Network team meeting pad, 6 November 2017
Welcome to our meeting! Every monday 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.)
Meeting notes from last week:
Announcements:
* Please do not forget about 0.3.2! How to get it stable?
Discussion topics:
- [dgoulet: "Hello71" on IRC has been profiling and chasing memleaks lately, very useful imo. Also, someone submitted a patch to torsocks to fix it with Rust program #24116]
* Still a few hackerone reports that need to be triaged by the fuzzer guy.
* Did we finish everything from october?
Updates: (Please use boldface to indicate discussion topics)
dgoulet:
[might be offline during the meeting, we'll see!]
Last week:
- HSv3 tickets in the 032 milestone.
- Review tickets in the 032 milestone.
- Worked on the prop224 control port specification which I just sent on
- Worked on a tor-dev@ thread about the connection, channel and scheduler
- From that thread ^, the improved scheduler interface came out for #23993
but not yet public as I'm waiting on possibly more feedback on the
thread before going further.
- We had a series of bad-relays last week that needed some investigation so
I had to put some times in the health of the network side of things.
- Profiling of 032 relay: #24127
This week:
- Continue the roadmap tasks I have that is the HS circuit timoue issues
and scheduler work.
- Prop224 control port work aimed at 033 so I want to start the review as
soon as possible for an "early" merge.
- 032 ticket review and address my tickets as well in that milestone.
asn
[Might be a bit late for the mtng but will be there]
Last week:
- Lots of microdescriptor work. Did some debugging and some bugfixing. Opened
#23985 and #24113 and #23989.
- Mainly worked on #23817: an important md issue. Pushing branch tonight. [NM: how complex will it be?]
- Also published prop224 blog post. We have now officially launched. Tracked
the blog comments and the hackernews discussion.
- Did some digging into the guard discovery stuff but not too much.
- Some prop224 reviewing and ticket handling.
This week:
- More microdescriptor fixing. This is the highest priority for me rn.
- Catching up with prop224 work and maybe guard discovery if time permits.
- Will learn more info about my future phd and my future schedule.
nickm:
Last week:
- Worked on revised (Shamir-based) privcount design
- Much review & merge
- Roadmapping coordination with other teams
- Reviewed prop283 (ipv6 address into consensus for Teor)
- Dealt with fuzzing false-positives from oss-fuzz; found one real memory leak (#24150)
-
This week:
- Several proposals, including:
- Revised privcount data collector behavior (#22955)
- Improved APIs for control on mobile/embedded systems (#24107, #24108)
- Ed25519 ID -> Consensus proposal (#23170, plus #24103)
- Release another 0.3.2 alpha on Tuesday or Wednesday, ideally including the following. (are they yours?):
- #23816 Use exponential backoff with jitter and/or tune its parameters
- #24150 Memory leak in v3 hsdesc parsing with empty encrypted data.
- #23414 rep_hist_format_hs_stats() should add noise, then round
- #23623 hs: Cache current time period number and SRV start time
- Review dgoulet's big scheduler/channel/connection email, time permitting
- Meeting w Mozilla people about Tor integeration
- Bug people about the 5 remaining tickets in review-group-24
- Absent much of Tuesday & Friday
ahf:
Last week:
Sponsor 8:
- Wrote a patch to use Android's logging subsystem directly in Tor (ahf/android-logging on GL.com).
- Looked into the syslog() inteface on Android as well and whether it works with adb's logging features too
on all devices.
- Created tickets for the performance and measurement work as well as controller interface bugs for
battery/bandwidth control.
- Wrote an introduction to Google's simpleperf tool for Tor:
(if you want to play around with the tool without doing all the device stuff you can download the sample file
- Started documenting `systrace` and `dumpsys` to get memory and battery information out of the device.
Misc:
- Started looking into Shawn's capsicum sandbox patches.
This week:
Sponsor 8:
- Finish docuumenting and testing, ideally get someone else to try it out too.
- Start working on #24065.
- If I have time: setup an fdroid repository with "daily" builds of Orbot with tor.git's master version of Tor.
Misc:
- See if there is anything I need to prepare for the OTF meeting next week. Follow up on Isa's email on the
topic.
catalyst:
Last week (2017-W44):
- HackerOne triage
- helped with bwauth trouble with control port oddities (#24110, #24128)
- reviewing #23816. download_status code is very hard to read. updated #23354 and filed 24152 in hopes we can improve this
- updated team wiki page with meetin time changes for DST stuff
This week (2017-W45):
- more jittered backoff review work (#23816)
- review other tickets (#23523, maybe others)
- try to coordinate more with others about HackerOne triage
- investigate bootstrap hangs (mostly clock skew related?) some more
- sponsor8 bookkeeping (ticket tagging, etc.)
Isabela:
Last week:
- worked on sponsor8 Q3 report (august and september)
- organized first meeting for teams to share their roadmaps
- pinged testers to test our network speed
This week:
- submit sponsor8 Q3 report (august and september)
- Tor Launcher work check in - moat implementation, check in with Browser team too, progress bar msgs. [check-in note for: nick, catalyst, isis]
- hope to get ux team with a roadmap to share soon - will share all roadmaps and its dicussions on tor-project ml
haxxpop:
This week:
- I will look at ESTRABLISH_INTRO, INTRODUCE1, and INTRODUCE2 and see if any of them is worth fuzzing
isis:
Last week:
- reviewed #22342, a new buf-to-string api
- worked mostly on #15967 and #22871 (moat stuff), which is coming along quite well but still has a couple kinks in it
- deployed new bridgedb so that bridge-distribution-request lines are respected
- planning/travel for mozilla all-hands in december
- got setup in hackerone
This week:
- reimbursements for montréal
- #22871 bugs and deploying a preliminary setup for testing
- wanted to talk with catalyst about writing down how we plan to use the github account, but other things might have to take priority this week
Mike:
- - Getting back up to speed on things and trying to normalize sleep schedule
- - Writing tests for #23114 and #23100
This week:
- Really trying to finish #23114 and #23100
- Paperwork, airfare, expense reports, time sheets for past couple months