[tor-bugs] #19958 [Core Tor/Tor]: Implement proposal 264 (protocol versioning)
Tor Bug Tracker & Wiki
blackhole at torproject.org
Mon Sep 12 16:11:19 UTC 2016
#19958: Implement proposal 264 (protocol versioning)
-----------------------------------------------+---------------------------
Reporter: nickm | Owner: nickm
Type: enhancement | Status:
| needs_revision
Priority: High | Milestone: Tor:
| 0.2.9.x-final
Component: Core Tor/Tor | Version:
Severity: Normal | Resolution:
Keywords: TorCoreTeam201608, review-group-8 | Actual Points: 2
Parent ID: #15055 | Points: 2
Reviewer: dgoulet | Sponsor:
-----------------------------------------------+---------------------------
Changes (by dgoulet):
* status: needs_review => needs_revision
Comment:
Review at https://gitlab.com/dgoulet/tor/merge_requests/11
So far, actual-review-point: 0.3
Comment on the proposal:
* `HSMid` is both for RP and IP. Taking proposal 224 where we'll have a
new version protocol for introduction point and directory. However, the RP
don't need a protocol bump. So if clients start looking at `HSMid` to pick
an RP, it's weird to me that they would be OK with version 1 for RP but
version 2 for IP. Do you think it's fine to do that? or should we have
different values for IP and RP?
* About hidden service. We are currently implementing version 3 of the
protocol where v0-1 are obsolete and 2 is the current one. Yet, this makes
it that `HSDir` and `HSMid` are currently at version 1. I think we should
have them in the proposal and enforce version 2 for both on the relay
side? So, with prop224, we'll make the recommended version be `HSDir=2-3`
but required should be `HSDir=0-3` because technically relay still support
those versions and with 224 we planned to remove the support. So when we
deploy 224, we can switch in the consensus to `HSDir=2-3` and voila. Does
that makes sense?
--
Ticket URL: <https://trac.torproject.org/projects/tor/ticket/19958#comment:7>
Tor Bug Tracker & Wiki <https://trac.torproject.org/>
The Tor Project: anonymity online
More information about the tor-bugs
mailing list