[tor-project] Notes from network team meeting, 3 April 2018
Nick Mathewson
nickm at freehaven.net
Wed Apr 4 14:09:18 UTC 2018
Hi!
You can find our chat logs from
http://meetbot.debian.net/tor-meeting/2018/tor-meeting.2018-04-03-16.59.html
Our pad notes follow below:
= Network team meeting pad, 3 April 2018 =
But above and beyond there's still one name left over,
And that is the name that you never will guess;
The name that no human research can discover--
But THE CAT HIMSELF KNOWS, and will never confess.
-- T. S. Eliot, "The Naming of Cats"
Welcome to our meeting! Mondays at 1700 UTC on #tor-meeting on OFTC.
[**April 2's meeting is one day late, for holiday reasons**]
(This channel is logged while meetings are in progress.)
Want to participate? Awesome! Here's what to do:
1. If you have updates, enter them below, under your name.
2. If you see anything you want to talk about in your updates, put
them in boldface!
3. Show up to the IRC meeting and say hi!
Note the meeting location: #tor-meeting on OFTC!
(See https://lists.torproject.org/pipermail/tor-project/2017-September/001459.html
for background.)
== Previous notes ==
5 March: https://lists.torproject.org/pipermail/tor-project/2018-March/001685.html
26 March: https://lists.torproject.org/pipermail/tor-project/2018-March/001695.html
== Stuff to do every week ==
* Let's check and update the roadmap. What's done, and what's coming up?
isabela will give a bit of orientation on irc
url to roadmap:
https://docs.google.com/spreadsheets/d/1Ufrun1khEo5Cwd6OwngERn829wU3W3eskdrriaYfUBQ/edit#gid=856122210
0.3.3 must fix bugs status:
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=~033-included-2018&milestone=Tor:+0.3.3.x-final&group=status&col=id&col=summary&col=owner&col=type&col=priority&col=component&col=version&order=priority
Needs information - no owner
#21394 - mike is reviewer and needs info
#3940 - isis is reviewer and needs info
#25316
Needs review and has no reviewer
#24454
Needs review, has reviewer and no owner
#24456
Also, have a look at the other tickets in the 0.3.3 milestone that are
NOT currently labelled 033-must. Should any of them get that label?
* Check reviewer assignments at
https://docs.google.com/spreadsheets/d/1Ufrun1khEo5Cwd6OwngERn829wU3W3eskdrriaYfUBQ/edit#gid=95993070
* Check rotations at
https://trac.torproject.org/projects/tor/wiki/org/teams/NetworkTeam/TeamRotations
== Announcements ==
* Beginning general 0.3.4.x triage this week.
* Remember to "/me status: foo" at least once daily.
* Remember that our current code reviews should be done by end-of-week.
* Make sure you are in touch with everybody with whom you are doing
0.3.4.x work.
* Important dates:
* Apr 15, 2018 -- 0.3.3.x stable is supposed to be ready!
* May 15, 2018 -- 0.3.4.x feature freeze!
* Cell crypto meeting in #tor-meeting2 at 1900 UTC on Tuesday.
* 0.3.4.x triage has begin!
== Discussion topics ==
* How did all the rotations go last week?
* We need to finish all the 033 tickets. How are we going to do this?
* We need to triage all the tickets in 034. Should we handle it the
same way we did 0.3.3.x? Can we share/divide responsibility somehow?
Tickets for 034 should be the ones (at least) that we organized at our
roadmap to be on 034. And yes we should triage it like 033.
* Let's discuss the 034-roadmap-proposed tickets?
I think this is related to the other 2 questions above (033 and 034)
and related to the roadmap check in. In short we need to review the
amount of work we have planed to be done by May (which I think we are
overcapacity but lets discuss that)
* Should we change our regression rule?
(Currently, we say that a regression is automatically 033-must or
034-must. But maybe this only makes sense for recent regressions? If
something broke in 0.2.9 and nobody has noticed it since, maybe it
isn't "must".)
