 18:39:54 -0500 (CDT)
Received: by mwgate02.mw.3com.com(Lotus SMTP MTA v4.6.7  (934.1 12-30-1999))  id 862568F0.00823B34 ; Wed, 31 May 2000 18:42:28 -0500
X-Lotus-FromDomain: 3COM@3COM-MWGATE
From: "Madhvi Verma" <Madhvi_Verma@mw.3com.com>
To: internet-drafts@ietf.org
cc: l2tp@ipsec.org, rverma@dc.com, james.d.carlson@east.sun.com,
        "Ching Kung" <Ching_Kung@mw.3com.com>
Message-ID: <862568F0.00823A12.00@mwgate02.mw.3com.com>
Date: Wed, 31 May 2000 18:40:19 -0500
Subject: draft-ietf-l2tpext-ppp-discinfo-00.txt
Mime-Version: 1.0
Content-type: text/plain; charset=us-ascii
Content-Disposition: inline
Sender: owner-l2tp@ipsec.org
Precedence: bulk





Network Working Group                                        Rohit Verma
Internet Draft                                       Deloitte Consulting
Expires November 2000                                       Madhvi Verma
                                                        3Com Corporation
                                                           James Carlson
                                                        Sun Microsystems
                                                                May 2000


                   L2TP Disconnect Cause Information
                <draft-ietf-l2tpext-ppp-discinfo-00.txt>

Status of this Memo

   This document is an Internet-Draft and is in full conformance with
   all provisions of Section 10 of RFC2026.

   Internet-Drafts are working documents of the Internet Engineering
   Task Force (IETF), its areas, and its working groups.  Note that
   other groups may also distribute working documents as Internet-
   Drafts.

   Internet-Drafts are draft documents valid for a maximum of six months
   and may be updated, replaced, or obsoleted by other documents at any
   time.  It is inappropriate to use Internet-Drafts as reference
   material or to cite them other than as "work in progress."

   The list of current Internet-Drafts can be accessed at
   http://www.ietf.org/ietf/1id-abstracts.txt

   The list of Internet-Draft Shadow Directories can be accessed at
   http://www.ietf.org/shadow.html.

Abstract

   The Layer 2 Tunneling Protocol (``L2TP'') defines a mechanism for
   tunneling PPP sessions.  The current RFC lacks a mechanism for the
   LNS to provide PPP related disconnect cause information to the LAC.
   This information, provided by the extension described in this
   document, can be useful for accounting and debugging purposes.


1.  Introduction

   L2TP [1] defines a general-purpose mechanism for tunneling PPP over
   various media.  By design, it insulates L2TP operation from the
   details of the PPP session that is being encapsulated by L2TP.  There
   are, however, cases where it may be desirable for PPP-specific



expires November 2000    Verma, Verma, Carlson                  [Page 1]

INTERNET DRAFT           L2TP Disconnect Cause                  May 2000


   disconnect information to be provided to an L2TP host (LAC/LNS) in a
   descriptive format. The lack of this information is especially a
   problem when the LAC and LNS are not managed by the same entities.

   The Result and Error Codes defined for L2TP specify only L2TP-
   specific disconnect information. This document provides an additional
   AVP that MAY be used by an L2TP host to provide PPP specific discon-
   nect information to its peer for accounting and debugging purposes.
   This AVP should be used in conjunction with, and not as a replacement
   for, the Result and Error Code AVPs.

   The PPP Disconnect Cause Code AVP can also be used to provide a
   human-readable disconnect reason to the user. The following AVP
   should not have any effect on either the functioning of the tunnel or
   the functioning of the PPP session.


2.  AVP


2.1.  PPP Disconnect Cause Code

   The AVP is valid in the CDN message only, and it MUST NOT be marked
   Mandatory.

   The PPP Disconnect Cause Code AVP is encoded as Vendor ID 43 (3Com
   Corporation), with an Attribute Type of PPP Disconnect Cause Code
   (46).  The attribute value is one of the values discussed in section
   2.1.1.  The length of the value field MUST be at least 11 octets.  If
   the length is more than 11 octets, the additional octets SHOULD con-
   tain a descriptive text in ASCII format that can be displayed to the
   user or in a log file.

