http://www.potaroo.net/ A weblog of Internet material by Geoff Huston. en-au Copyright 1989-2008 Geoff Huston Tue, 23 Jun 2009 23:50:00 +1000 http://feedvalidator.org/docs/rss2.html 130 http://www.potaroo.net/ispcol/2009-06/twenty-years.html http://www.potaroo.net/ispcol/2009-06/twenty-years.html As I write this, on the 23rd June 2009, I've been reminded that some 20 years ago, on the night of the 23rd June 1989, Robert Elz of the University of Melbourne and Torben Neilsen of the University of Hawaii completed the connection work that bought the Internet to Australia with a permanent connection via a 56kbps satellite circuit. Since that day we've evidently connected some 56.8% of the local population, or 12,073,852 Australians, to the Internet (according to recent Internet user statistics published by the ITU-T). While that~s an impressive outcome, I suppose I should say at the outset that when we started down this path in Australia some twenty years ago we had no intention of achieving this scale of outcome. Indeed we never thought that this type of data networking would ever cross the boundary from an esoteric tool to assist a select group of computer literate researchers and academics into the mainstream of society, and current concepts like twitter and social networking were completely foreign to us. In truth all we were trying to do was to save a bit of money for the universities and have some fun experimenting with some pretty novel technology on the way. At best this was all just an experiment and if there was ever going to be some mainstream commercial outcome, then that was the for the telephone companies to work on, and was certainly none of our business. So how and why did all this get started? Tue, 23 Jun 2009 23:50:00 +1000 http://www.potaroo.net/ispcol/2009-04/sharing.html http://www.potaroo.net/ispcol/2009-04/sharing.html It looks like we will need to operate a dual-stack Internet for many years yet, and the associated implication is that that we are going to have to make the existing IPv4 Internet span a billion or more new deployed services, and do so without any additional address space. So how are we going to make the IPv4 address pool stretch across an ever-larger Internet? Given that the tool chest we have today is the only one available, then it looks like there is only one answer to this question: Network Address Translation, or NATs. Mon, 18 May 2009 18:50:00 +1000 http://www.potaroo.net/ispcol/2009-02/mtu.html http://www.potaroo.net/ispcol/2009-02/mtu.html When looking at network performance the conventional belief is that larger parkets are far better than smaller packets. So whenever and however possible you should raise your packet size, or MTU, as larger as possible. "Conventional beliefs" always intrigue me, in that while sometimes these beliefs express basic constraints and truths, at other times they are misleading and just plain wrong. So the question I'd like to look at in this article is: Just how bad is a smaller MTU setting? Sun, 15 Feb 2009 20:15:00 +1100 http://www.potaroo.net/ispcol/2009-01/mtu6.html http://www.potaroo.net/ispcol/2009-01/mtu6.html I have seen a number of commentaries and presentations in recent times that claim that IPv6 is identical to IPv4 in every respect except one: namely more addresses. But there is one more rather critical difference, and that is the deliberate change in the IPv6 with respect to MTU handling and packet fragmentation, and this relatively minor change in IPv6 has some really quite critical implications. In this article I'd like to illustrate some of the implications of this change with respect to the IPv6 treatment of packet fragmentation by taking an in-depth look at the IPv6 packet flows and why and how this change to packet fragmentation management can cause service-level disruption. Sat, 20 Dec 2008 07:25:00 +1100 http://www.potaroo.net/ispcol/2008-12/resourcecertificates.html http://www.potaroo.net/ispcol/2008-12/resourcecertificates.html In November 2008 APNIC publically released its resource certification toolkit. This system allows a holder of APNIC-allocated IP number resources (IP addresses and AS numbers) to have these resources certified by APNIC as being currently held by the entity through the issuance and publication of a digital resource certificate. In this article I will look at resource certificates in some detail, looking at the technology that underpins certification structures, and the potential use of these certificates in securing inter-domain routing and potentially in the area of the emerging need to support address transfers in IPv4. Sat, 29 Nov 2008 14:55:00 +1100 http://www.potaroo.net/ispcol/2008-11/transfers.html http://www.potaroo.net/ispcol/2008-11/transfers.html The RIRs' policy forums are now considering what the appropriate registry policies should be when the allocation function for IPv4 addresses has terminated. Attempting to apply the same policy principles that were used to guide the address allocation function to the registry function is causing some level of confusion. Here previous practice becomes confused with the generic principles, and the differing characteristics of allocation and registry functions are not being clearly distinguished. Mon, 20 Oct 2008 12:00:00 +1000 http://www.potaroo.net/ispcol/2008-10/v4depletion.html http://www.potaroo.net/ispcol/2008-10/v4depletion.html It should be entirely unsurprising that the next phase of the Internet's story, that of the transition of the underlying version of the IP protocol from IPv4 to IPv6, refuses to follows the intended