[tor-project] Network Team Meeting Notes, 23 September 2019
Alexander Færøy
ahf at torproject.org
Tue Sep 24 03:23:13 UTC 2019
Hello,
Here is a short summary of the network team meeting today:
1) We went over our Kanban board to ensure that it is up-to-date.
2) We went over the set of code reviews that needs to happen this week.
Everyone should have gotten around 2.5 tickets in average according to asn.
3) We looked at the list of tickets for the upcoming 0.4.2 Tor release
on Trac. This is a follow up item from the meeting we held last week.
We decided that Gaba and ahf distributes the remaining 042 tickets
amongst the team members and discuss the distribution next week if,
for example, anybody wants to swap tickets with somebody else.
The goal for 0.4.2 is to have an RC with no known bugs ready for
around the 15th of November. Alpha releases will be coming out over
the next couple of weeks as the team continues to eliminate bugs.
4) Continuation of the discussion last week around "What does an ideal
PR looks like?"
We continued to discuss this item and according to the network team
meta-policy any member of the team can object to a policy. We decided
to not continue with turning the current proposal into a team
policy as some team members expressed enough concern to be willing
to object to it. Mike expressed interest in working on an alternative
proposal that we can discuss next week. The team decided this was our
actionable item to this matter for now.
5) We discussed moving our retrospective next week. The retrospective
is an internal network team discussions where we meet over a voice
communication service and discuss the work that have happened in the
team over the past month.
We decided to move the meeting to the 1st of October at 20:00 UTC.
6) We discussed how to handle code reviews that are requested over the
cause of the week. The network team have two volunteers who are
distributing the code review load each Monday, but sometimes people
have code they want reviewed during the cause of the week.
We decided that people should either reach out to each other directly
or use the "network-team" label in #tor-dev. If nobody responds,
people can ask ahf to help them find somebody who can review their
patches.
Please remember that our next meeting is on Wednesday the 2nd of October at
23:00 UTC!
--- end of summary ---
You can read today's network team meeting log at:
http://meetbot.debian.net/tor-meeting/2019/tor-meeting.2019-09-23-16.59.html
Below are the contents of our meeting pad:
gaba: (updated on september 23rd)
Last week (actual):
. scalability funding proposal
. other small tasks
. gitlab migration
This week (planned):
. sponsor 30 project
. gitlab migration
. sponsor 31 follow up
. 1:1:1s
Help with:
teor: (online first week of the month, offline at the usual meeting time)
Week of 16 September (planned):
Leave:
- May need to take some leave for the climate strike on 20 September
Urgent:
-
Take Time for:
- gitlab signup / config
- team policies:
- commit bit and roles?
- finish add tor controller trace logging to diagnose stem hangs (#30901)
- try nickm's suggested a simpler implementation
- split off bugs I found while writing the control trace code
Roadmap:
- Sponsor 31 possible coding tasks for me
Other:
Week of 16 September (actual):
Leave:
- Short week for climate strike
Urgent:
-
Take Time for:
-
Roadmap:
- Sponsor 31 config refactor
- #31338 practracker bug
- bug smash on logging bugs #31594 and children (maybe some of these are bugs on other sponsored work)
- bug smash on #31408 - config include dir bug
- help with onion balance v3
Other:
- bug triage
- backports
- Climate Strike blog post & comments
- try to recover failing dev environment a few times, then rebuilt a new one
- review gitlab transition document
Week of 23 September (planned):
Urgent:
-
Take Time for:
- gitlab signup / config
- team policies:
- faster reviews
- commit bit and roles?
- finish add tor controller trace logging to diagnose stem hangs (#30901)
- try nickm's suggested a simpler implementation
- split off bugs I found while writing the control trace code
Roadmap:
- Sponsor 31 outstanding bugs / tasks
- Sponsor 31 possible coding tasks for me - follow up email
Other:
- finish off bug smash logging bugs
Nick:
Week of 16 September (planned):
- Release 0.4.1.6
- release 0.4.2.1-alpha
- Start 0.4.2 stabilization process
- Read proposal 308 (counter galois onion)
- Discuss C style polling with Teor and Taylor
- Finish PETS review
- Work on circuitpadding doc stuff?
- Pending reviews
Week of 16 September (actual):
- Released 0.4.1.6
- Released 0.4.2.1-alpha
- Sorted 0.4.2.x tickets that were not must/should/can into Unspecified and 0.4.3.x. I used the keyword 042-deferred-20190918 to mark them.
- Squashed/answered a bunch of 042-must/should bugs:
- 31107: unexpected versions cell
- 31483: single-bucked issues (analyzed and deferred)
- 31769: -Wextra-semi failures (seems to be ephemeral, analyzed)
- 31338: --list-overbroad issues
- 31759: line-width on annotate_ifdef_directives
- 30916: assertion failure in dimap_add_entry()
- 31466: rate-limiting ".exit is disabled" messge
- PETS review
- Started but did not finish circuitpadding doc edits
- Answered emails, got to inbox 0.