      0                   1                   2                   3
      0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 0 1
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |M|H| rsvd  |      Length       |           Vendor ID         |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |         Attribute Type        |       Disconnect Code       |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+
      |  CP Number                    | Direction     | Message...  |
      +-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+-+

                Figure 1.1 PPP Disconnect Cause code AVP

   Mandatory (M) bit: MUST be 0.

   Hidden (H) bit: MAY be 1 if the attribute is hidden.



expires November 2000    Verma, Verma, Carlson                  [Page 2]

INTERNET DRAFT           L2TP Disconnect Cause                  May 2000


   Length: The length of the entire attribute in octets, expressed as a
   single octet.  MUST be at least 11.

   Vendor ID: A two octet value in network byte order; set to 43 (3Com
   Corporation).

   Attribute Type: A two octet value in network byte order; set to 46
   (PPP Disconnect Cause Code).

   Disconnect Code: A two octet value in network byte order.

   CP Number: Primary failing Control Protocol, if any.  This field may
   be 0 unless mentioned otherwise in the description of the Disconnect
   Cause Codes in section 2.1.1.

   Direction(D) Flag: A single octet value; specifies the direction in
   which the Disconnect code applies.

           The valid values of this field are:

                   0: global error
                   1: at peer
                   2: at local
                   3-255: Reserved

   This field may be 0 unless mentioned otherwise in the description of
   the Disconnect Cause Codes in section 2.1.1.


2.1.1.  Disconnect Cause Codes

   This section contains the list of well-known values of the Disconnect
   Code field in the PPP Disconnect Cause Code AVP. The IANA will main-
   tain a registry of the up-to-date values. These values should be used
   in conjunction with the Direction flag and the Control Protocol
   Number to get the specific error code.


2.1.1.1.  Global Errors

   The global error codes, given in the list below, are Disconnect Codes
   that do not relate to one particular control protocol. The CP Number
   for these errors MUST be set to 0 and the Direction (D) flag MUST
   also be set to 0 unless specified otherwise for a given code.

   0    No information available.

   1    Administrative disconnect.



expires November 2000    Verma, Verma, Carlson                  [Page 3]

INTERNET DRAFT           L2TP Disconnect Cause                  May 2000


   2    LCP renegotiation at LNS disabled; LNS expects proxy LCP
        information, LAC did not send it.

   3    Normal Disconnection, LCP Terminate-Request sent.

        Valid Direction flags:

           1: By peer
           2: By local

   4    Compulsory encryption required by a PPP peer was refused by the
        other.

        Valid Direction flags are:

           1: Required by local; refused by peer
           2: Required by peer; refused by local


2.1.1.2.  LCP Errors

   The LCP error codes, listed below, are disconnect reasons that are
   directly related to failure of PPP peers to negotiate mutually agree-
   able link parameters. The CP Number for such errors MUST be set to
   0xCO21 and the direction flag to the direction of the failure.

   5    FSM Timeout error.

   6    No recognizable LCP packets were received.

   7    LCP failure-Magic number error; link possibly looped back.

   8    LCP link failure-Echo Request timeout.

   9    Peer has unexpected Endpoint-Discriminator for existing MP
        bundle.

   10   Peer has unexpected MRRU for existing MP bundle.

   11   Peer has unexpected Short-Sequence-Number option for existing
        MP bundle.

   12   Compulsory callback required by a PPP peer was refused by the
        other.

        Valid Direction flags are:

           1: Required by local; refused by peer



expires November 2000    Verma, Verma, Carlson                  [Page 4]

INTERNET DRAFT           L2TP Disconnect Cause                  May 2000


           2: Required by peer; refused by local


2.1.1.3.  Authentication Errors

   The authentication errors codes, listed below, are disconnect reasons
   that are directly related to authentication failures between the PPP
   peers. The CP Number for such errors MUST correspond to the control
   protocol number for the authentication protocol in use and the direc-
   tion flag to the direction of the failure.

