Hi!
Our notes from the meeting follow below:
===================================
= Network team meeting pad, 16 April 2018 =
- "That's my name, don't wear it out!"
Welcome to our meeting! Mondays at 1700 UTC on #tor-meeting on OFTC.
(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!
for background.)
== Previous notes ==
== Stuff to do every week =* Let's check and update the roadmap. What's done, and what's coming up?
* Check reviewer assignments at
== Announcements ==
* 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 was supposed to be ready!
* May 15, 2018 -- 0.3.4.x feature freeze!
* Remember: don't spend more than a day working on anything that isn't on the 033 or 034 milestones.
== Discussion ==
[asn: Seems like we are having a hard time deciding on the intricate tradeoffs of prop#291. Doing it over [tor-dev] makes it even harder due to the bandwidth and the latency. Two suggestions:
a) Let's have a meeting next week about it. How about Wednesday 17:00 UTC?]
b) Let's leave it for the Seattle hackfest.]
(Decision: We're going with Wednesday, 1700 UTC.)
[mike: Jenkins hook changes+role documentation]
== Updates ==
Nick:
* Last week:
* Fixed 25691 and possibly 25692 (assertion failure bugs in 033)
* Worked on compilation failures affecting Jenkins, especially on 32-bit
* Various patches to libevent subsystems merged; new backend for token buckets merged. (Thanks for the reviews and discussion, David!)
* Inbox zero!
* Merged various fixes in 033 and 034.
* 0.3.3.5-rc released.
* Planned #25500 with dgoulet
* This week:
* CI rotation!
* Hoping to finish #25373 coding (no more every-100-ms timers for token buckets)
* Keeping an eye out for any bugs reported in 0.3.3.5-rc.
Mike:
* Last week:
* Did a bunch of paperwork; had to help some friends move a bunch of stuff. Took longer than I expected
* Discussed prop#291 on list
* Reviewed + edited #25248
* Did jenkins triage, somewhat poorly. Have some thoughts:
* Can we add a commit hook to mail the committer(s) since last build (and maybe a weekly/group triage alias?) when a build failes?
* Several of our builds are failing because of missing/broken platform stuff, I think. I guess I should be filing Service:Jenkins for these?
* We should probably tag all jenkins tickets with jenkins, so for long-time failing builds we don't dup/waste effort..
* We should document all of this like reviewing stuff.
* This week:
* Down to meet re prop #291+vanguards (prop#291 decisions impact how we do children of #25546)
* Work on vanguard script
catalyst:
* Last week (2018-W15):
- bug triage rotation
- #25756 clock skew complaints because dizum (dirauth) is 65 seconds in the future; do we want to relax the "consensus is coming from the future" tolerance to account for the earliest that a clock-skewed dirauth could be skewed yet still be distributing a valid consensus? e.g., by taking the voting schedule into account so "enough" dirauths would have to have clocks actually in the future for a client to produce a consensus-informed clock skew warning? [outcome: roger suggests "yes, 5 minutes is good." Nick seconds.][catalyst will make a ticket]
- worked on #25061, mostly tracing and annotating call trees
- tried out the new calltool to help with #25061; it doesn't play nicely with separate build dirs yet
- adventures in taxes and medical billing
* This week (2018-W16):
- dealing with winter storm aftermath (nothing too serious, only annoying)
- #25061
- 033 work and reviews as needed
- follow up on #25511 (minor revisions to test probably needed)
isis:
last week:
- refactored the randomness functions from crypto.c into a new crypto_rand.c module #24658
- wrapped our (P)RNG in Rust #24660
- began writing the encoding/decoding code for wide create cells #25647
- began writing FFI code for ed25519-dalek and x25519-dalek #23886
this week:
- finishing encoding/decoding wide create cells #25647
- ??? more wide create cell stuff? maybe #25648 and/or #25649 and/or #25653
- whatever else needs to do for #25517 (i think this is my only 033 task remaining)
dgoulet:
* Last week:
- Tickets work (including 033), see timeline for more info.
- Worked with nickm on the token bucket new API.
- Worked on the reducing client CPU usage with #25762.
- We merged #25226 this morning so great success!
- Some Bad Relays attention as well including the tor-team@ email about
some relay guidelines.
* This week:
- I'm done with 033 tickets so I'll go mostly in Roadmap items land.
Mostly #25762 (client CPU) and #25610 (modularization).
- I would like to work on #25761 HS regression before 033 stable!
- I'm the Bug Triager this week!
ahf
I'm gonna be missing during the meeting, but will read backlog once I'm
home again. See #tor-internal.
Last week:
Sponsor 8:
- Experimenting with GNU cflow on caller/callee relationship in
dirserv/dirauth code.
Misc:
- Began review of #20522 and #23846
- Code landed for #24782 and #24854
- Did an interview about Tor with the big Danish IT news site
- Talked a bit with Antonela about having TeX "templates" of the
different designs the UX team comes up with.
This week:
Sponsor 8:
- Move to 0.3.4 features.
Misc:
- Finish #25245
- Review #25140
- Figure out what to do with travel for Mozilla all hands.
haxxpop:
Last week:
- The descriptor is already encrypted using clients' pubkey
- I started working on the client side of the onion service client authorization
- Succesfully load HidServAuth config line to the daemon at the client side
This week:
- Try to decrypt the descriptor at the client side
Isabela:
- Helping with movig things forward related to the censorship team project
- Working with tommy on .onion grant proposal (deadline May 1st)
- FYI .onion ux ticket #23247 is being implemented by TB team :)
- ahf - reminder about new .onion address for the test of this month (april)
pastly:
/me is all about that sbws
last week:
- Compared sbws v3bw files to those from the real bwauths. Discovered
that my helper relay's bandwidth is a limiting factor in measurements
- Write integration tests
- Make one minor modification needed for IPv6 compatibility
- Merge code style guide
- Merge tests for code style and static ananlysis
- Change logging library from home-grown PastlyLogger to standard python
logging library (which is way better)
this week:
- Exercise sbws after logging changes to weed out bugs
- Write/merge additional docs
- Setup diratuh in test net, ask for other dirauths to trust it
- Other misc. stuff in 1.0 milestone
future:
- ~2 weeks get sbws paper work back saying it can be open source
asn:
Last week:
- Spent lots of time thinking and discussing the 2-guard proposal (#25754):
- Investigated engineering part of switching to two guards:
- Raised some concerns about the threat model of prop#291:
- Found a guard bug that might hit us if we reduce the number of our
primary guards (like prop#291 suggests): #25783
- Reviewed #25705, #25024, #23247.
- Prepared patch for #24544 and got it merged.
- Did snowflake meeting on Thursday.
This week:
- Continue work on 034 deliverables (2-guards, vanguards, etc.).
- More reviews.