[tor-project] Network team meeting notes, 3 July 2017

Nick Mathewson nickm at torproject.org
Tue Jul 4 00:31:12 UTC 2017

Hi; we had our regular meeting.  Logs here:


And updates below:

Network team pad, for 3 July Meeting (or 4 July, for those in UTC+5 or later)

Notes from last week's meeting:
    * https://lists.torproject.org/pipermail/tor-project/2017-June/001217.html
      (Did we do what we had planned?)

Announcements and reminders:
    * review-group-19 ?
    * There are still ? tickets in 0.3.1.x, including ? in status
"new" with no owner. We need to get 0.3.1.x done!
    * Timesheets and expenses today, please!!!
    * Remember to take slots in the rotation:
    * Reminder - isabela will be off the rest of week

Things we should talk about:
    * I want a Windows tor developer for Christmas, but I think we'll
get mobile first -teor
    * TROVE-2017-007 is openbsd only; need backport advice. Could use
a backport team in general. -nickm
    * We should talk about regression tests, timing, etc -nickm

Tasks for after the meeting:

teor (not online):
    Last week:
        * Started collecting entry and exit stats with PrivCount 1.0.2
        * Wrote draft PrivCount Tor code to count middle circuits and cells
        * Helped out with tor security fixes
        * Tried to help out with relays stuck with low measurements
#22741 #22798 #22744
        * Reviewed some of the circuit padding code
    This week:
        * Implement and test rendezvous circuit counts in PrivCount
          (We'll find out the rough usage of Tor2web and Single Onion Services)

  [Will not be home today during the meeting times.]
  Last week:
   - Did lots of reviewing on #21979. Still not done but getting
there. Comment posted.
   - Fixed #22006 based on nick/isis review and now it's merged upstream.
   - Addressed coverity issue #22746.
   - Reviewed #15554, #22726, #22727.
   - Continued prop247 simulation. Started making first graphs.
Adversary models are still pretty rough.
   - Opened #22735 and #22736 about prop224 and desc overlap period.
   - Wrote Wilmington blog post. Still unpublished.
  This week:
   - Finish #21979 review. Start next service-side review.
   - Continue prop247 simulation.
   - Publish wilmington blog post.

 Last week:
     * Security updates for TROVE-2017-006; new releases for 0.3.0 and 0.3.1.
     * Review, triage, merge, misc bugs.
     * Fuzzed tor_inet_pton, to no avail, since I was on linux.
     * Hardware issues ate a bit of my time.
     * Learned about harvest some.
 This week:
     * More 0.3.1 bugfix work. :p
     * Isabela sent me the time allocation spreadsheet betwen now and
end-of-yearish. I should get it out to everyone today!
     * Prioritizing code review, then sponsor8 preliminaries and other
code triage for 0.3.2
     * It would be nice to program something.
     * Review time-allocation thing with everyone
     * Figure out whether I'm supposed to be reviewing timesheets?

Last week:
    Sponsor 4:
        - Landed fix for #22702 and #22751.
        - Reviewed: #22752, #22719, #22797
        - Started a measurement for Sponsor 4 with a Tor compiled without LZMA
          and Zstandard support.
        - Started to land updates to dir-spec.txt for prop #278 in
          GL/ahf/torspec/tree/prop278-in-dir-spec (#22275)
        - Wrapped my head around our release process. Tried to help Nick out
          with the release for 0.3.0.x and 0.3.1.x.
          - Testing of releases on OS X, FreeBSD, and Linux.
        - Switched from text file notes for the week to Harvest. Seems
good so far.
        - Did everything needed before the end of the fiscial year (I
hope! Items are "pending approval").
This week:
    - Help out with 0.3.1 issues:
      - Go over the Sponsor 4 measurement issues and figure out: which is
        solved by the current knowledge and data we have, which can be solved
        easily with some modifications, and figure out if there is any of them
        that requires any modifications to Tor that we don't currently have.
    - Look into the mobile proposal email from Isa.

    last week:
        - tried to understand fuzzing framework, but i still don't
understand it much
    this week:
        - I will try to continue doing fuzzing

    Last week:
        * bug triage
        * Updated TorOnTrac keywords from pad
        * #22745 Started regression testing discussion
        * #22762 reviewed komlo's edits (coding standards make check
        * #22773 stem fix for random nicknames on 64-bit platforms
        * [sponsorM] Tor Launcher UX meeting
        * Reproduced #20532 pluggable transport bypass; can anyone
else replicate?
        * Opened #22743 for ed25519 point validation documentation
    This week:
        * Nail down which days I'm taking off this month for moving
and notify people
        * Still some stem issues with lack of system tor
        * Look more at 0.3.1 tickets


last week: PrivCount

this week: rebase + squash KIST in a reasonable way

  Last week:
    - Put #21979 in needs_review. Big step towards prop224 service merge.
    - Couple of prop224 groundwork tickets have been filed which are in
    - Worked on #17242 specifically the client introduction code. Out of this
      work, #22781 was borned and I finalized that branch this morning.
    - A discussion with asn on engineering details of prop224 including the
      overlap period and service management of introduction point.
    - Review some tickets in the review group 19.
  This week:
    - Continue forward with service branch merge/review
    - Keep going on client prop224 implementation (#17242).
    - I'm on triage/coverity duty this week.
    We are a month away from our soft deadline to have a working prototype
    which is not that much so expects LOTS of prop224 on my side in the coming

    Last week:
        - Worked on #22762, will finish this week
        - Had to deal with a fair amount of work/personal things so
focused on that.
    This week:
        - Writing a follow up to the Rust team about our next steps,
what we need to work out on our end, any ways they can help
        - Adding logging to Rust protover
        - Wondering what is the best way to mock in Rust for testing,
it would be nice to learn if Servo has any good practices
        - Look at #22745

    last week:
        - sent sponsor4 report - sent nick time distribution
spreadsheet - met f2f with roger
    this week:
        - coordination of nsf reports today and off the rest of the week

More information about the tor-project mailing list