16:59:25 <ahf> #startmeeting Network team meeting, 1st november 2021
16:59:25 <MeetBot> Meeting started Mon Nov  1 16:59:25 2021 UTC.  The chair is ahf. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:59:25 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:59:31 <nickm> hello!
16:59:31 <ahf> hello hello welcome
16:59:36 <juga> o/
16:59:39 <ahf> our pad is at: https://pad.riseup.net/p/tor-netteam-2021.1-keep
16:59:53 <ahf> it's the first meeting of the month so our agenda is a bit different than the other meetings
17:00:02 <eta> o/
17:00:33 <ahf> let's look at boards:
17:00:42 <ahf> https://gitlab.torproject.org/groups/tpo/core/-/boards?scope=all&assignee_username=YOURNAMEHERE
17:01:58 <dgoulet> clean here
17:02:03 <nickm> good by me.
17:02:32 <ahf> excellent
17:02:47 * eta still needs to fill hers with stuff
17:03:11 <ahf> eta: have you been introduced to how we use icebox, backlog, next, doing, etc. labels?
17:03:16 <eta> planning on doing so imminently though
17:03:23 <ahf> gitlab's board is just a fancy interface on top of the label system :-S
17:03:23 <eta> ahf, no, but it seems relatively self-explanatory (?)
17:03:29 <ahf> yeah, there is not much magic to it
17:03:36 <ahf> and i think the arti project is way more clean than tor.git
17:03:46 <ahf> so you probably have a higher crunch rate there and a smaller ticket-space to deal with
17:04:33 <ahf> ok
17:04:42 <mikeperry> I wanna close https://gitlab.torproject.org/tpo/core/tor/-/issues/40427 as wontfix (looking in my backlog on my board)
17:05:24 <mikeperry> I don't think it will help, maybe it hurts, and the filer reacted with reaction "timer emoji" when I asked about it
17:05:37 <ahf> lol, i was just looking at what the timer was
17:05:50 <nickm> mikeperry: they also edited the issue to explain why they think it's a good idea
17:06:09 <nickm> so probably double-check the new description
17:06:30 <mikeperry> yeah it is still not helpful
17:06:43 <ahf> https://gitlab.torproject.org/pseudonymisaTor they have a bunch of random projects here
17:06:45 <nickm> fine to close as wontfix IMO, just explain in a little more detail what's up
17:07:02 <ahf> guardsets in tor using as relationships sounds like something i looked at some years ago :o and congestion aware routing in tor sounds like the CC work now :-S
17:07:03 <ahf> funky
17:08:32 <ahf> yeah, wont fix is OK to close tickets at
17:08:34 <ahf> ok
17:08:56 <ahf> so for release stuff we got a ticket from weasel on the alpha release of 0.4.7 that we have some sandbox test failures on x86-32 in tor#40505
17:09:09 <ahf> i think i have solved the chown() case, but the stat() thing seems a bit more naughty to me
17:09:16 <ahf> but i think we can get a fix in for that before next release of 0.4.7
17:09:48 <nickm> I wonder if we can get an i386 builder on gitlab CI.
17:09:53 <ahf> we can
17:10:00 <ahf> i just reproduced this bug in i386 debian docker image
17:10:06 <ahf> so that will be part of the solution to this ticket
17:11:14 <nickm> ack
17:11:40 <ahf> i could use your brain on the stat part of this issue, but i also pinged Simon who seemed to be looking into some sandbox stuff recently
17:11:46 <ahf> and Simon did the tests for the sandbox
17:12:13 <ahf> ok, otherwise i don't see anything interesting re. releases
17:12:29 <ahf> https://gitlab.torproject.org/tpo/core/team/-/wikis/NetworkTeam/CoreTorReleases <- 0.4.7 stable is set to Dec 15 right now
17:12:30 <nickm> ahf: sounds good. probably the best solution is to just do it the stupid way: condition based on the architecture to allow stat64 vs newfstatat :/
17:12:42 <ahf> nickm: yeah
17:12:53 <nickm> That date doesn't seem likely if we're still hoping to get cc in.
17:13:04 <ahf> this is before newfstat tho, since i am on a glibc before that i think
17:13:15 <ahf> i am on glibc 2.31 and i think the change with stat() was in 2.33
17:13:28 <nickm> ahf: maybe you need to adjust the glibc version check.  We have code for that that gets used in the sandbox
17:13:33 <ahf> no, i also don't think the date is realistic, it will be a new year release
17:13:36 <nickm> ahf: (We usually need a couple solid months to stabilize after we are done merging.)
17:13:41 <ahf> ya
17:13:43 <ahf> for sure
17:13:59 <ahf> nickm: we can talk about the i386 bug during the week, it's no rush i think
17:14:03 <nickm> ok
17:14:34 <ahf> https://gitlab.torproject.org/tpo/core/tor/-/issues?scope=all&utf8=%E2%9C%93&state=opened&label_name[]=Backport <- we are down to two tickets with Backport now after the pre-release backport crunch
17:14:45 <ahf> nothing interesting there except the number i think
17:15:42 <ahf> (i assume people interrupt me if i say something wrong/bad/something worth commenting on 8)
17:15:59 <ahf> https://gitlab.torproject.org/tpo/core/team/-/issues <- we have no new team notes. we closed the eta onboarding one \o/
17:16:28 <ahf> with roger's hacking on one of the s30 items, i have one big s30 thing on my plate for this week, then i could dive into tpo/team#19 and probably try to get a volunteer to help with some of the web/css stuff
17:16:34 <ahf> ugh
17:16:36 <ahf> tpo/core/team#19
17:17:10 <ahf> we have no discussion items
17:17:15 <ahf> no announcements
17:17:23 <ahf> no s61 part of the meeting since that was in the bbb earlier today
17:17:40 <ahf> have everybody updated the pad with the exciting they have been doing in the last week? :-)
17:18:39 <eta> looks like nickm already wrote something about the arti release for me (thanks!)
17:18:57 <ahf> \o/
17:19:26 <ahf> shall we return to the interesting things we want to hack on instead of being in a meeting? :-)
17:19:35 <ahf> i might prod you all for some questions on the pad over the week, but shouldn't be much
17:20:20 <ahf> https://c.tenor.com/3Bro6ws6vSwAAAAC/quiet-desert.gif
17:20:23 <ahf> ok, let's call it!
17:20:25 <ahf> see you all
17:20:26 <ahf> #endmeeting