   13   FSM Timeout error.

   14   Peer has unexpected authenticated name for existing MP bundle.

   15   PPP authentication failure-Authentication protocol unacceptable.

        Valid Direction flags are:

           1: All local authentication protocols were rejected by the
              peer.

           2: All authentication protocols requested by peer were
              unacceptable or unimplemented locally.

   16   PPP authentication failure-Authentication failed (bad name,
        password, or secret).

        Valid Direction flags are:

           1: Authentication of peer identity by local system.
           2: Authentication of local identity by peer system.


2.1.1.4.  Network Control Protocol Errors

   NCP Errors are disconnect reasons that are directly related to the
   failure of PPP peers to negotiate a mutually agreeable set of parame-
   ters for the network protocols. The CP Number for such errors MUST
   correspond to the network control protocol number in use and the
   direction flag to the direction of the failure. Where multiple net-
   work protocols are in use, multiple copies of this AVP may be given
   to indicate failure reasons for each NCP.

   The Control Protocol field in the following cases will be the last or
   primary failing NCP.

   17   FSM Timeout error.



expires November 2000    Verma, Verma, Carlson                  [Page 5]

INTERNET DRAFT           L2TP Disconnect Cause                  May 2000


   18   No NCPs available (all disabled or rejected); no NCPs went to
        Opened state.

   19   NCP failure: failed to converge on acceptable addresses.

   20   NCP failure: user not permitted to use any addresses.


3.  Notes

   This AVP MAY always be sent by the LNS to the LAC. However, at the
   LAC, this AVP would be more useful diagnostically in the compulsory
   case than in the voluntary case. In a voluntary model, it can be
   presumed that local PPP traces are always present.


4.  IANA Considerations

   This AVP includes an enumerated cause code value.  Values 0 through
   20 are described in this document.  New values may be assigned by the
   IANA.


5.  References

      [1]   W. Townsley, A. Valencia, A. Rubens, G. Pall, G. Zorn,
            B. Palter, ``Layer 2 Tunnel Protocol (L2TP)'', RFC2661,
            August 1999.

6.  Acknowledgements

   The Authors would like to thank W. Mark Townsley for his useful com-
   ments to an earlier version of this document.


7.  Contacts


7.1.  L2TP Working Group Chair

   W. Mark Townsley
   Cisco Systems
   7025 Kit Creek Road
   PO Box 14987
   Research Triangle Park, NC 27709
   Email:  <townsley@cisco.com>





expires November 2000    Verma, Verma, Carlson                  [Page 6]

INTERNET DRAFT           L2TP Disconnect Cause                  May 2000


7.2.  Authors

   Rohit Verma
   180 N. Stetson Avenue
   Chicago IL 60601
   Phone: +1 312 374 2475
   Fax:   +1 312 870 2475
   Email:  <rverma@dc.com>

   Madhvi Verma
   1800 W. Central Road
   Mount Prospect IL 60056
   Phone: +1 847 797 6528
   Fax:   +1 847 222 2424
   Email:  <Madhvi_Verma@mw.3com.com>

   James Carlson
   Sun Microsystems
   1 Network Drive MS UBUR02-212
   Burlington MA  01803-2757
   Email:  <james.d.carlson@east.sun.com>
   Voice:  +1 781 442 2084
   Fax:    +1 781 442 1677



8.  Notices

   The IETF takes no position regarding the validity or scope of any
   intellectual property or other rights that might be claimed to per-
   tain to the implementation or use of the technology described in this
   document or the extent to which any license under such rights might
   or might not be available; neither does it represent that it has made
   any effort to identify any such rights.  Information on the IETF's
   procedures with respect to rights in standards-track and standards-
   related documentation can be found in BCP-11.  Copies of claims of
   rights made available for publication and any assurances of licenses
   to be made available, or the result of an attempt made to obtain a
   general license or permission for the use of such proprietary rights
   by implementers or users of this specification can be obtained from
   the IETF Secretariat.

