commit 58000de4faea96761d9e87dcd7cb1fcef63e961e Author: teor teor@torproject.org Date: Mon Feb 10 12:46:12 2020 +1000

Prop 311: Add Optional IPv6 Bandwidth Self-Tests

Add an optional change that makes relays do bandwidth self-tests over IPv6.

This change increases the connection and bandwidth IPv6 statistics. Related to proposal 313, which covers IPv6 statistics.

Related to 33159. --- proposals/311-relay-ipv6-reachability.txt | 21 +++++++++++++++++++++ 1 file changed, 21 insertions(+)

diff --git a/proposals/311-relay-ipv6-reachability.txt b/proposals/311-relay-ipv6-reachability.txt index 978dc9b..e4c9414 100644 --- a/proposals/311-relay-ipv6-reachability.txt +++ b/proposals/311-relay-ipv6-reachability.txt @@ -573,6 +573,27 @@ Ticket: #24404 network makes it more likely that an existing connection can be used for the final hop of a relay IPv6 ORPort reachability check.

+4.4.5. Relay Bandwidth Self-Tests Over IPv4 and IPv6 + + In this proposal, we only make relays (and bridges) use IPv6 for their + reachability self-tests. + + We propose this optional change, to improve the accuracy of relay (and + bridge) bandwidth self-tests: + + Relays (and bridges) perform bandwidth self-tests over IPv4 and IPv6. + + If we implement good abstractions for relay self-tests, then this change + will not need much extra code. + + If we implement IPv6 extends for all relay circuits (see section 4.4.4), + then this change will effectively be redundant. + + Doing relay bandwidth self-tests over IPv6 will create extra IPv6 + connections and IPv6 bandwidth on the tor network. (See + [Proposal 313: Relay IPv6 Statistics].) In addition, some client circuits + may use the IPv6 connections created by relay bandwidth self-tests. + 4.5. Alternate Reachability Designs

We briefly mention some potential reachability designs, and the reasons that