Hello
", "time": "2026-07-24T07:01:13.000Z"}, {"author": "Nick Buraglio", "text": ":wave:
", "time": "2026-07-24T07:04:34.000Z"}, {"author": "Tim Wicinski", "text": "Welcome from EST (UTC-4)
", "time": "2026-07-24T07:07:37.000Z"}, {"author": "Jen Linkova", "text": "Chairs: as I wasn't fat enough to join the queue after the chairs slides, putting this idea here.....those who attended the iabopen probably heard about the outreach program, when the IETF trying to reach out to operators. In that context: what do you (and the group) think about having an interim colocated with the RIPE meeting next May? It will be right between two IETF meetings, it will bring operators in basically for free. I'll be attending, if any onsite assistance is needed
", "time": "2026-07-24T07:08:59.000Z"}, {"author": "\u00c9ric Vyncke", "text": "s/fat/fast/ ;-)
", "time": "2026-07-24T07:12:23.000Z"}, {"author": "Franck Martin", "text": "This is dual stack, not IPv6-only yet?
", "time": "2026-07-24T07:12:31.000Z"}, {"author": "\u00c9ric Vyncke", "text": "but good idea and suggestions
", "time": "2026-07-24T07:12:33.000Z"}, {"author": "Jen Linkova", "text": "Lol
", "time": "2026-07-24T07:12:43.000Z"}, {"author": "Franck Martin", "text": "cool presentation
", "time": "2026-07-24T07:12:54.000Z"}, {"author": "Jen Linkova", "text": "Fat....fast fingers ;))
", "time": "2026-07-24T07:12:55.000Z"}, {"author": "Franck Martin", "text": "I'm not sure there is a solution out there or even documentation to deal with an IPv4 application on a dual stack server, where all the applications are talking with IPv6-only services on other servers.
\nit would mean only some tcp/udp port needs to got through some sort of nat46...
DNS for routing just sounds a bit fragile
", "time": "2026-07-24T07:27:00.000Z"}, {"author": "Franck Martin", "text": "yeah, performance is more a routing issue... than a server issue
", "time": "2026-07-24T07:38:24.000Z"}, {"author": "Franck Martin", "text": "we should also check with what NIST do...
", "time": "2026-07-24T07:40:47.000Z"}, {"author": "Nick Buraglio", "text": "@jen regarding the colocation with RIPE, I think we should talk about that more.
", "time": "2026-07-24T07:41:03.000Z"}, {"author": "Tim Wicinski", "text": "Getting Changes to stub resolvers can be a long tail
", "time": "2026-07-24T07:49:03.000Z"}, {"author": "Tom Hill", "text": "+1` on the interim co-located with RIPE
", "time": "2026-07-24T07:50:24.000Z"}, {"author": "Tim Wicinski", "text": "Ripe co-located interim sounds good.
", "time": "2026-07-24T07:50:59.000Z"}, {"author": "Jen Linkova", "text": "We definitely should talk more about interim. I'm bringing this now only as I suspect the Mac number of people are paying attention..:) we still have two ietfs before that
", "time": "2026-07-24T07:56:28.000Z"}, {"author": "XiPeng Xiao", "text": "@meetecho: battery dead lin v6ops, the remote
", "time": "2026-07-24T07:59:15.000Z"}, {"author": "Lorenzo Miniero", "text": "Help is on the way
", "time": "2026-07-24T07:59:47.000Z"}, {"author": "Franck Martin", "text": "I agree, regression is a big pain to handle when moving to IPv6-only. My overall short comment, I don't see an issue as long as this is all optional, many applications don't care, but some care and do their own \"load balance/IP selection\"
", "time": "2026-07-24T08:13:19.000Z"}, {"author": "Tim Wicinski", "text": "Managers care about engineering time spent and if it eats into other projects
", "time": "2026-07-24T08:14:50.000Z"}, {"author": "Franck Martin", "text": "Tim Wicinski said:
\n\n\nManagers care about engineering time spent and if it eats into other projects
\n
Managers care about the KPI their manager sets.