   The IETF invites any interested party to bring to its attention any
   copyrights, patents or patent applications, or other proprietary
   rights, which may cover technology that, may be required to practice
   this standard.  Please address the information to the IETF Executive
   Director.




expires November 2000    Verma, Verma, Carlson                  [Page 7]

INTERNET DRAFT           L2TP Disconnect Cause                  May 2000





















































expires November 2000    Verma, Verma, Carlson                  [Page 8]




From owner-l2tp@ipsec.org  Wed May 31 21:27:10 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id VAA29265
	for <l2tpext-archive@odin.ietf.org>; Wed, 31 May 2000 21:27:10 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id OAA04541; Wed, 31 May 2000 14:38:11 -0700 (PDT)
Received: from zendo.com (bodhi.zendo.com [205.187.71.2]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id OAA04535 for <l2tp@ipsec.org>; Wed, 31 May 2000 14:37:35 -0700 (PDT)
Received: from vandys-pc.zendo.com (dialup-63.214.15.3.Seattle1.Level3.net [63.214.15.3])
	by zendo.com (8.10.1/8.10.1) with ESMTP id e4VGGL702355
	for <l2tp@ipsec.org>; Wed, 31 May 2000 16:16:22 GMT
Received: from vandys-pc.zendo.com (localhost.zendo.com [127.0.0.1])
	by vandys-pc.zendo.com (8.9.1/8.9.1) with ESMTP id RAA00360
	for <l2tp@ipsec.org>; Wed, 31 May 2000 17:00:54 -0700 (PDT)
	(envelope-from vandys@vandys-pc.zendo.com)
Message-Id: <200006010000.RAA00360@vandys-pc.zendo.com>
To: l2tp@ipsec.org
Subject: L2TPHC changes
Date: Wed, 31 May 2000 17:00:54 -0700
From: Andy Valencia <vandys@zendo.com>
Sender: owner-l2tp@ipsec.org
Precedence: bulk

From review comments on L2TPHC, a couple of slightly-more-than-trivial
changes appear to be appropriate for the draft.  They won't be hard to put
in, but Mark Townsley suggested I post them to the list first.

1. Session ID's rather than Tunnel ID's

Originally, L2TPHC was designed to permit a single session in a single
tunnel.  Early review input resulted in consensus that more than one session
should be supported.  Since multiple tunnels was a superset of everything
which could be done with multiple sessions, I put in the ability for more
than one tunnel, while keeping the requirement of a single session within
the tunnel.

Now that the dust has settled on L2TP, it seems like I should've done it the
other way around--one (and only one) tunnel, with sessions within.  The key
points Mark brought up are that (1) this saves you a number of messages, (2)
you don't have to do authentication, and (3) (this is the key one!) the flow
control mechanisms for control messages work a lot better when you're
dealing with numerous messages under a single tunnel.  Remember, flow
control is per tunnel, so if you bring up thousands of tunnels quickly,
you'll basically have no flow control to help you out.

(This is a problem latent in L2TP, too.  But since it's also latent in TCP,
I'm just not going to worry about it until we see specific problems.)

2. IP protocol number is implicit

The AVP for enabling L2TPHC indicated the IP protocol number.  I did this
primarily because I was having trouble getting an official protocol number.
A timely event got us this protocol number, so Mark thought it would be
tidier to keep it out of the way (we were mostly thinking about "things
which would make the IESG nervous).  So now the AVP does not have a Value
indicating the assigned protocol number.  The IP protocol number to use is
(implicitly) 115.

In tandem with the previous point (session versus tunnel ID's) this really
nails down the fact that you can not have distinct SA's (QOS associations,
or whatever) for different L2TP flows.  My own desire in the original L2TP
was to accomodate such associations this by permitting distinct UDP ports.
A similar desire in L2TPHC would have been to leave open the possibility of
using more than one IP protocol number (some sort of dynamic allocation
scheme, I guess).  Neither of these, in practice, looks like a viable avenue
given the current state of affairs.  If we want distinct associations, we
now require L2TP-specific knowledge in each protocol which wanted to
maintain such associations.  This is because there's no IP-generic way to
make distinct sessions within L2TP visible.

