[tor-bugs] #23493 [Core Tor/Tor]: IPv6 v3 Single Onion Services fail with a bug warning

[Tor Bug Tracker & Wiki]

#23493: IPv6 v3 Single Onion Services fail with a bug warning
-------------------------------------------------+-------------------------
 Reporter:  teor                                 |          Owner:  teor
     Type:  defect                               |         Status:
                                                 |  needs_revision
 Priority:  Medium                               |      Milestone:  Tor:
                                                 |  0.3.2.x-final
Component:  Core Tor/Tor                         |        Version:
 Severity:  Normal                               |     Resolution:
 Keywords:  prop224, tor-hs, single-onion, ipv6  |  Actual Points:  0.7
Parent ID:                                       |         Points:  1
 Reviewer:                                       |        Sponsor:
-------------------------------------------------+-------------------------

Comment (by dgoulet):

 Thanks for this comment asn and your great work teor.

 I do share these concerns as well. I'll also take blame here on adding
 half of a SoS feature while we were re-writing the entire HS subsystem, we
 shouldn't have done that as we see now that it is much more complicated in
 terms of interface, algorithm and integration.

 I've been silent in the last week about this because I've been thinking
 about it on what is best. My initial idea was to sit down at the Montreal
 meeting and discuss what would be best to go forward.

 I propose we do that in any case but in the meantime, I'm slightly tempted
 to actually go with (c) here so we do not go in the direction of
 accumulating technical debt that we are unhappy with in the hope of
 refactoring in a later version. It usually doesn't work that smoothly...
 The only thing we must be sure of is for a 032 client to be able to reach
 a SoS which it should by design because it doesn't behave differently in
 the case of a SoS or non-SoS service in theory.

 Then, in 033 we introduce the service functionality in a clean well
 discussed interface and for which we all agree about how to proceed with a
 SoS. The number of conditions to switch from one hop to 3 hop or v4 vs v6
 is not that simple, we should nail it down in a spec or doc before we do
 anything more. Sprinkling `direct_conn` in many places of the code
 honestly worries me and I would like us to take a step back and sit down
 together so we can all be on the same page.

--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/23493#comment:17>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online