Re: [renum] IPv4 deployment scenarios to minimise renumbering costs

Brian E Carpenter <brian.e.carpenter@gmail.com> Fri, 03 December 2010 22:10 UTC

Return-Path: <brian.e.carpenter@gmail.com>
X-Original-To: renum@core3.amsl.com
Delivered-To: renum@core3.amsl.com
Received: from localhost (localhost [127.0.0.1]) by core3.amsl.com (Postfix) with ESMTP id 283633A67AB for <renum@core3.amsl.com>; Fri, 3 Dec 2010 14:10:21 -0800 (PST)
X-Virus-Scanned: amavisd-new at amsl.com
X-Spam-Flag: NO
X-Spam-Score: -103.224
X-Spam-Level:
X-Spam-Status: No, score=-103.224 tagged_above=-999 required=5 tests=[AWL=0.375, BAYES_00=-2.599, RCVD_IN_DNSWL_LOW=-1, 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 KTYEiPxVfAwx for <renum@core3.amsl.com>; Fri, 3 Dec 2010 14:10:20 -0800 (PST)
Received: from mail-yw0-f44.google.com (mail-yw0-f44.google.com [209.85.213.44]) by core3.amsl.com (Postfix) with ESMTP id D6DFF3A659C for <renum@ietf.org>; Fri, 3 Dec 2010 14:10:19 -0800 (PST)
Received: by ywi6 with SMTP id 6so959908ywi.31 for <renum@ietf.org>; Fri, 03 Dec 2010 14:11:37 -0800 (PST)
DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=gmail.com; s=gamma; h=domainkey-signature:received:received:message-id:date:from :organization:user-agent:mime-version:to:cc:subject:references :in-reply-to:content-type:content-transfer-encoding; bh=y8fsMXXcjbdRwg85glg47D9rvXGUuD7DzwoBGptzMTg=; b=gdxgFJleI/WsUQxlvMCVjoJicN8l+WyTrLugGQF74LWcLh9fNYFN2V8oIflsaGpxUF F/IqUa+4Mh7w2QrMdQUzMNN6O4aM8KRzpVk4rXPanEv6HnxQR73hzhhUr+CYN0Mqwp5f Yox++shVT1jWVA2Rlw8FijPkNSXwKahCOMiTw=
DomainKey-Signature: a=rsa-sha1; c=nofws; d=gmail.com; s=gamma; h=message-id:date:from:organization:user-agent:mime-version:to:cc :subject:references:in-reply-to:content-type :content-transfer-encoding; b=Xo5jbpBRy/oMjHU9+/hPzJj9wq0V4BewRnh0RwMKupQPwMUaqdlpJ3QLFgtghCkLIN 6n8SmaarFEGzTJcoLmHrOfsEmsi0OrtMbiIXAsV3Gik6b7D0Op3l4QPmWXNLS8VpGO0B ConkgZPpxpGOg/6iflzqOSP0zPMp0GwZq7coc=
Received: by 10.220.171.9 with SMTP id f9mr520786vcz.275.1291414296336; Fri, 03 Dec 2010 14:11:36 -0800 (PST)
Received: from [10.1.1.4] ([121.98.190.33]) by mx.google.com with ESMTPS id w30sm428506vcr.16.2010.12.03.14.11.34 (version=SSLv3 cipher=RC4-MD5); Fri, 03 Dec 2010 14:11:35 -0800 (PST)
Message-ID: <4CF96B11.8020405@gmail.com>
Date: Sat, 04 Dec 2010 11:11:29 +1300
From: Brian E Carpenter <brian.e.carpenter@gmail.com>
Organization: University of Auckland
User-Agent: Thunderbird 2.0.0.6 (Windows/20070728)
MIME-Version: 1.0
To: RJ Atkinson <rja.lists@gmail.com>
References: <EBE6787B-849F-49A1-A467-84C18F52FE10@gmail.com>
In-Reply-To: <EBE6787B-849F-49A1-A467-84C18F52FE10@gmail.com>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: 7bit
Cc: renum@ietf.org
Subject: Re: [renum] IPv4 deployment scenarios to minimise renumbering costs
X-BeenThere: renum@ietf.org
X-Mailman-Version: 2.1.9
Precedence: list
List-Id: "Renumbering discussion mailing list." <renum.ietf.org>
List-Unsubscribe: <https://www.ietf.org/mailman/listinfo/renum>, <mailto:renum-request@ietf.org?subject=unsubscribe>
List-Archive: <http://www.ietf.org/mail-archive/web/renum>
List-Post: <mailto:renum@ietf.org>
List-Help: <mailto:renum-request@ietf.org?subject=help>
List-Subscribe: <https://www.ietf.org/mailman/listinfo/renum>, <mailto:renum-request@ietf.org?subject=subscribe>
X-List-Received-Date: Fri, 03 Dec 2010 22:10:21 -0000