script. Where we are now, in mid-2008, with IPv4 unallocated address pool exhaustion looming within the next 18 to 36 months, and IPv6 still largely not deployed in the public Internet, is a situation that was entirely uncontemplated and, even in hindsight, entirely surprising to encounter. Mon, 15 Sep 2008 17:00:00 +1000 http://www.potaroo.net/ispcol/2008-09/ietf72-ipv6trans.html http://www.potaroo.net/ispcol/2008-09/ietf72-ipv6trans.html The IETF's developmental work on IPv6 has included the study of the particular issues associated with transition to IPv6 from the outset. Starting with the TACIT work on transition in 1994-95, then NGTRANS from 1995 to 2002, and, more recently, the V6OPS Working Group which met first in early 2003 at IETF56. This study has now broadened in scope and today a number of IETF working groups are examining aspects of transition to IPv6 including the SOFTWIRE, BEHAVE and INTAREA Working Groups, in addition to V6OPS. In this article I'd like to look at the background to this IETF activity and review the current state of the IETF Working Groups that are looking at transition tools. Sat, 19 Jul 2008 21:00:00 +1000 http://www.potaroo.net/ispcol/2008-08/ipv6addr.html http://www.potaroo.net/ispcol/2008-08/ipv6addr.html If you have enabled IPv6 on your computer, and in an idle moment you've browsed through the interface configuration information for IPv6 addresses you may have been a little surprised by the fact that there's not just one IPv6 address that's been loaded, but many. With IPv4 there was a single address that was bound to each interface, but when using IPv6 its not so clear, and an interface can have a number of IPv6 addresses simultaneously. Its also common to have automatic IPv6 over IPv4 tunnelling interfaces be created, and they also are configured with IPv6 addresses. The result can be impressive in terms of the number of IPv6 addresses that are configured into a single host system. Which IPv6 addresses are useable, and in which context? In this article I'd like to look at the IPv6 address plan, and describe the various address prefixes. Sat, 19 Jul 2008 21:00:00 +1000 http://www.potaroo.net/ispcol/2008-07/foi.html http://www.potaroo.net/ispcol/2008-07/foi.html Lets face it, gathering a collection of ministerial delegations to laboriously recite prepared speeches to each other sounds about as exciting as watching paint dry. And observing meetings where the major outcome appears to be limited to the scheduling of the next meeting can become somewhat tedious after a while. It should not be surprising that the level of expectation of tangible outcomes for such governmental meetings is invariably abysmally low. So what's the value of adding yet another meeting to governments' schedule? What makes the OECD-hosted ministerial meeting on the Future of the Internet Economy so unique in the context of the Internet's current political landscape and its political future? Why would a meeting about the dismal science of economics hold any interest at all? Wed, 25 Jun 2008 05:00:00 +1000 http://www.potaroo.net/ispcol/2008-06/10years.html http://www.potaroo.net/ispcol/2008-06/10years.html In 1998 any lingering doubts about the ultimate success of the Internet were dispelled. There was nothing else left standing in the data communications landscape that could serve our emerging needs for data communications. IP was now the communications technology of the day, if not the coming century, and the industry message of the time was to adopt the Internet or imperil your entire future in this business. By 1998 the job was apparently done, and the Internet had prevailed. But the story was not over. Communications continued to drive our world, and the Internet continued to evolve and change. What has happened in the last decade of the Internet? What aspects of internet technology has changed, and why? Fri, 30 May 2008 12:00:00 +1000 http://www.potaroo.net/ispcol/2008-05/eoe2e.html http://www.potaroo.net/ispcol/2008-05/eoe2e.html The model of a clear and simple Internet where end hosts can simply send packets across a transparent network is largely an historical notion. These days we sit behind a dazzling array of so-called "middleware", including Network Address Translators, Firewalls, Web caches, DNS interceptors, TCP performance shapers, and load balancers, to name but a few. So the question is: Have we gone past the end-to-end argument? Are we heading back to a world of bewilderingly complex and expensive networks? Thu, 24 April 2008 12:00:00 +1100 http://www.potaroo.net/ispcol/2008-04/ipv6.html http://www.potaroo.net/ispcol/2008-04/ipv6.html In this article we'd like to look at some measures of the use of IPv4 and IPv6 protocols in today's Internet and see if we can draw any conclusions about just how far down the track we are with the IPv6 part of dual stack deployment. We'll use a number of measurements that have been made consistently since 1 January 2004 to the present, where we can distinguish between the relative levels of IPv4 and IPv6 use in various ways. Mon, 31 March 2008 21:30:00 +1100 http://www.potaroo.net/ispcol/2008-03/routehack.html http://www.potaroo.net/ispcol/2008-03/routehack.html I suppose it had to happen one of these days. Sooner or later a routing hijack would get ts 15 seconds of fame in the industry press, and the incident relating to the YouTube prefix just happened to be the one that was selected by the media because of the players involved rather than the rather mundane characteristics of the routing leak itself. Tue, 4 March 2008 11:00:00 +1100