[tor-dev] Student project to work on IPv6 support (was: [tor-relays] About relay size)

[Santiago R.R.]

(Moving this thread from tor-relay)

On Tue, 3 Oct 2017 09:53:46 -0400

For interposing dual-protocoled nodes along the way, how many do there
have to be for it to become "not too limiting"?

This is one of the questions we need researchers to answer.

I can't help but feel you are overcomplicating this.
Clients create a circuit by randomly picking 3 nodes out of the all-nodes
pile, right? If all 3 happen to be IPv6-capable, then the circuit can go over
IPv6 and all is fine. If some of the 3 happen to be IPv6-only while others are
IPv4-only, the whole selection can be thrown away and repeated.
That way IPv6-only relays could get some usage on a totally random basis, with
no compromises and no restraining "of the next hop based on the previous one",
not hurting anonymity. Clients just need to know which nodes are IPv4-only,
IPv6-only or dual-stack, to not attempt unworkable combinations, discarding
them instead.

Discarding unworkable combinations and restraining node choices seem
equivalent to me, although the relay weights may be different.

And as there are more and more dual-stack or IPv6-only relays, the "throw
away" step will be needed less and less often.

If you think this will work and is safe for client anonymity, then the next step
is to write a tor proposal. Having a concrete design could help with
analysing the anonymity implications as well.
I think IPv6-only relays are a lower priority than better IPv6 and dual-stack
client support, and IPv6-only bridge support But we could do both in the
same release.

Hello tor-dev,

With my colleague JC Bach (in CC), we have proposed a last-year student
project to address IPv6-related issues in Tor for the upcoming semester,
at IMT Atlantique engineering school. There will be two students working
on it. It is hard to say now how far we will arrive, especially because
this is our first approach to Tor entrails.

So this message is to say we have good chances to come back here looking
for help :-)

Cheers,

-- Santiago

[teor]

Post by Santiago R.R.
(Moving this thread from tor-relay)

On Tue, 3 Oct 2017 09:53:46 -0400

For interposing dual-protocoled nodes along the way, how many do there
have to be for it to become "not too limiting"?

This is one of the questions we need researchers to answer.

I can't help but feel you are overcomplicating this.
Clients create a circuit by randomly picking 3 nodes out of the all-nodes
pile, right? If all 3 happen to be IPv6-capable, then the circuit can go over
IPv6 and all is fine. If some of the 3 happen to be IPv6-only while others are
IPv4-only, the whole selection can be thrown away and repeated.
That way IPv6-only relays could get some usage on a totally random basis, with
no compromises and no restraining "of the next hop based on the previous one",
not hurting anonymity. Clients just need to know which nodes are IPv4-only,
IPv6-only or dual-stack, to not attempt unworkable combinations, discarding
them instead.

Discarding unworkable combinations and restraining node choices seem
equivalent to me, although the relay weights may be different.

And as there are more and more dual-stack or IPv6-only relays, the "throw
away" step will be needed less and less often.

If you think this will work and is safe for client anonymity, then the next step
is to write a tor proposal. Having a concrete design could help with
analysing the anonymity implications as well.
I think IPv6-only relays are a lower priority than better IPv6 and dual-stack
client support, and IPv6-only bridge support But we could do both in the
same release.

Hello tor-dev,
With my colleague JC Bach (in CC), we have proposed a last-year student
project to address IPv6-related issues in Tor for the upcoming semester,
at IMT Atlantique engineering school. There will be two students working
on it. It is hard to say now how far we will arrive, especially because
this is our first approach to Tor entrails.
So this message is to say we have good chances to come back here looking
for help :-)

Hi Santiago,

This is great! We would like some help with Tor's IPv6 support.
And we are happy to help you and your students.

How many students?
How much time?
What are your goals for the project?
How much do you expect to get done?

We are at a Tor meeting this week.
We are revising Tor's IPv6 roadmap for the next year.
Next week, this page will be updated:
https://trac.torproject.org/projects/tor/wiki/org/meetings/2017Amsterdam/IPv6Hackfest

We want to help people get code accepted into tor.
Here is how we write code:
https://gitweb.torproject.org/tor.git/tree/doc/HACKING/GettingStarted.md
https://gitweb.torproject.org/tor.git/tree/doc/HACKING/CodingStandards.md

It can help to start by submitting a small change, so you can see how we work.
Then you can make larger changes more easily.
Our bug tracker is:
https://trac.torproject.org/

We are also in #tor-dev IRC on irc.oftc.net.

Please ask questions early, and ask often!
We would love to help you help tor.

Tim (teor)

[Santiago R.R.]

…

Post by Santiago R.R.
Hello tor-dev,
With my colleague JC Bach (in CC), we have proposed a last-year student
project to address IPv6-related issues in Tor for the upcoming semester,
at IMT Atlantique engineering school. There will be two students working
on it. It is hard to say now how far we will arrive, especially because
this is our first approach to Tor entrails.
So this message is to say we have good chances to come back here looking
for help :-)
Hi Santiago,
This is great! We would like some help with Tor's IPv6 support.
And we are happy to help you and your students.

Great, thanks!

Post by Santiago R.R.
How many students?

There will be two.

Post by Santiago R.R.
How much time?

From now until mid-March. Students will have 135h in their schedules to
work on their projects.

Post by Santiago R.R.
What are your goals for the project?

For now, it's still open, but addressing IPv6 support. We should limit
the scope soon, according to open related tickets that could be feasible
to work on.

Post by Santiago R.R.
How much do you expect to get done?

At least, choose a couple of easy-tagged IPv6 tickets, and close them.
However, it's difficult to state on this right now.