On 2010-12-04 03:45, RJ Atkinson wrote:
> 
> There is at least one IPv4 deployment scenario that
> is pretty common right now in several areas of the 
> world, works pretty well for enterprise users,
> provides practical amounts of provider-independence,
> and probably would be worth documenting.
> 
> 
> 
> A previous employer of mine *by choice* (specifically, 
> due to internal security policy choices) decided to put 
> nearly the whole organisation into private address space.
> The exception was that public-facing content servers
> were in global-scope address space, on a network segment
> outside the "trusted zone" of the organisation.  By
> policy, access initiated from outside the trusted
> zone to any node inside the trusted zone was prohibited.
> So it was actively undesirable to have interior nodes
> with global-scope address space.  I believe these are
> not unusual choices for an enterprise to make these days.

Sad but true. I believe that draft-mrw-nat66 is aimed at
this model, and draft-troan-multihoming-without-nat66 is
aimed against it. We could consider documenting this approach,
but of course it doesn't affect the subnet-reorganization
form of renumbering (i.e. renumbering regardless of the
"address independence" property).

    Brian

> 
> While they easily could have obtained sufficient IPv4
> address space (and likely still could today) from
> their upstream providers, they preferred to deploy
> using private addressing with NAT-capable IPsec-capable
> firewalls at the edge(s) of each site within the
> organisation.
> 
> Each site within the organisation had its own upstream
> connectivity -- managed/selected/paid by centralised
> IT services folks.  This let the IT folks reduce cost
> by selecting lower-cost providers on a site-specific
> basis.
> 
> Connectivity between sites of the organisation was
> enabled using IPsec encrypted tunnels between those
> same edge firewall devices.
> 
> This had the effect of offering provider-independence,
> because only the (1 or 2) Firewall devices at a given site 
> needed to renumber their exterior network interfaces 
> if that site's upstream provider set changed.  
> 
> Crucially, by using MBOX-style identity types 
> (e.g. site-11@organisation.foo) in the IKE configurations, 
> there was no need to reconfigure firewalls or nodes 
> at other sites if a given site's upstream connectivity
> changed.
> 
> For M&A, an acquired organisation's sites would be
> renumbered into a suitable chunk of private address
> space and migrated into the above schema.  Because
> most organisations now use DHCP, the amount of 
> manual configuration involved in site renumbering
> was minimised and turned out to be quite tractable
> to undertake manually.
> 
> 
> This doesn't help everyone and isn't completely general, 
> but it does apply to a very common type of user organisation.
> It could be very helpful to document the above for the
> benefit of those organisations who choose to use it.
> 
> I suspect there are other specific items that would be
> useful to folks with IPv4 deployments.  So I think that
> IPv4 efforts ought to remain within-scope, even if IPv6
> were perceived as the "sweet spot" for improvements.
> 
> Yours,
> 
> Ran
> 
> 
> _______________________________________________
> renum mailing list
> renum@ietf.org
> https://www.ietf.org/mailman/listinfo/renum
>