== Updates ==
Nick:
Last week:
* Start PETS reviews
* Finished first round of ED screening
* Review and merge a lot
* Released 0.3.3.4-alpha
*
This week:
* Finish PETS reviews
* Review and merge a lot
* Review all relevant SoP applications
* Work on libevent mainloop-related refactoring for sponsor8
* Work on identifying triaged-in tickets for 0.3.4
* Maybe start on bucket-refill refactoring for sponsor8
teor:
Last week:
* Deployed research code (3.5 codebases, 3 machines, ~10 instances)
* Worked with juga and pastly on bwauth scaling
* Reviewed a Tor Summer of Privacy proposal
* Collected streams-per-circuit for arma's DoS limits
* Started a data collection for a paper
This week:
* Analyse streams-per-circuit and send an email
* Post-travel admin
* General employment admin / research admin
* Maybe collect streams-per-circuit with less noise
* Revise the PrivCount in Tor noise code after Isis' review
isabela:
Last week:
* finished FB proposal
* first meeting following new agenda format
This week:
* organizing pad for todays meeting
* 1:1:1 with folks
* prep work for sponsor8 Q1 2018 report (due end of april)
Mike:
last week:
- Reviewed tickets #21394, #25505, #25355; commented on #25347, #24456
- Wrote two guards proposal
- Started looking at #25545
- Emails about QUIC
this week:
- Discuss #25545 with asn
- Sit in on relay crypto meeting
- Work on vanguard roadmap tickets (#25546)
pastly:
Special thanks to juga for all the help so far, and hope you can continue
helping make Simple Bandwidth Scanner (sbws) a reality :)
last week:
* Make sbws (torflow replacement) into even more of a proper package.
Namely, give it a config file and it runs out of a data directory, not
the current working directory
* Write 95% of a specification for sbws
* Close many usability tickets
* Add version to client/server handshake and to result output
* Begin running sbws in testnet with a variety of configurations
* Choose a license, CC0
* SUBMIT PAPERWORK NECESSARY TO RELEASE SBWS AS AN OPEN SOURCE PROJECT
* SUBMIT PAPERWORK NECESSARY TO RELEASE SBWS AS AN OPEN SOURCE PROJECT
* SUBMIT PAPERWORK NECESSARY TO RELEASE SBWS AS AN OPEN SOURCE PROJECT
this week:
* Close some more UX tickets
* Hopefully compare sbws to torflow in the testnet, with help from teor
* Moar documentation!!!
next month:
* Get paper work back stating I can/cannot release sbws
asn:
[we should roadmap #25552 (rev counters hsv3) and #20212 (client-side
guard dicsovery)]
Last weeks:
- Made patch for the 033 ticket I took over last week: #25347. After
discussions with people, it seems like behind that ticket there are hidden
complex security-related decisions and we should be really careful on how
to resolve it. More discussion with at least Mike and Roger is required to
make sure we come up with something decent here.
- Worked on my vanguard simulator (#25545), polished the code, and
pushed it on github:
https://github.com/asn-d6/vanguard_simulator
There is still work to be done here so that the results are useful for
taking decisions, but publishing the code is an important first step.
- Did roadmapping on the Penn censorship project.
- Discussed switching to 2 entry guards with Mike (#25668).
- Reviewed #24989, #25024, #25432, #24767.
- Reviewed #25581 but it's still unclear what should happen here. I
think Mike's input is required.
- Finished community guide duty for this week
- Acked exarkun's patch on #25627
- Got filipo's patch merged in #24342.
- Discussed bw scanner work with juga.
- Merge_readied #25432 by fristonio.
- Helped researcher I met in RWC to use Chutney to test his patches.
This week:
- Bug triage duty.
- More work on 033 ticket #25347.
- More work with mike on 2-guard proposal and vanguard simulator.
- Meeting with David/Arlo on Thursday about Snowflake and the penn
censorship project.
- Reviews as needed.
ahf
Last week:
Sponsor 8:
- Talked with Nick about Idle API.
- Looked at DirAuth call graph for the directory server interface to
figure out where DirAuth specific code is called from.
Misc:
- Submitted patch for #24782.
- Fixed review comments for #25355.
- Bug triage rotation role.