3. 16 bits of Session ID's

The ATM folks have argued (strongly) that they would like to L2TPHC on an
access concentrator in a mandatory tunneling environment.  I fiddled the
text around to accomodate this possibility a bit better--no big deal.

However, there is a need for more than 8 bits of sessions in such an
environment.  So I'm also planning on adding a second bit in the packet
header so the header can specify 0, 1, or 2 bytes of session ID.  This will,
of course, be required to match what was specified in the
L2TPHC-Assigned-Session-ID (previously known as L2TPHC-Assigned-Tunnel-ID,
and--of course--now exchanged during session establishment).

OK.  Break out your flame throwers... I'm ready. :->

BTW, I have departed Cisco, so make sure you direct mail to my current
account--vandys@zendo.com.

Regards,
Andy Valencia


From owner-l2tp@ipsec.org  Wed May 31 22:17:13 2000
Received: from ns1.sailpix.com (ns1.sailpix.com [155.53.1.250])
	by ietf.org (8.9.1a/8.9.1a) with ESMTP id WAA00794
	for <l2tpext-archive@odin.ietf.org>; Wed, 31 May 2000 22:17:12 -0400 (EDT)
Received: from localhost (localhost [[UNIX: localhost]]) by ns1.sailpix.com (8.8.8/8.6.9) id PAA04622; Wed, 31 May 2000 15:35:11 -0700 (PDT)
Received: from franklin.cisco.com (franklin.cisco.com [171.70.156.17]) by ns1.sailpix.com (8.8.8/8.6.9) with ESMTP id PAA04618 for <l2tp@ipsec.org>; Wed, 31 May 2000 15:34:01 -0700 (PDT)
Received: from gwz (ssh.cisco.com [171.69.10.34]) by franklin.cisco.com (8.8.6 (PHNE_17190)/CISCO.SERVER.1.2) with SMTP id RAA02582; Wed, 31 May 2000 17:51:59 -0700 (PDT)
Reply-To: <gwz@cisco.com>
From: "Glen Zorn" <gwz@cisco.com>
To: "Ly Loi" <LLL@shastanets.com>, "'Skip Booth'" <ebooth@cisco.com>
Cc: <aboba@internaut.com>, <l2tp@ipsec.org>, <wdixon@MICROSOFT.com>,
        <townsley@cisco.com>
Subject: RE: Comments/questions to l2tp over ipsec draft
Date: Wed, 31 May 2000 17:35:56 -0700
Message-ID: <NDBBIHMPILAAGDHPCIOPCENGCFAA.gwz@cisco.com>
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 (9.0.2910.0)
Importance: Normal
In-Reply-To: <4DEC824F6F1FD3119BEF0000E86CEA0101E49779@mailserver.shastanets.com>
X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200
Sender: owner-l2tp@ipsec.org
Precedence: bulk
Content-Transfer-Encoding: 7bit

Ly Loi [mailto:LLL@shastanets.com] writes:

> Glen, the LNS is the tunnel terminator; e.g. the security gateway.

An LNS doesn't terminate IPSEC tunnels, though.  The LNS and "Security
Gateway" may reside on the same box, but they are logically distinct, right?

