Re: [pcp] On the interaction between PCP sever and NAT (or multiple NATs)

"Dan Wing" <dwing@cisco.com> Thu, 14 October 2010 04:07 UTC

Return-Path: <dwing@cisco.com>
X-Original-To: pcp@core3.amsl.com
Delivered-To: pcp@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id EAFF53A68CF for <pcp@core3.amsl.com>; Wed, 13 Oct 2010 21:07:36 -0700 (PDT)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -109.519
X-Spam-Level:
X-Spam-Status: No, score=-109.519 tagged_above=-999 required=5 tests=[AWL=-0.720, BAYES_00=-2.599, J_CHICKENPOX_23=0.6, J_CHICKENPOX_33=0.6, J_CHICKENPOX_42=0.6, RCVD_IN_DNSWL_HI=-8, USER_IN_WHITELIST=-100]
Received: from mail.ietf.org ([64.170.98.32]) by localhost (core3.amsl.com [127.0.0.1]) (amavisd-new, port 10024) with ESMTP id HfK6iC3VDr1U for <pcp@core3.amsl.com>; Wed, 13 Oct 2010 21:07:33 -0700 (PDT)
Received: from sj-iport-4.cisco.com (sj-iport-4.cisco.com [171.68.10.86]) by core3.amsl.com (Postfix) with ESMTP id 34F083A68BB for <pcp@ietf.org>; Wed, 13 Oct 2010 21:07:32 -0700 (PDT)
Authentication-Results: sj-iport-4.cisco.com; dkim=neutral (message not signed) header.i=none
X-IronPort-Anti-Spam-Filtered: true
X-IronPort-Anti-Spam-Result: Av0EADoftkyrR7Ht/2dsb2JhbACUT4xUcaFanFqFSASEUlU
X-IronPort-AV: E=Sophos;i="4.57,328,1283731200"; d="scan'208";a="200486514"
Received: from sj-core-1.cisco.com ([171.71.177.237]) by sj-iport-4.cisco.com with ESMTP; 14 Oct 2010 04:08:50 +0000
Received: from dwingWS ([10.32.240.198]) by sj-core-1.cisco.com (8.13.8/8.14.3) with ESMTP id o9E48ocs016541; Thu, 14 Oct 2010 04:08:50 GMT
From: Dan Wing <dwing@cisco.com>
To: songlinjian@163.com, 'pcp' <pcp@ietf.org>
References: <4CB67A53.78DC56.21897@m50-135.163.com>
In-Reply-To: <4CB67A53.78DC56.21897@m50-135.163.com>
Date: Wed, 13 Oct 2010 21:08:49 -0700
Message-ID: <1e1a01cb6b55$81695ff0$843c1fd0$@com>
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: 7bit
X-Mailer: Microsoft Office Outlook 12.0
Thread-Index: ActrUbAHF3p6BCQ5SDuYYRA/4iYlewAAyYHw
Content-Language: en-us
Subject: Re: [pcp] On the interaction between PCP sever and NAT (or multiple NATs)
X-BeenThere: pcp@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: PCP wg discussion list <pcp.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/pcp>, <mailto:pcp-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/pcp>
List-Post: <mailto:pcp@ietf.org>
List-Help: <mailto:pcp-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/pcp>, <mailto:pcp-request@ietf.org?subject=subscribe>
X-List-Received-Date: Thu, 14 Oct 2010 04:07:37 -0000

> -----Original Message-----
> From: pcp-bounces@ietf.org [mailto:pcp-bounces@ietf.org] On Behalf Of
> ???
> Sent: Wednesday, October 13, 2010 8:37 PM
> To: pcp
> Subject: [pcp] On the interaction between PCP sever and NAT (or
> multiple NATs)
> 
> Hi,ALL
> 
> 	 In "draft-wing-softwire-port-control-protocol-02" the Fig.2
> shows that the NAT devices can work with separate PCP Server. But I
> only find the detailed specification between the PCP server and
> clients. The interation between PCP server and NAT is obscure and the
> suggested xml-style communication is not clear. How dose it works?
> Based on TCP,UDP or other mature transmission technology? what
> information should it contain ?

Undefined.  The PCP server and NAT could be implemented in the same
device (which I expect will be common), and share whatever 
data structures are necessary between them using whatever they
find natural (perhaps a socket-based API, perhaps shared memory,
etc.).  If PCP server and NAT are implemented in separate devices,
it would be a proprietary protocol between the PCP server and
the NAT(s).

> Below is some question and considerations on the interation between PCP
> sever and NAT. Comments and suggestions are welcome and appreciated.
> 
> 1.Is the interation between PCP server and NAT private or public ?  In
> another word,is it possible that PCP server and NAT are producted by
> differnt network equipment vendors. IF we expect two devices from
> different vendors can work together without trouble, the specification
> of the interation is necessary.

It's a private interface.

-d

> 2. In the PCP draft, NAT device only receives address binding
> information from PCP server. Is there any message that NAT should tell
> PCP server? Intuitively, PCP server should detect the failure of NAT
> device, expecially in multiple NATs scenario. Tranditional  echo or
> 'hello' protocol is available. on the purpose redundancy and load-
> balancing the NAT group id and the current load is expected to be known
> by PCP server. (plz see 3)
>
> 3. Can a single PCP server interact with multiple NATs? Because the
> reliability is very important for CGN devices, mulitple NATs is highly
> desireable for both load balacing and redundancy. If the multiple NAT
> scenario makes sense in PCP context, the specification of the
> interation should be more carefully designed.
> plz see http://tools.ietf.org/html/draft-xu-behave-stateful-nat-
> standby-04 and http://tools.ietf.org/html/draft-wang-behave-nat64-load-
> balancer-02 for more information
>
> 4.TCP or UDP? I'm not sure about that part. If reliable connection is
> need between PCP server and NAT ,TCP is prefered.
>
> 
> 
> 
> 
> 
> Best regards.   2010-10-14
> 
> ********************************************************
> *    Linjian Song ,Ph.D candidate                      *
> *    Department of Computer Science & Technology       *
> *    Tsinghua University, Beijing, P.R.China(100084)   *
> *    Tel: (8610)-62795818-6864                         *
> *    Email: songlinjian@csnet1.cs.tsinghua.edu.cn      *
> ********************************************************
> 
> 
> _______________________________________________
> pcp mailing list
> pcp@ietf.org
> https://www.ietf.org/mailman/listinfo/pcp