Maybe something from:
https://trac.torproject.org/projects/tor/query?status=accepted&status=assigned&status=merge_ready&status=needs_information&status=needs_review&status=needs_revision&status=new&status=reopened&keywords=~ipv6+easy&max=20&col=id&col=summary&col=status&col=owner&col=priority&col=component&col=severity&col=time&col=changetime&col=reporter&desc=1&order=changetime

Post by Santiago R.R.
We are at a Tor meeting this week.
We are revising Tor's IPv6 roadmap for the next year.
https://trac.torproject.org/projects/tor/wiki/org/meetings/2017Amsterdam/
IPv6Hackfest

Good to know about this!

Post by Santiago R.R.
We want to help people get code accepted into tor.
https://gitweb.torproject.org/tor.git/tree/doc/HACKING/GettingStarted.md
https://gitweb.torproject.org/tor.git/tree/doc/HACKING/CodingStandards.md
It can help to start by submitting a small change, so you can see how we work.
Then you can make larger changes more easily.

Understood.

Post by Santiago R.R.
https://trac.torproject.org/
We are also in #tor-dev IRC on irc.oftc.net.
Please ask questions early, and ask often!
We would love to help you help tor.
Tim (teor)

Cheers,

-- Santiago

[teor]

Post by Santiago R.R.



Post by Santiago R.R.
With my colleague JC Bach (in CC), we have proposed a last-year student
project to address IPv6-related issues in Tor for the upcoming semester,
at IMT Atlantique engineering school. There will be two students working
on it. It is hard to say now how far we will arrive, especially because
this is our first approach to Tor entrails.


This is great! We would like some help with Tor's IPv6 support.
And we are happy to help you and your students.




Post by Santiago R.R.
How many students?

There will be two.

Post by Santiago R.R.
How much time?

From now until mid-March. Students will have 135h in their schedules to
work on their projects.

Post by Santiago R.R.
What are your goals for the project?

For now, it's still open, but addressing IPv6 support. We should limit
the scope soon, according to open related tickets that could be feasible
to work on.

Post by Santiago R.R.
How much do you expect to get done?

At least, choose a couple of easy-tagged IPv6 tickets, and close them.
However, it's difficult to state on this right now.

135h is enough to submit a small, one-line change to get used to the tor
patch process, and then do something more substantial with some testing.

Understanding the code, and testing and documenting the fix can take more
time than writing the patch.

Your students could work on getting IPv6 bridges working with private IPv4
addresses. The bridge needs to include a placeholder IPv4 address in its
descriptor, and then the bridge client needs to ignore this address.

Or they could work out why Tor Browser often fails IPv6-only sites like
ipv6.google.com?

We think it's because IPv4 exits don't resolve AAAA addresses. Failed
addresses should've resolved and sent back to the client. And then the
client can use the address to pick its next exit.

We've made some progress on both of these issues, but then ran out of
time.

Post by Santiago R.R.

Post by Santiago R.R.
We are at a Tor meeting this week.
We are revising Tor's IPv6 roadmap for the next year.
https://trac.torproject.org/projects/tor/wiki/org/meetings/2017Amsterdam/
IPv6Hackfest

Good to know about this!

Sorry, this was the wrong link.

Please use:

https://trac.torproject.org/projects/tor/wiki/org/meetings/2017Montreal/IPv6Hackfest

https://trac.torproject.org/projects/tor/wiki/org/roadmaps/Tor/IPv6Features

(We are still adding features to the matrix.)

Post by Santiago R.R.

Post by Santiago R.R.
We want to help people get code accepted into tor.
https://gitweb.torproject.org/tor.git/tree/doc/HACKING/GettingStarted.md
https://gitweb.torproject.org/tor.git/tree/doc/HACKING/CodingStandards.md
It can help to start by submitting a small change, so you can see how we work.
Then you can make larger changes more easily.

Understood.

Post by Santiago R.R.
https://trac.torproject.org/
We are also in #tor-dev IRC on irc.oftc.net.
Please ask questions early, and ask often!
We would love to help you help tor.

[teor]

Hi,

How are these student projects working out?
Is there anything we can do to help?

(There is no need to identify the students, unless they want to be named
on a public mailing list.)

Post by teor

Post by Santiago R.R.



Post by Santiago R.R.
With my colleague JC Bach (in CC), we have proposed a last-year student
project to address IPv6-related issues in Tor for the upcoming semester,
at IMT Atlantique engineering school. There will be two students working
on it. It is hard to say now how far we will arrive, especially because
this is our first approach to Tor entrails.


This is great! We would like some help with Tor's IPv6 support.
And we are happy to help you and your students.




Post by Santiago R.R.
How many students?

There will be two.

Post by Santiago R.R.
How much time?

From now until mid-March. Students will have 135h in their schedules to
work on their projects.

Post by Santiago R.R.
What are your goals for the project?

For now, it's still open, but addressing IPv6 support. We should limit
the scope soon, according to open related tickets that could be feasible
to work on.

Post by Santiago R.R.
How much do you expect to get done?

At least, choose a couple of easy-tagged IPv6 tickets, and close them.
However, it's difficult to state on this right now.

135h is enough to submit a small, one-line change to get used to the tor
patch process, and then do something more substantial with some testing.

Have the students tried small patches?
How did they go?

Post by teor
(Snip)

Post by Santiago R.R.

Post by Santiago R.R.
https://trac.torproject.org/
We are also in #tor-dev IRC on irc.oftc.net.

It might be worth reminding your students that replies on IRC can take hours.
Some people are disappointed when they don't get an instant reply.

Post by teor

Post by Santiago R.R.

Post by Santiago R.R.
Please ask questions early, and ask often!
We would love to help you help tor.

T