Hey Relay Operators, I'll be at the upcoming WHY2025 hacker camp in the Netherlands. WHY2025 is an excellent opportunity to do a Relay Operator meetup at some point during the camp days. Who else from here is going? At the latest CCCamp, we did a meetup at the BornHack village, but we did receive some negative feedback on the noise level in that general area, so I was thinking of investigating if there is a quieter place with some good seating where we could host the meetup this time, and announce the time and location here on the list. Suggestions of locations are most welcome. I recently received a link from tornth on #tor-relays on IRC/Matrix to https://wiki.why2025.org/Project:Tor_Relays, where it's relatively easy to add "Tor Relays" as a project from your WHY2025 wiki account. Cheers, Alex -- Alexander Hansen Færøy
Hey folks, Let's meet in 2 days, on Sunday, 2025-08-10, from 17:00 to 18:00, at the BornHack star tent in the Loud North cluster[1]. Both BornHack and ANDERSTORPSFESTIVALEN won't be blasting out music during this time, so we'll hopefully have a meeting where we can hear each other. Remember, Morgan has a talk[2] tomorrow on their work on Gosling and Ricochet-Refresh at 19:00 in Brachium. Hope you have a joyful festival! Cheers, Alex [1]: https://map.why2025.org/?building=63&building_table=villages#maphash=19/52.6... [2]: https://program.why2025.org/why2025/talk/TMS3DC/ -- Alexander Hansen Færøy
On 08/08/2025 19.13, Alexander Hansen Færøy via tor-relays wrote:
Let's meet in 2 days, on Sunday, 2025-08-10, from 17:00 to 18:00, at the BornHack star tent in the Loud North cluster[1]. Both BornHack and ANDERSTORPSFESTIVALEN won't be blasting out music during this time, so we'll hopefully have a meeting where we can hear each other. Hello Tor Relay Operators,
We had our meetup yesterday here at WHY2025. Here are my notes, based on my memory and the whiteboard we used during the meeting. Thanks to everybody who participated! Announcements: - Alex talked about the upcoming community event in Denmark, and information on the event is available at https://tcg2025.4711.se/ Discussions: - The Future of Tor: - The group discussed concerns about Tor ("the company") and its base in the US with the current political situation there. - Alex described the current funding situation, particularly around the grants that he is aware of, that is primarily related to the Arti project, and how they are currently considered stable. - Discussions took place on whether it would be beneficial to establish a European alternative to Tor ("the company") in the future, considering the desired resilience properties that would help mitigate potential negative impacts if the US situation deteriorates. - What is the IPv6 situation with Tor? - Alex mentioned that he and GK had a discussion earlier this year about both a potential solution for having IPv6-only Middlenodes (in a network where all guards and all exits are dual-stack) and how one could do something similar to Happy Eyeballs within the Tor ecosystem. The former is largely a policy change, whereas the latter requires some changes in C Tor and Arti. A Tor spec proposal is needed, along with funding to build it. - It was mentioned that we should address the bridge situation with IPv6, as it is not in a good state. - Mixnets/Store and Forward inside the Tor ecosystem - We briefly discussed whether there were plans to add anything like that to the Tor ecosystem. Alex said that we had no such plans in the near future, with the ongoing work all being focused on getting Arti into the Tor Browser, VPN, and the Relay operators. - Alex added that if people attended Morgan's talk from the day before, they could imagine a world where Onion Service-based Instant Messengers would benefit from being able to do offline messaging. - Anonymity Properties and NetFlow trading - The group discussed the recent news about companies actively purchasing NetFlow data for analysis purposes. - Alex mentioned the recent work related to Website Traffic Fingerprinting and the padding layer in C Tor. The Network Team is currently working on integrating MaybeNot[1] in Arti for even stronger padding machines. - Status of Arti Relay: - Alex gave the group a short update on the current status of Arti Relays: - The Network Team is currently working on: - Simulation tooling with Shadow and Chutney. - Directory Authority support in Arti, including handling the large number of different documents available via the directory layer of Tor. - Core Relay functionality. - (Unrelated to Arti Relays) Conflux, Congestion Control, and PoW (for services) have also been priorities recently, as they overlap significantly, with some components serving both client and relay-side needs. - We will have an open call for volunteers at a later point to help create a parallel Tor test network. Initially, it will consist of C Tor nodes, but we plan to experiment with Arti Relays there as development happens. We will likely need to reboot the network multiple times to prepare for the transition to the production Tor network. - Stateless Relays - The group discussed shortly the desire by some folks to experiment with stateless relays. The Italian group has been working in this space recently and provided a brief update on their progress. - We talked about problems related to the (relay) Onion Key when that is updated from a restarted stateless node, and what could be done about that. Cheers, Alex [1]: https://github.com/maybenot-io/ -- Alexander Hansen Færøy
Quoting Alexander Hansen Færøy via tor-relays (2025-08-11 15:48:52)
- It was mentioned that we should address the bridge situation with IPv6, as it is not in a good state.
