February 2010 - ipv6-wg - mailman.ripe.net

IPv6 Act Now: News Update
by Chris Buckridge 11 Feb '10

11 Feb '10

Hi all, A couple of news items from the IPv6 Act Now newsfeed over the last week that might be of interest: First accreditation issued for testing to US IPv6 product specs (9 Feb 2010) Amid all the discussion of government leading the way on IPv6 through procurement, news from the US that the University of New Hampshire's test lab is among the first wave of labs to be accredited to test networking products for compliance with government IPv6 specifications. <http://www.ipv6actnow.org/2010/02/ipv6-and-its-future-prospect-in-developin… > AfriNIC embraces Internet challenges in Africa (8 Feb 2010) An extensive interview with AfriNIC CEO Adiel Akplogan, discussing the challenges facing the region's Internet community and some of the new roles AfriNIC is taking on, including working with law enforcement agencies and participating in the Africa Union heads of state summit. <http://www.ipv6actnow.org/2010/02/afrinic-embraces-internet-challenges-in-a… > Pakistan needs to shift to Internet Protocol Version 6 (9 Feb 2010) IPv6 makes it into the Pakistani press, with the IPv6 Monitory Group of Pakistan Telecommunication Authority (PTA) making a number of recommendations for the widespread adoption of IPv6 in the country (including establishing a National Internet Registry). <http://www.ipv6actnow.org/2010/02/pakistan-needs-to-shift-to-internet-proto… > Regards, Chris Buckridge, RIPE NCC

1 0

Request for approval of the minutes for wg session RIPE 59 (20100207)
by David Kessens 10 Feb '10