", "time": "2026-07-24T08:16:15.000Z"}, {"author": "Franck Martin", "text": "python legacy calls do not support IPv6, you need to change the code to support IPv6...
", "time": "2026-07-24T08:18:54.000Z"}, {"author": "Ond\u0159ej Caletka", "text": "@Franck which calls you mean? The legacy C-calls like gethostbyname()? Nobody is supposed to use them :D
", "time": "2026-07-24T08:22:25.000Z"}, {"author": "\u00c9ric Vyncke", "text": "'need to be brave enough' by Knight Philip ;-)
", "time": "2026-07-24T08:22:45.000Z"}, {"author": "Franck Martin", "text": "IPv6 needs a culture change supported by the C-suite.
", "time": "2026-07-24T08:23:16.000Z"}, {"author": "Franck Martin", "text": "Ond\u0159ej Caletka said:
\n\n\n@Franck which calls you mean? The legacy C-calls like gethostbyname()? Nobody is supposed to use them :D
\n
yeah, so many software engineers did not get the memo... even the AIs...
", "time": "2026-07-24T08:25:12.000Z"}, {"author": "Daniel Bernier", "text": "+1 Tim Wicinski
", "time": "2026-07-24T08:25:30.000Z"}, {"author": "\u00c9ric Vyncke", "text": "I was missing the fonts
", "time": "2026-07-24T08:25:36.000Z"}, {"author": "Mark Andrews", "text": "Hundreds of millions of home users have gone from IPv4 only to dual stack without knowing they have gone to dual stack. Basically it was just turned on by the ISP and CPE announced IPv6 was available. The world didn't fall over.
", "time": "2026-07-24T08:26:20.000Z"}, {"author": "Daniel Bernier", "text": "Managers care about ROI of a technology shift either in OPEX savings or Revenu increase.
", "time": "2026-07-24T08:27:01.000Z"}, {"author": "Mark Andrews", "text": "The boxes they have are dual stack capable to day.
", "time": "2026-07-24T08:28:00.000Z"}, {"author": "Daniel Bernier", "text": "Infrastructure people want to transition because it eases the IPv4 maintenance, overlap challenges with common services and scarcity of available addresses.
", "time": "2026-07-24T08:28:19.000Z"}, {"author": "Mark Andrews", "text": "Going IPv6 only is harder because you have to ensure all your equipment supports it.
", "time": "2026-07-24T08:28:39.000Z"}, {"author": "Donald Sharp", "text": "it's been my experience transition is mostly held up by C-suite as opposed to the actual feet on the ground. This is a people/political problem
", "time": "2026-07-24T08:29:00.000Z"}, {"author": "Franck Martin", "text": "going IPv6-only means you are doing an IPv4 outage, and no one like any kind of outage...
", "time": "2026-07-24T08:29:39.000Z"}, {"author": "Daniel Bernier", "text": "end users (tenants, app devs, consumers) could not care less but infrastructure salvation is at their depends.
", "time": "2026-07-24T08:29:40.000Z"}, {"author": "Mark Andrews", "text": "Time to tell the C-suites \"You are most probably running IPv6 at home and don't even know it.\"
", "time": "2026-07-24T08:31:16.000Z"}, {"author": "Franck Martin", "text": "you need to convince the c-suite, OMB used to be scary, but it is not... best way to convince is to talk about K8S NATmargedon!
\nLike Philipp mentioned in his presentation
IPv6 is a clear opportunity for infrastructure simplification IF the basics are simple to implement (i.e. the basic reqs) are simple enough.