I would like to hear more about this. I know there are quirks on configuring IPv6 bridges c-tor. The mayor one for me is that the transport line is published only for IPv4 in dual stack bridges. On rdsys we get the IPv6 address from networkstatus and the port from the transport and try them, if it works we publish this bridge. So most of dual stack bridges work. I'll be happy to hear if we can do something from rdsys side to improve the situation. I assume the c-tor situation will stay as it is and we'll work on make it better in arti. -- meskio | https://meskio.net/ -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- My contact info: https://meskio.net/crypto.txt -=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=- Nos vamos a Croatan.
We talked about problems related to the (relay) Onion Key when that is updated from a restarted stateless node, and what could be done about that.
That's a problem also experienced by the Snowflake bridges (being more than a single tor instance per bridge), and something i have interest in in the context of webtunnel + geodistributed websites (so again multiple tor instances per bridge). In https://www.petsymposium.org/foci/2023/foci-2023-0011.pdf section 3.2, the Snowflake operators describe a workaround (basically freezing the file), meaning the key becomes about as valuable as the identity key (according to the paper), though you could imagine somehow re-creating it manually and updating your static seed from time to time. regards, trinity-1686a On Mon, 11 Aug 2025 at 15:49, Alexander Hansen Færøy via tor-relays <tor-relays@lists.torproject.org> wrote:
On 08/08/2025 19.13, Alexander Hansen Færøy via tor-relays wrote:
Let's meet in 2 days, on Sunday, 2025-08-10, from 17:00 to 18:00, at the BornHack star tent in the Loud North cluster[1]. Both BornHack and ANDERSTORPSFESTIVALEN won't be blasting out music during this time, so we'll hopefully have a meeting where we can hear each other. Hello Tor Relay Operators,
We had our meetup yesterday here at WHY2025. Here are my notes, based on my memory and the whiteboard we used during the meeting. Thanks to everybody who participated!
