IPv6 Working Group Minutes - RIPE 78

Date: 23 May 2019, 09:00 - 10:30
WG Co-Chairs: Jen Linkova, Benedikt Stockebrand, Raymond Jetten
Scribe: Boris Duval
Status: Final

1.Welcome / Agenda Bashing

The presentation is available at: https://ripe78.ripe.net/archives/video/82

There were no questions or comments.

2.Update on ITU SG-20 IPv6 Deployment Plan

Marco Hogewoning

The presentation is available at:https://ripe78.ripe.net/archives/video/84/

There were no questions or comments.

3.IPv6 Reliability Measurements

Geoff Huston

The presentation is available at: https://ripe78.ripe.net/archives/video/85

Jan Žorž (Internet Society) thanked the presenter and agreed with his findings. He also suggested that SYNC/ACK might be happening because prefixes allocated dynamically via DHCPv6-PD are not getting properly withdrawn when the prefix changes. He mentioned that this was documented in RIPE 690 and will be discussed further in the RIPE-690 Follow-up.

Jen Linkova (Google) asked the presenter why DNS 6.4 failed with IPv6 but succeeded with IPv4. She also asked if it could be run with IPv6 only.

The presenter answered that while he saw ‘rock-solid’ performance from clients that deployed IPv6 with 464XLAT, he could not vouch for other transition technologies such as DNS 6.4. He added that IPv6 only was not a viable option at the moment.

Kostas Zorbadelos (Canal+ Telecom) agreed with the presenter on his findings and asked if he had more data on transition mechanisms and network breakage.
The presenter replied that this kind of data was not public and that only a few companies published data on the subject.

Jan Žorž asked if it was possible to track prefix changes while an ad is played in a browser.

The presenter replied that it was impossible to track because ads usually play for less than ten seconds and measuring prefix changes takes would necessitate more time.  

There were no more questions or comments.

 4. Large-Scale Deployment of IPv6-Enabled Wi-Fi Hotspots

Enno Rey

The presentation is available at: https://ripe78.ripe.net/archives/video/88/

Tim Wattenberg (University of Düsseldorf) mentioned that the IETF is aware of the MDNS traffic issue and is thinking of shifting from Multicast to Unicast for service discovery to solve it.

Blake Willlis (Zayo / iBrowse / L33) thanked the presenter.

There were no more questions or comments.

5. Mobile IPv6 BCOP Doc

Massimiliano Stucchi

The presentation is available at: https://ripe78.ripe.net/archives/video/93

Jen Linkovasuggested that the RIPE NCC organises a tutorial on how mobile networks work. The presenter said that they will discuss the possibility.  

There were no more questions or comments.

6. RIPE-690 Follow-up: SLAAC’s Reaction to Renumbering Events - IETF I-D

Jan Žorž

The presentation will be available shortly. 

Jen Linkova agreed with the presentation and added a few remarks on CPEs and prefix delegation and suggested a Hackathon. The presenter agreed that something must be done.  

Kostas Zorbadelos (Canal+ Telecom) shared his experience with the audience about a similar CPE related issue that was solved. He also added that nothing was wrong with current IPv6 specifications.

There were no more questions or comments.

Technology Deep Dive on Modern Router Architectures

Ignas Bagdonas

The presentation is available at: https://ripe78.ripe.net/archives/video/115/

In addition to the IPv6 Working Group agenda, Ignas Bagdonas (Equinix) gave a “Deep Dive” lecture on Modern Router Architectures. His presentation was followed by an open discussion.