10 Feb '10

Please find below the minutes from the IPv6 WG session at RIPE 59. The minutes will be declared final if no objections, except for minor editorial corrections that will be fixed before publication, will be received by the end of Sunday February 7, 2010 in a timezone of your choice. Finally, I would like to thank Alex Band for the excellent minutes! Thanks, David Kessens --- RIPE Meeting: 59 Working Group: IPv6 Chair: David Kessens Date: Tuesday October 6, 2009, 16:00-18:00 David welcomed the attendees and opened the session. A. Administrative Matters o Welcome o Select scribe: Alex Band o Select Jabber Monitor: Rumy Kanis o Finalize agenda o Approval of minutes from previous working group meeting B. RIPE NCC IPv6 Update James Aldridge, RIPE NCC http://www.ripe.net/ripe/meetings/ripe-59/presentations/aldridge-ncc-update… There were no questions. C. Future of the Working Group: Charter Discussion Shane Kerr, David Kessens. Input from the audience http://www.ripe.net/ripe/meetings/ripe-59/presentations/kessens-v6-wg-futur… Shane mentioned that the initial idea for an IPv6 Working Group came from looking at the statistics Geoff Huston publishes. "Make IPv6 Happen", was the summary of the charter of the WG. Most of the discussion on IPv6 is actually in the Address Policy WG, so he proposed to shut down the WG and move on to a bigger and better thing. David asked the audience for input. Olaf Kolkman (NLNetlabs) asked how the activities of the WG tied into the IPv6ActNow website. Shane said that there is not really any tie in, and that he thought that is was a shame that something like IPv6ActNow is necessary at all. Rob Blokzijl (RIPE Chair) mentioned that the message that is coming out of the WG is heard by the community at large, however, it doesn't seem to be received and digested by the industry on a wide scale. He proposed that the WG is not shut down, because there is important work to be done, especially in the coming two years and when IPv4 has run out. Gert Doering (Spacenet) commented that the WG is not the place where people go nowadays to learn about IPv6. At the RIPE Meeting, those kinds of discussions are held in the Plenary. Rob gave an example: there were two presentations about the general implementations of IPv6 at customer premises. In one case, the CPE worked fine, in the other it didn't. He would have liked to have seen a technical presentation in the IPv6 WG session on why that is. Shane responded there are definitely opportunities for the WG in that capacity, and even other topics related to IPv6, like Multi-layer NAT. Denesh Babutha (Aexiomus//Cyberstrider) agreed that the WG had reached the end of its useful life as originally envisaged. He said that, as such, the WG needs to evolve, and he thought that this needed to happen through the the charter. Denesh added that the WG needed to move away from repeating the IPv6 situation in every single meeting and move onto more useful discussions. The new charter should be inclusive of other industry areas too. David said that there were many topics covered by the WG that were not originally in the charter. Bernard Tuy said that there is an enormous need for technical training and that kind of knowledge can come from the WG. David asked the audience how to move forward with the WG and whether it should continue in it's current form? Ruediger Volk (Deutsche Telekom) commented that the current form seems to be "quasi-plenary". Rob commented that the WG should be modest in its aims and not try to reach the end goal without some intermediate steps. He said that the WG should focus on real world implementations. Shane commented that the WG did not need to come up with a new charter here and now. Rob proposed that a new draft charter is written. David summarized the feedback for a new charter as more focus on education and outreach activities. Maarten Botterman said that not all of those activities have to be within the WG. David asked the audience if preparing for a hybrid IPv4/IPv6 world should be part of the charter. Rob commented that outreach should lie with the RIPE NCC efforts, such as the IPv6ActNow initiative. He said that training is a sensitive issue because the RIPE NCC does not want to compete with its members. Rumy Kanis (RIPE NCC Training Services Manager) commented that the RIPE NCC offers IPv6 training, but careful steps were taken to ensure the RIPE NCC does not compete with the membership. Marco Wertejuk (Binconsult) said that he believes the focus for the new charter should be "how to encourage conservation of IPv4 and deployment of IPv6". David said that he assessed from the audience feedback that there isn't much support for the WG to focus on non-technical outreach activities. David asked Rob for a time line. Rob said that the WG should take things slowly, and propose a new charter at the next RIPE Meeting and that further discussions should continue on the mailing list. D. Future RIPE Network Experiments - David Kessens. Input from the audience David asked the audience if there was further interest in IPv6 network experiments during a RIPE Meeting. Gert said that forcing IPv6 on people is actually a good idea. He said we should even go as far as making the RIPE Meeting network IPv6 only for the whole week. He added that we should do the experiment without NAT-PT. David said that he would like to see a completely new experiment. Marco Hogewoning (XS4all) said that although IPv6 connectivity in general is good, tunneling is needed in some places in order to get full connectivity. David Wilson (HEAnet) said that the IPv6 hour should be repeated, and that the WG should