>
> - Ly
>
> > -----Original Message-----
> > From: Glen Zorn [mailto:gwz@cisco.com]
> > Sent: Saturday, May 27, 2000 11:16 AM
> > To: Ly Loi; 'Skip Booth'
> > Cc: aboba@internaut.com; l2tp@ipsec.org; wdixon@MICROSOFT.com;
> > townsley@cisco.com; Glen Zorn
> > Subject: RE: Comments/questions to l2tp over ipsec draft
> >
> >
> > Ly Loi [mailto://LLL@shastanets.com] writes:
> >
> > > Skip, security between the client and the LAC isn't what I was
> > > referring to.
> > > In my scenario, the PPP and the LAC can reside on the box,
> > or different
> > > boxes. However, the LAC doesn't reside on the same box as
> > the IPSEC tunnel
> > > initiator.
> > > To be more specific, what I have in mind is a group of LACs
> > > sharing 1 IPSEC
> > > SA. This IPSEC SA is between the gateway and the LNS.
> > >
> > > LAC1---------gateway------------ipsec_tunnel--------------LNS
> > > LAC2-----------|
> > > LAC3-----------|
> >
> > It appears that you are missing a "security gateway" at the LNS end.
> >
> > >
> > > Unlike the cases covered by the draft where each L2TP tunnel is
> > > mapped to an
> > > IPSEC SA, here, multiple L2TP tunnels map to one IPSEC SA.
> > If the draft
> > > could also cover this case, it'd be more complete.
> >
> > I don't think that the point of the draft is to present an
> > exhaustive list
> > of every method (no  matter how lame) that can be used to
> > "secure" L2TP
> > tunnels.
> >
> > >
> > > A few more comments:
> > >
> > > 1. Could you not require that L2TP start the retransmission timer
> > > only after
> > > the IPSEC SA has come up (section 9.2.2)? I stated the reason in
> > > my original
> > > email. In addition, in the above scenario, this requirement
> > wouldn't be
> > > applicable.
> > >
> > > 2. The draft lists the IPSEC filters as if they represent
> > both the IPSEC
> > > policies and the IPSEC SAs, which is rather confusing since
> > IPSEC SAs are
> > > derived from IPSEC policies. If you could decouple them in
> > your examples,
> > > it'd be great.
> > >
> > > 3. I don't know how others feel about having an external entity
> > > (e.g. L2TP)
> > > injecting some filters into the IPSEC policy database, but
> > this appears to
> > > me to be a local implementation issue. Also, this wouldn't
> > be feasible for
> > > the above scenario. I however appreciate the problems the draft
> > > is trying to
> > > solve.
> > >
> > > Thanks,
> > >
> > > - Ly
> > >
> > >
> > >
> > >
> > >
> > > > -----Original Message-----
> > > > From: Skip Booth [mailto:ebooth@cisco.com]
> > > > Sent: Friday, May 26, 2000 12:36 PM
> > > > To: Ly Loi
> > > > Cc: 'aboba@internaut.com'; l2tp@ipsec.org; gwz@cisco.com;
> > > > wdixon@MICROSOFT.com; townsley@cisco.com
> > > > Subject: RE: Comments/questions to l2tp over ipsec draft
> > > >
> > > >
> > > >
> > > >
> > > > On Fri, 26 May 2000, Ly Loi wrote:
> > > >
> > > > > Skip, I was referring to the case where the LAC and the
> > ipsec tunnel
> > > > > initiator didn't reside on the same device:
> > > > > LAC---gateway(ipsec tunnel
> > initiator)---------------LNS(ipsec tunnel
> > > > > terminator)
> > > > >
> > > > > Is the draft supposed to cover this case too? This case
> > > > would be considered
> > > > > compulsory tunneling, ipsec tunneling :-)
> > > >
> > > > The draft discusses security concerns in the compulsory
> > > > model.  Specifically, it
> > > > talks about how the link between the client and the LAC is
> > > > potentially not
> > > > secure and that clients may require end to end security.  If
> > > > the client is
> > > > running end to end security than potentially no encryption is
> > > > required for L2TP
> > > > data packets between the LAC and LNS, but authentication may
> > > > be.  As you noted
> > > > however, currently there is no way standardized method
> > for IPsec to
> > > > differentiate between L2TP data and control packets.
> > > >
> > > > Does the draft need to go into more detail in this area?  If
> > > > so what more would
> > > > you like to be discussed?
> > > >
> > > > > If L2TP and IPSEC reside on the same device, then yes, we
> > > > all agree IPSEC
> > > > > would run in transport mode and I trust this will be stated
> > > > in a future
> > > > > version of the draft, thanks.
> > > >
> > > > Yes.
> > > >
> > > > -Skip
> > > >
> > > > >
> > > > > - Ly
> > > > >
> > > >
> > > > [rest of text snipped]
> > > >
> > >
> > >
> >
>
>