- Read proposal 308; seems well-formed but it'll need a reference implementation
- Merges and reviews
- Read new employee manual
Week of 23 September (planned):
- Keep an eye on blog comments
- Finish edding circuit-padding document
- Start a C style poll for the team
- Meet about sponsor 31 on tuesday
- attend ED tea time later Monday
- Respond if needed to expected reviews on paper with Ian and Chelsea
- Work on 042-must/should items
- Review and merge
Mike:
Week of 9/16 (planned):
- Funding proposals
- Circpad doc tweaks and ticket #s -> git & trac
- Patch for #31653 (0ms circuitpadding circuit failure bug)
Week of 9/09 (actual):
- Funding proposals
- Circpad doc tweaks and ticket #s -> git & trac
- Patch for #31653 (0ms circuitpadding circuit failure bug)
- Funding proposals
- Research discussion
Week of 9/23 (planned):
- Research discussion
- Work with Nick to clean up final XXX's in circpad doc and merge/cite the quickstart guide
- Cloudflare bot fight?
- Maybe have time to start on Firefox review? <-- gaba: please or let pili know that you can't do it this week -->
Need help with/at risk of dropping this month: <-- do you still need help with this?
- Not doing any code reviews this month, unless on circpad patches. Got enough for Tor Browser
- Circpad doc followup/fold-in? (nickm is helping here, thx!)
- Firefox ESR code review
- (I am working on this, but unlikely to finish it this month at this rate of interruption+distraction)
- Firefox new feature review (Geko might be taking this, it sounds like. Thx!)
- Funding proposal review, meetings, coordination, cross-checking estimations, strategy, etc.
- Deep-thought-required research project followup
- (Google masque, BGP, ECN, Rob's bw experiments, Dennis's Mozilla video, etc etc...)
- Relay community drive/mgmt (and related LTS herding)
catalyst:
week of 09/16 (2019-W38) (planned):
- reviews
- coordinate sponsor31 documentation work
- respond to C style thread
- respond to review guidelines thread
- review gitlab migration plans
- #30984
week of 09/16 (2019-W38) (actual):
- reviews
- provided input to grant writing stuff
- caught up on review guidelines thread
- updated some CI failure notes
week of 09/23 (2019-W39) (planned):
- reviews
- provide feedback about C style
- more feedback to review guidelines thread as needed
- more feedback re gitlab migration as needed
- check in with swati on Summer of Docs project
- #30984
asn:
Week of 16/09 (planned):
- Get close to finalizing #31648. This involves cross-testing the Python
implementation of the HSv3 hashring with the little-t-tor one.
- Maybe start working on HSv3 public key blinding using the hazmat crypto
library in stem.
Week of 16/09 (actual):
- Pretty much done with #31648. I have unittest vectors ready for
little-t-tor, and I cross-test my Python implementation with them. Only
finishing touches need to be done (a few bugfixes, code docs, etc.).
- Start looking at HSv3 public key blinding for onionbalance. This seems
harder than anticipated due to the hazard crypto lib not exposing the
necessary primitives, but I need to look deeper into this and possible
solutions.
- Figured out a more precise (personal) roadmap for onionbalance and related
s27 work.
- Lots of reviews and merges.
- Talked with atagar about progress in #31369 and #31823.
Week of 23/09 (planned):
- Figure out how to do key blinding in onionbalance.
- Forward progress in #31369 now that atagar has given me feedback
ahf:
Week of 16th of September (planned):
- Follow up on "Good PR" discussion on network-team@ and turn it into a proposal
- Keep the discussions on GL migration active and hopefully continue to make progress.
- Finish #5304
- Figure out my task plan for the rest of the year'ish for tickets.
Week of 16th of September (actually):
- Been trying to come up with a solution to the anonymous users/moderation issue with Gitlab migration.
- Came up with a solution to #31810 that I think wont need any spec changes and should solve it.
- We had a very produtive Gitlab migration meeting where we identified some new ways of doing ticket migration that might be smarter for us.
- Rebased changes from #28930 to head, started refactoring some of the common code that Tim had found.
- Continuing to wrap my head around the new role and zoom out a bit of my former role.
Week of 23rd of September (planned):
- Do first round of 1:1:1 with Gaba and other people on the team.
- Work on #31810
- Continue to work on #28930
- Friday is Gitlab day where I will continue with the migration project.
dgoulet: - on vacation
Week of 02/09 (actual):
- Most of my time has been on #30200 parent tasks working on building and
running an onion service health tool (#28841).
- From the above, several tickets have been opened, see the child tickets
of #30200.
- I've mostly worked on the service side health. I've moved now on writing
the tool analysis part for the client side.
- One ticket I opened which was needed for my development branch in #30200
that affects the wider tor (and found a bug): #31608. It is good that we
are all awayre of this imo.
Week of 09/09 (planned):
- Continue making sure #31549 is on track and doesn't stall.
- Help write blog post about the above ^ in order to be as loud as we can
about this big change in the network.
- Continue on #30200 to dig into client side behavior (s27 stuff).
swati (Updated on September 16)
What I did last week:
- Spent time learning how to use Asciidoc.
- Set up the Atom Editor by installing plugins to be used for
editing/writing Asciidoc documents.
- Sent my questions regarding cloning/forking and the process followed. Gaba and Taylor answered my queries.
- Forked and cloned the repo: https://github.com/torproject/tor/
- Saved a local copy of Tor Manual (doc/tor.1.txt) on my machine.
- Added a Table of Contents and regrouped topics to create a better flow and structure.
What I plan to do this week:
- Continue working on the tasks planned for month 1 (https://developers.google.com/season-of-docs/docs/participants/project-tor)
- Create a pull request and request the network team to review the first set of changes.
Help with:
- Might have questions about the categorization of configuration options and might need info about config files.
- Need to get an account to use Trac (https://trac.torproject.org/) for creating tickets for pull requests. (gaba looking at it)
All the best,
Alex.
--
Alexander Færøy
More information about the tor-project
mailing list