get new experiences because things have changed since the last time it was done. Izumi from JPNIC that for their experiments, they created different testbeds for different kinds of implementations. David asked the audience if the IPv6 Hour should be rerun at future meetings.. There is consensus that it should be. E. IPv6 Deployment: What are the Remaining Issues and Bottlenecks? - A panel discussion with interactive input from the audience. Moderator: Maarten Botterman, GNKS Consult http://www.ripe.net/ripe/meetings/ripe-59/presentations/botterman-towards-v… Gert Doering introduced himself and asked the question "What kind of Internet do we want to have in five years?". Geoff Huston introduced himself and in a slide set proposed the question "Is the IPv6 transition an example of market failure?" http://www.ripe.net/ripe/meetings/ripe-59/presentations/huston-ipv6-transit… Kurtis Lindqvist introduced himself and explained his own experiences with IPv6 deployment. He said that doing a slow and gradual implementation has been very beneficial. Maarten Botterman did a live survey, with the question "What would you gain by postponing, and what would you gain by stepping up to the plate, earlier rather than later?". 40 votes were cast, most go to: o By stepping up to the plate I have time and space to do the transition, which will allow me to do it gradually and well thought through o By stepping up to the plate I confirm my brand's image as 'state-of-the-art' or 'top-of-the-wave' Geoff commented that the respondents are most likely non-representative. Gert said that he is not surprised with the numbers. Kurtis said that looking at the responses, he is optimistic. Maarten said that some people responded with 'none of the above'. He asked three of those people to comment: Ruediger said that for all of the relevant players there is still enough time for doing a proper job. Bernard said that there was no vendor support, so ISPs will never start until that happens. Jan Hugo Prins said that there will never be end user demand. Deployment is not picking up because of security issues. Geoff reiterated that he thought that the situation we are finding ourselves in right now is no accident. It is in many companies' interests to delay IPv6 adoption. Shane agreed with Geoff. Gabriella Paolini (GARR) said that the biggest issue is that middleware doesn't support IPv6 and that changes very slowly. Maarten did another live survey with the following question: "There is perception and experience of/with a number of hurdles that seem to be difficult to overcome. Which are important to you?". o As there is very little customer demand, I cannot justify the investment to my management, even if I would like to move o I have the experience that my vendors cannot deliver what I need o None of the above (highest score) Ruediger said that the options are very extreme. Slow and gradual deployment is a real option. Geoff commented that in his days at a telco, the biggest cost turned out to be customer support. IPv6 could become expensive in that respect with end users calling in because of issues. Maarten asked people to comment: - Wilfried Woeber (Vienna University) commented that he would have to select 'none of the above' for all questions. His issue is that the questions are too black or white. The biggest problem from his point of view is in layer 7 or 8: infrastructure, applications etc. The backbone has been done for years. - Marco said that he agreed with Geoff, the cost is not in the hardware. IPv6 is not a product, don't pay extra for it. - An attendee said that in a lot of cases, even though the product spec sheet says that IPv6 is supported, there is no feature parity with IPv4. Gert said he agreed with most of the comments. Geoff said that any vendor is willing to give you what you want, as long as you put down the money. Kurtis added that there was also a big push from vendors for MPLS. Marco added that some IPv6 functionality used to be free, but after vendors realized that there was a market for it, they started charging for a license. Maarten did another live survey, the results with the most votes were: o The deployment of IPv6 is unavoidable for my organization in due time o Technically, everything is in place to deploy IPv6 Geoff commented that openness is the key in the end. IPv6 is the only way to guarantee that. Kurtis added that he was surprised that so many people responded "I will deploy in due time", because that is pretty close. Geoff re-stated that the market is well informed, and that this situation is intentional in order to maximize shareholder profits at large corporations. Ruediger agreed that the industry is flooded with information. But he asked how enlightened are people? Geoff responded that it's all about vendor lock in. Brian Nisbet (HEAnet) said that the WG keeps having the same discussion meeting after meeting. He said that he could not believe that people are not well informed. Kurtis said that CFO's cannot be convinced of the urgency of something that will happen two years from now. An attendee said that for mobile operators, the vendors of mobile terminals are the blockers. Maarten showed the results of the last survey. The panel shared some closing thoughts. The general thought was that the community should stop distributing messaging, and start moving packets. Maarten closed the discussion. F. Developments/Initiatives Regarding IPv6 in the RIPE Region and Beyond - Input from the audience No input was given. Y. Input for the RIPE NCC Activity Plan - Input from the audience No input was given. Z. AOB None. David closed the session. ---------