Announcements: - Alex talked about the upcoming community event in Denmark, and information on the event is available at https://tcg2025.4711.se/
Discussions: - The Future of Tor: - The group discussed concerns about Tor ("the company") and its base in the US with the current political situation there. - Alex described the current funding situation, particularly around the grants that he is aware of, that is primarily related to the Arti project, and how they are currently considered stable. - Discussions took place on whether it would be beneficial to establish a European alternative to Tor ("the company") in the future, considering the desired resilience properties that would help mitigate potential negative impacts if the US situation deteriorates. - What is the IPv6 situation with Tor? - Alex mentioned that he and GK had a discussion earlier this year about both a potential solution for having IPv6-only Middlenodes (in a network where all guards and all exits are dual-stack) and how one could do something similar to Happy Eyeballs within the Tor ecosystem. The former is largely a policy change, whereas the latter requires some changes in C Tor and Arti. A Tor spec proposal is needed, along with funding to build it. - It was mentioned that we should address the bridge situation with IPv6, as it is not in a good state. - Mixnets/Store and Forward inside the Tor ecosystem - We briefly discussed whether there were plans to add anything like that to the Tor ecosystem. Alex said that we had no such plans in the near future, with the ongoing work all being focused on getting Arti into the Tor Browser, VPN, and the Relay operators. - Alex added that if people attended Morgan's talk from the day before, they could imagine a world where Onion Service-based Instant Messengers would benefit from being able to do offline messaging. - Anonymity Properties and NetFlow trading - The group discussed the recent news about companies actively purchasing NetFlow data for analysis purposes. - Alex mentioned the recent work related to Website Traffic Fingerprinting and the padding layer in C Tor. The Network Team is currently working on integrating MaybeNot[1] in Arti for even stronger padding machines. - Status of Arti Relay: - Alex gave the group a short update on the current status of Arti Relays: - The Network Team is currently working on: - Simulation tooling with Shadow and Chutney. - Directory Authority support in Arti, including handling the large number of different documents available via the directory layer of Tor. - Core Relay functionality. - (Unrelated to Arti Relays) Conflux, Congestion Control, and PoW (for services) have also been priorities recently, as they overlap significantly, with some components serving both client and relay-side needs. - We will have an open call for volunteers at a later point to help create a parallel Tor test network. Initially, it will consist of C Tor nodes, but we plan to experiment with Arti Relays there as development happens. We will likely need to reboot the network multiple times to prepare for the transition to the production Tor network. - Stateless Relays - The group discussed shortly the desire by some folks to experiment with stateless relays. The Italian group has been working in this space recently and provided a brief update on their progress. - We talked about problems related to the (relay) Onion Key when that is updated from a restarted stateless node, and what could be done about that.
Cheers, Alex
[1]: https://github.com/maybenot-io/
-- Alexander Hansen Færøy _______________________________________________ tor-relays mailing list -- tor-relays@lists.torproject.org To unsubscribe send an email to tor-relays-leave@lists.torproject.org
Am 11.08.2025 um 15:48:52 Uhr schrieb Alexander Hansen Færøy via tor-relays:
Alex mentioned that he and GK had a discussion earlier this year about both a potential solution for having IPv6-only Middlenodes (in a network where all guards and all exits are dual-stack) and how one could do something similar to Happy Eyeballs within the Tor ecosystem.
How does this work for machines that use ClientPreferIPv6ORPort 1 Could that be default in torrc without breaking anything? I have that set on my guard/middle relays and detected no issues. https://gitlab.torproject.org/tpo/applications/tor-browser/-/issues/41742 Here someone mentions a case where issues might occur, but without any details. -- Gruß Marco Send unsolicited bulk mail to 1754920132muell@cartoonies.org
How does this work for machines that use ClientPreferIPv6ORPort 1
Sadly ClientPreferIPv6ORPort is somewhat broken with IPv4 only clients today. See https://gitlab.torproject.org/tpo/core/tor/-/issues/40913#note_2994762 for a concrete example of something that can easily fail badly for end users (but there might be other cases, possibly without using bridges). trinity-1686a On Tue, 12 Aug 2025 at 00:01, Marco Moock via tor-relays <tor-relays@lists.torproject.org> wrote:
Am 11.08.2025 um 15:48:52 Uhr schrieb Alexander Hansen Færøy via tor-relays:
Alex mentioned that he and GK had a discussion earlier this year about both a potential solution for having IPv6-only Middlenodes (in a network where all guards and all exits are dual-stack) and how one could do something similar to Happy Eyeballs within the Tor ecosystem.
How does this work for machines that use ClientPreferIPv6ORPort 1
Could that be default in torrc without breaking anything?
I have that set on my guard/middle relays and detected no issues.
https://gitlab.torproject.org/tpo/applications/tor-browser/-/issues/41742 Here someone mentions a case where issues might occur, but without any details.
-- Gruß Marco
Send unsolicited bulk mail to 1754920132muell@cartoonies.org _______________________________________________ tor-relays mailing list -- tor-relays@lists.torproject.org To unsubscribe send an email to tor-relays-leave@lists.torproject.org
participants (4)
-
Alexander Hansen Færøy -
Marco Moock -
meskio -
trinity pointard