RIPE 84

Archives

Ekaterina Cherkasova - 19-05-2022 10:59:48
Good afternoon everyone, I’m Ekaterina from the RIPE NCC. This chat panel is meant for discussion ONLY. If you have questions for the speaker and you want the session chair to read it out, please write it in the Q&A window also stating your affiliation. Otherwise, you can ask questions using the microphone icon. Please note that all chat transcripts will be archived and made available to the public on https://ripe84.ripe.net/. The RIPE Code of Conduct: https://www.ripe.net/publications/docs/ripe-766

Kurt Kayser - 19-05-2022 11:13:07
BTW link-local is not fe80::/64 rather fe80::/10

Eva Lauren Kelly - 19-05-2022 11:15:11
Regarding RA's -- I think the firewall rules are still applicable for the (more modern) 'Mobile Hotspot' feature in Windows 10+, though I would agree they should probably be enabled/disabled only as required along with that feature

Petra Zeidler - 19-05-2022 11:18:55
the really funny part is that when you have no connection to the Internet (because isolated network for protection) you are on a dangerous public network

Tim Chown - 19-05-2022 11:26:26
"random users without Windows". probably most of us.

Petra Zeidler - 19-05-2022 11:26:44
"enterprise admins knwo what they do" is an aspirational stance :)

christian bretterhofer - 19-05-2022 11:42:18
please point me to enterprise admins with network or fw knowledge

Tim Chown - 19-05-2022 11:43:02
there are lots in universities. but not so many universities deploy ipv6 except for specific science communities

Jen Linkova - 19-05-2022 11:43:06
>please point me to enterprise admins with network or fw knowledge
I'm offended ;))))

Tim Chown - 19-05-2022 11:43:51
i dont think there was anything new in this talk, as such. sadly.

Tim Chown - 19-05-2022 11:45:45
you need a reason to deploy. cern had one, and as a result here on Janet we now see Imperial College with a 100Gbps link to their DC and regularly filling that and most of the time 100% IPv6 traffic. That came from 4-5 years of work by the HEPiX IPv6 WG doing the groundwork so the CERN distributed storage nodes would support v6, and the tools use it for transfers.

Tim Chown - 19-05-2022 11:46:15
the reason was wanting direct addressing between all compute and storage nodes, no NAT.

Tim Chown - 19-05-2022 11:47:59
but overall, uk v6 traffic is 35%, on janet it's under 10%.

Petra Zeidler - 19-05-2022 11:49:00
yes, the reason enterprises don't migrate is that networking is barely on life support, there's usually no drive at all to invest into IT in general and the network in specific

Jen Linkova - 19-05-2022 11:50:21
at the end of the day people migrate to ipv6 because they ran out of IPv4. SMBs can live behind a single IPv4 address (they can run IPX or APpletalk or any other crazy thing inside their networks, if they want ;)

Jen Linkova - 19-05-2022 11:51:02
ISPs and *large* enterprises do have a very good reasons for IPv6 - and they are doing this, some of them at least.

Tim Chown - 19-05-2022 11:51:38
this is true, cern ran out, or knew sites who take part in their experiments were running out.

Tim Chown - 19-05-2022 11:52:41
having flashbacks to 7872 Jen? :)

Jen Linkova - 19-05-2022 11:53:06
@Tim: oh yeah ;)) But you see - 94% for short DO now!!

Jen Linkova - 19-05-2022 11:53:10
we *are* moving

Petra Zeidler - 19-05-2022 11:53:12
the question "what set of addresses will the packet have on that leg of its trip" alone is reason to migrate, but needs selling too

Tim Chown - 19-05-2022 11:53:40
it seems so jen.

Jen Linkova - 19-05-2022 11:54:24
and look - short DO does goes through even we do not have any use case for that. Now imagine we come up with a use case.

Christian Marx - 19-05-2022 11:55:00
shorter link for survey would help :)

Pawel Worach - 19-05-2022 11:55:18
or a QR code :)

Tim Chown - 19-05-2022 11:55:29
the problematic drops in R&E networks are the firewalls that drop icmpv6 ptb and prevent deployment of 9k mtu for the big data transfers (like cern data)

Jen Linkova - 19-05-2022 11:55:34
https://ripe84.ripe.net/programme/meeting-plan/ipv6-wg/ - links to slides ;)

Christian Marx - 19-05-2022 11:56:13
thx

Ekaterina Cherkasova - 19-05-2022 12:22:00
This session has now ended. The next session is the Database WG and it will start at 14.00. More info on the RIPE 84 meeting plan: https://ripe84.ripe.net/programme/meeting-plan/