2 5

Last Call: Confirmation of co-chairs (20100217)
by David Kessens 10 Feb '10

10 Feb '10

The discussion phase regarding new co-chair positions has now ended. It seems that there is strong support to appoint: Shane Kerr http://www.ripe.net/ripe/maillists/archives/ipv6-wg/2010/msg00038.html Marco Hogewoning http://www.ripe.net/ripe/maillists/archives/ipv6-wg/2010/msg00041.html as additional co-chairs for the ipv6 working group. Please let us know if you have objections to such a solution by the end of Wed Feb 17, 2010 in a time zone of your choice. There is no need to restate your supportive position again as there was already a significant number of people that expressed support for this approach. Thanks, David Kessens ---

1 0

announcing DNS64-NAT64 opensource implementations
by Marc Blanchet 08 Feb '10

08 Feb '10

<sorry for cross-posting/> This is to announce the availability of two NAT64-DNS64 open-source implementations by Viagenie, as follows: ====== NAT64 ====== BSD PF: this implementation of NAT64 is made by modifying the PF packet filter available on BSD systems. A new 'nat64' statement is created in the pf.conf file to enable the nat64 functionality. Linux Netfilter: The implementation of NAT64 for linux is available as a kernel module. It uses Netfilter's facilities for packet interception. The configuration is done with parameters provided at module insertion. ====== DNS64 ====== As announced in july 2009 during IETF Stockholm, the companion DNS64 functionality is also available in two implementations, as follows: BIND: this implementation of DNS64 is made by modifying the BIND DNS server. A new "dns64-prefix" statement in options is created in the named.conf file to enable DNS64 functionality. Unbound: this implementation of DNS64 is made by adding a module to the Unbound DNS server. To enable the DNS64 functionality, the module should be declared and the dns64-prefix statement should be added in the unbound.conf file. The two functionalities (NAT64 and DNS64) make a complete system for translating IPv6-IPv4 packets. The source code, some pre-compiled packages and project description are available at: http://ecdysis.viagenie.ca. This project was funded by NLNet foundation and Viagenie. We are looking for feedback, patches, suggestions from the community. http://ecdysis.viagenie.ca Regards, Marc Blanchet, Simon Perreault, Jean-Philippe Dionne Viagenie

1 0

IPv6 experiments at future RIPE Meetings
by James Aldridge 08 Feb '10

08 Feb '10

Reading the minutes of the IPv6 WG at RIPE 59 I read: > David asked the audience if the IPv6 Hour should be rerun at future > meetings.. There is consensus that it should be. Rob Blokzijl has asked that we treat the RIPE Meeting network as a production network. This effectively means that we won't be turning off the dual-stack network for another "IPv6 Hour". What we can do is to build the IPv6 only networks as we did in Berlin and make these available to anyone who wants to use them, but without the complications caused by the IPv6 Hour. One question that remains is whether, with rfc2766 now "historic", providing NAT-PT on the IPv6-only networks is worthwhile for RIPE 60 and future meetings. Would it be enough just to provide the IPv6 networks but without any translation to be able to reach the IPv4 legacy Internet? Any feedback would be appreciated. Regards, James -- James Aldridge, Senior Systems & Network Engineer, RIPE NCC RIPE Meeting technical team

17 24

Re: [ipv6-wg] Discussion period for co-chair selection until 20100208
by Ralph Smit 08 Feb '10

08 Feb '10

Marco Hogewoning wrote: > On 3 feb 2010, at 23:44, David Kessens wrote: > >> Basically, knowing our candidates, I would like to hear from you >> whether you prefer to appoint one or two co-chairs, and whether you >> have any preference for either candidate in case you believe one >> co-chair is enough. > > > Hi All, > > I had a short chat with Shane this afternoon and we would like you to > know we as candidated would prefer the option to appoint two co-chairs. > Shane and I have known eachother for quite some time and our goals and > objectives for the future of the IPv6 working group are the same, we > both are looking forward working with David and the community to live up > to the new charter en make IPv6 a succes. > > Thanks for all the support so far, > > MarcoH and Shane > > > I support both candidates and agree with many who replied before me (and the candidates themselves) that 2 co-chairs would be a good idea. Regards, Ralph Smit.

2 1

Re: RFC 1918 in "production networks"
by Geert Jan de Groot 06 Feb '10

06 Feb '10

> If we use 192.168.1.0/24 for the RIPE meeting, and your corporate network > at home uses 192.168.1.0/24 as well, most VPN clients will NOT be able > to build a working connection. Try it. > Given the number of attendees, the chances for address collisions with > at least one participant's home network can be assumed to be near 100%. One workaround would be to have two (vastly different) 1918 prefixes and giving people the possibility to switch (so, 2 SSID's, 2 prefixes, say 192.168.224.0/20 and 10.210.32.0/20). Chances that someone will conflict with both, are small. For better or worse, we will need to learn how to deal with this - denying these issues doesn't fix the fact that these situations will happen, without the escape of availability of public IPv4 addresses, in the near future. And, of course, if nobody puts pressure on the VPN-sellers, these issues will stick around for *years*. Geert Jan

