RIPE 55

• content to the Chair of the working group.
• format to [email protected].

Minutes IPv6 Working Group session

Meeting: RIPE 55, Amsterdam
Date: Thursday, 25 October 2007
Time: 16:00 - 17:30 (UTC +0200)
Chair: David Kessens
Minutes: Robert Kisteleki
Jabber: http://www.ripe.net/ripe/meetings/ripe-55/jabber-files/ipv6.txt
J-Scribe: Timothy Lowe
Audio-1: http://www.ripe.net/ripe/meetings/ripe-55/podcast-files/ipv6.mp3
WG URL: http://www.ripe.net/ripe/wg/ipv6/index.html
Material: http://www.ripe.net/ripe/meetings/ripe-55/presentations/thursday.html
Agenda: http://www.ripe.net/ripe/meetings/ripe-55/agendas/ipv6.html

A. Administrative Matters (David Kessens)

No additonal agenda items.

B. Experience with IPv6 in the RIPE NCC network - Mark Guz, RIPE NCC
http://www.ripe.net/ripe/meetings/ripe-55/presentations/guz-practical-IPv6-use.pdf

Question: Have you considered firewall v5.4 -> v6 upgrade?

Mark responded that, according to Juniper, there are no improvements in terms of IPv6, so we felt we don't need to the upgrade at the moment.

Question: Does Juniper promise to do IPv6 in ASICs?
Mark answered that, according to them, v6 is done in ASICs.

Question (David Kessens): v6 for mail services would give statistics over deployed IPv6 mail servers (in Europe). Did you consider collecting statistics on the v6 mail usage?
Mark responded that yes, it's a good idea.

Question (Lorenzo Colitti, Google): How many external complaints did you have because of v6?

Mark responded that there were mostly internal complaints; a lot of them. With regards to the external connectivity, our staff (probably) experienced delays in some cases (eg. waiting for AAAA timeouts). If we had WWW behind the firewall, we'd have had
many more tickets.

Comment (Jordi Palet, Consulintel): Consider 6to4 / Teredo relay on the RIPE NCC infrastructure; it can improve reachability and the delay problems for some clients.

The WG Chair encouraged members to come up with similar presentations at a next RIPE Meeting about their experiences with IPv6

C. Two jokes, one and half proposal (IPv4->IPv6 transition in DSL environment) - Geza Turchanyi
http://www.ripe.net/ripe/meetings/ripe-55/presentations/turchanyi--two-jokes.pdf

There were questions.

D. Report(s) About Actual v6 Traffic Volume Compared to v4? What's real out there, not what's on PowerPoint? - Henk Steenman, AMS-IX and input from the audience
http://www.ripe.net/ripe/meetings/ripe-55/presentations/steenman-ipv6.pdf

Question (David Kessens): Is the traffic equally distributed between members,
or just two specific members.
Henk responded that it's not known but it's possible to find out.

The WG Chair asked the audience if anyone has seen Vista playing a part here?
Jordi Palet responded that he will give a presentation about this in another session which shows a big increase in traffic.

Question (Jordi Palet): The graphs did not include encapsulated IPv6 traffic? that probably influences the result a lot.

Henk responded that this is correct.

Real IPv6 traffic - Carlos Friacas
http://www.ripe.net/ripe/meetings/ripe-55/presentations/friacas-real-ipv6-traffic-as1930.pdf

Comment (Gert Doering, SpaceNet AG): The 2007 traffic graphs seems to be stable. That looks weird. Is the DNS traffic incoming ueries or overall incoming packets?
Carlos answered that it's only incoming traffic.

Question: About e-mail, what is the rate of spam in the v6 graph?
Carlos responded that the rate was probably not much.

Comment (David Kessens): I definitely received spam over v6.

Comment (Jordi Palet): One explanation for the the v6 traffic not growing can be the effect
of transition traffic.

Comment: Traffic peaks seen in the 2005/2006 graphs seems to relate to
holidays.

Question (Max Tulyev, NetAssist LLC): Are there statistics about what kind of services use IPv6 more?

Carlos answered that there were no statistics yet, but we're working on it.

Comment (David Kessens): The peeks might corelate with Fedora Core release schedule, and some sites encourage v6 usage for downloading.

E. Follow-up: Global IPv6 routing table status (discussion) - Gert Doering and input from the audience
http://www.ripe.net/ripe/meetings/ripe-55/presentations/doering-ipv6-routing.pdf

There were no questions.

F. RIPE Community Resolution on IPv4 Depletion and Deployment of IPv6 msg00047 and follow-up mails on ipv6 and address-policy mailing list - Gert Doering, David Kessens
http://www.ripe.net/ripe/maillists/archives/ipv6-wg/2007/msg00047.html

There was an extensive and detailed discussion on what the exact text should look like, with comments like:

• Replace 'extra address space' with 'address space growth'.
• The text should be as broad as possible.
• 'Policy' might mean different things to different people - especially for governments.
• This whole third point could be dropped
• 'operators' might not have the intended meaning, could be interpreted differently.

The term needs to be revisited.

• Rewording for plainer English might help.

quick survey of the audience supported Rob Blokzijl's suggestion that Sander Steffan (Address Policy Working Group Co-Chair), Gert Doering (Address Policy Working Group Co-Chair)  and David Kessens, supported
by a technical writer from the RIPE NCC, reword the proposal and present the new version at the Friday morning plenary. The proposal that was subsequently presented can be found at:
http://www.ripe.net/ripe/meetings/ripe-55/presentations/friday.html

G. Developments/initiatives regarding IPv6 in the RIPE region and beyond - Input from the audience

There were no questions.

Y. Input for the RIPE NCC Activity Plan (input from the audience)

This item was postponed.

Z. AOB

None.