", "time": "2026-07-24T08:32:20.000Z"}, {"author": "Daniel Bernier", "text": "currently they are not
", "time": "2026-07-24T08:32:43.000Z"}, {"author": "Donald Sharp", "text": "on why access points do not have the ability to send unicast, I would guess RA implementers have always just thought of the interface as an abstract thing underneath them in the stack. There never was a consideration
", "time": "2026-07-24T08:33:56.000Z"}, {"author": "\u00c9ric Vyncke", "text": "@Lorenzo can you be more specific about the SNAC issue ? on the AIL (Wi-FI) it is mostly a normal router
", "time": "2026-07-24T08:38:56.000Z"}, {"author": "Antonios Chariton", "text": "IPv6 and WiFi Multicast has always been a bit challenging. In RADIUS assigned VLANs for many WiFi deployments by default devices will get addresses via SLAAC from all VLANs with at least one wireless device in them. They won\u2019t work, and Apple devices will ignore them, but it breaks Android. If these were converted, too, it would also be fine.
", "time": "2026-07-24T08:45:08.000Z"}, {"author": "Ond\u0159ej Caletka", "text": "Oh yeah, leaking broadcasts/multicasts between VLANs on some APs has been a problem for years already. Which is a bit sad.
", "time": "2026-07-24T08:49:40.000Z"}, {"author": "Nick Buraglio", "text": "I have also seen this across certain merchant silicon managed switches. It's quite horrible
", "time": "2026-07-24T08:50:42.000Z"}, {"author": "Nick Buraglio", "text": "Specifically, a few L2 managed switches that are popular in hospitality and WISP networks
", "time": "2026-07-24T08:51:38.000Z"}, {"author": "Kyle Rose", "text": "That is pretty awful. Having to be bug-for-bug compatible with the long tail of broken hardware is not a good way to run a railroad, and yet... here we are.
", "time": "2026-07-24T08:52:20.000Z"}, {"author": "Franck Martin", "text": "I saw a France Telecom presentation on YT, talking about the issue of roaming, when you are on an IPv4-only network, trying to reach back your IPv6-only home phone network...
", "time": "2026-07-24T08:52:25.000Z"}, {"author": "Antonios Chariton", "text": "Yes, the switches do it too, but I was more surprised by the top tier enterprise APs designed for this purpose :)
", "time": "2026-07-24T08:52:34.000Z"}, {"author": "Nick Buraglio", "text": "\"why does my every have an IPv6 address from every VLAN on the switch?\" was a fun day.
", "time": "2026-07-24T08:53:34.000Z"}, {"author": "Nick Buraglio", "text": "every host*
", "time": "2026-07-24T08:53:44.000Z"}, {"author": "Franck Martin", "text": "if I understand correctly, roaming, encapsulates your traffic back to your home mobile network, that then put it on the wire...
", "time": "2026-07-24T08:54:09.000Z"}, {"author": "Kyle Rose", "text": "I ran into a variant of this problem recently, with Kea responding to DHCP(v4) requests on Linux on both the VLAN interface and the parent untagged interface, resulting in a race condition where the client would half the time end up with the wrong IP. This bug somehow survived for five years before being resolved.
", "time": "2026-07-24T08:54:21.000Z"}, {"author": "Nick Buraglio", "text": "@Franck I know some systems do the tunnels but I don't know that all of them do it that way. I'm also pretty out of data on wifi tech, so probably just missing it
", "time": "2026-07-24T08:55:03.000Z"}, {"author": "Franck Martin", "text": "@Nick Buraglio I'm not talking about wifi but about telecom mobile roaming
", "time": "2026-07-24T08:55:57.000Z"}, {"author": "Nick Buraglio", "text": "ahhh, apologies. missed that detail.
", "time": "2026-07-24T08:56:18.000Z"}, {"author": "Brian Carpenter", "text": "Mobile IPv6 needed a home agent. RIP.
", "time": "2026-07-24T08:57:07.000Z"}, {"author": "Franck Martin", "text": "Nick Buraglio said:
\n\n\nahhh, apologies. missed that detail.
\n
no need to apologize.
", "time": "2026-07-24T08:58:17.000Z"}, {"author": "Franck Martin", "text": "if you understand French, similar concerns re ipv6-only on mobile: https://www.youtube.com/watch?v=ahlY1vwM8qE
\n