7 8

IPv6 Act Now update
by Chris Buckridge 04 Feb '10

04 Feb '10

Hi all, A couple of items from the IPv6 Act Now newsfeed over the last week that may be of interest: YouTube Turns On IPv6 support, Net Traffic Spikes (2 Feb 2010) Martin Millnert has already posted some useful information to the list on this subject. Google has quietly turned on IPv6 support for its YouTube video streaming Web site, which sent a spike of IPv6 traffic across the Internet. <http://www.ipv6actnow.org/2010/02/youtube-turns-on-ipv6-support-net-traffic… > IPv6: the new platform (3 Feb 2010) Some perspectives on IPv6 deployment from India, the second largest telecom market in the world. The article includes discussion of the IPv6 survey done in the Asia Pacific recently by TNO/GNKS, who worked with the European Commission and the RIPE NCC on an IPv6 Deployment Monitoring Survey last year: <http://www.ipv6actnow.org/2010/02/ipv6-the-new-platform/> Separate to IPv6 Act Now, the IPv6 adoption issue continues to gain exposure in the mainstream press. Dutch news site nu.nl ran an article this week based on the recent announcement from the NRO and ICANN that the pool of available IPv4 address space has reached 10%: <http://www.nu.nl/internet/2175537/internetadressen-raken.html> [in Dutch] Regards, Chris Buckridge, RIPE NCC

1 0

How polluted is 1/8?
by Mirjam Kuehne 03 Feb '10

03 Feb '10

[Apologies for duplicates] Dear colleagues, After 1/8 was allocated to APNIC last week, the RIPE NCC did some measurements to find out how "polluted" this block really is. See some surprising results on RIPE Labs: http://labs.ripe.net/content/pollution-18 Please also note the call for feedback at the bottom of the article. Kind Regards, Mirjam Kuehne

1 0

Youtube over IPv6!
by Martin Millnert 03 Feb '10

03 Feb '10

Hello fellow IPv6 admirers, (I'm sending this here on behalf of Shane. :) ) As you might know already, Google has been so *tremendously* kind to, as of ~22 hours ago, initiate the enabling of their Youtube site over IPv6. It is currently available to participants in their Google over IPv6 program, http://www.google.com/intl/en/ipv6/ . The www.youtube.com user interface so far does *not* have any AAAA published, but the *much* more important (traffic wise) image and video servers do. Use tcpdump to verify. Eg., anticimex@hsa:~$ host s.ytimg.com s.ytimg.com is an alias for static.cache.l.google.com. static.cache.l.google.com has address 74.125.13.213 static.cache.l.google.com has IPv6 address 2001:4860:4001:402::15 anticimex@hsa:~$ host v1.lscache1.c.youtube.com v1.lscache1.c.youtube.com is an alias for v1.lscache1.l.google.com. v1.lscache1.l.google.com has address 74.125.97.80 v1.lscache1.l.google.com has IPv6 address 2001:4860:4001:402::10 The time when it began is clearly visible in our own IPv6 graphs, http://stats.csbnet.se/public/ipv6/csbnet-ipv6-traffictypes.html , and you can see similar correlated data at for example http://www.de-cix.net/content/network.html . To figure out all the mappings they give you, if you want to trace etc, something similar to: "for i in `seq 1 24 `; do for y in `seq 1 8 `; do host v$i.lscache$y.c.youtube.com ; done ; done | grep 'IPv6 address' | cut -d' ' -f 1,5 | sort -uk2" could be used - at least from where I sit. Adjust for correctness. I'm sure that as this matures and bugs are wiped out, there will no longer be any need for the Google over IPv6 program and everybody will be able to access this. Until then, join! :) I bow to the Google IPv6 team, their netops and everybody there that is working so hard to give the Internet true CONTENT on IPv6. Yippie! From 48514, much of love! Thank you guys (and gals)! Regards, -- Martin Millnert <millnert(a)csbnet.se>

1 1