- Worked together with Isis on review of #25425.
- Reviewed #23846 and couldn't find anything problematic there, except
for the Rust warning that Simone had seen himself.
- Out Thursday/Friday and Monday for easter with friends and family.
This week:
Sponsor 8:
- Talk with David (and maybe Nick?) about the DirAuth module project.
- Begin splitting things up for DirAuth module project.
- Discuss with the people from the guardian project some of
the things we discussed last week for the Idle interface.
Misc:
- Work on more 033 tasks.
dgoulet:
- Last week:
* Reviewed and worked on 033 tickets (most of them 033-must). See Trac
Timeline for more details there.
* Had a planning meeting with Nickm about roadmap item: Reduce client CPU
usage.
* Had a planning meeting with ahf about roadmap item modularization
which is extracting dirauth into its own module. (#25610)
* Investiguated #25616 which needs feedback about the possible solutions
(staring at asn :).
* Got #24767 merged so exciting. Thanks to the reviewers!
* Discussed with arma the #25226 and I got a review from him so \o/
- This week:
* Continue on 033-must tickets, I think I still have one to be merged.
* Roadmap items, mostly modularization. From which, would be nice if people
would read this small comment about the "plan" for dirauth code. If you
think some of it is crazy or we could do better, good time to say so
because after that we start in that direction.
https://trac.torproject.org/projects/tor/ticket/25610#comment:3
haxxpop:
- Last week
* Start implementing hsv3 client authorization (progress
https://github.com/torproject/tor/pull/36)
* It includes loading HiddenServiceAuthorizeClient in torrc to the service
* It also includes generating a new curve25519 key pair if we do
not have that client's pubkey yet.
Or just load the key if we already have that client's pubkey.
- This week
* Properly encrypt the descriptor if the client authorization is enabled
catalyst:
- Last week (2018-W13)
* fixed two Coverity defects (#25629, #25675) (weekly rotation)
* meeting on large create cells
* fixed some torspec typos and got commit rights in that repo
* did some follow up about the Rome cell crypto session -- meeting
1900 UTC today
* looked at some build system patches (#23846 -- libtool, #25549 -- Appveyor)
* looked into #25061 some more
* worked with isis on some configuration of our github org
* talked with people about the two github org issue
* experimented with improving my work space ergonomics
- This week (2018-W14)
* cell crypto meeting today
* community advocate rotation
* write up my thoughts on #25061 and try something
* code review
* investigate the deleted github branches that keep reappearing
* talk more with people about the two github org issue
* more ergonomics
juga:
- Last week:
* sbws: followed progress, talked with pastly and teor, add some docs
* funding: wrote and submitted application for F3
* worked on #25515
- Next week:
* sbws: solve some bugs, more docs
* work on another test for #25515
- Maybe we could have a bwauth meeting once a month? (is difficult
to find more than 1 or 2 persons to discuss in #tor-dev cause of
timings)
- Is it fine for all i switch (at least for now) to sbws instead of
bwscanner?,
- and that i integrate in it later some functionality that was in
Torflow/bwscanner?
- Does it makes sense i work on sbws?, or is it enough with 1 person?
[pastly: thinks it makes sense to focus on sbws instead of
bwscanner; also wonders
what torflow/bwscanner does that sbws doesn't do and should do]
[juga: ^^^ i'm not sure, we can discuss further...]
[isis: where is the code for sbws? may i see it?]
[pastly re isis: in a private GitHub repo, yes if you pinky
promise to keep it private for now]
isis:
last week:
- revised bridge.c tests #25425
- revised #24031 after two reviews
- did the patches for #25517, now needs revision after teor's review
- reviewed new appveyor configs #25549
- reported some potential windows bugs based on #25549
- reviewed build system patches for testing our C code from Rust #25386
- added more tests for C code that our Rust calls #25605
this week:
- probably exposing our RNG to Rust #24660
- revising my branch for wide create cell creation #24986
- revising my function pointery handshake composition thingie #24987
- finish the sha2 thing since it's also blocking the handshake
work #24659
- more prop#269 thinking/writing
More information about the tor-project
mailing list