17:01:29 <nickm> #startmeeting network team meeting 6 May 17:01:29 <MeetBot> Meeting started Mon May 6 17:01:29 2019 UTC. The chair is nickm. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:01:29 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic. 17:01:36 <nickm> sorry I'm late! :) 17:01:37 <ahf> hey 17:01:48 <nickm> meeting pad at https://pad.riseup.net/p/tor-netteam-2019.1-keep 17:01:49 <catalyst> o/ 17:02:18 <mikeperry> hi. I'm a little late.. still loading pad 17:02:24 <dgoulet> hello 17:02:51 <asn> hello 17:02:55 <nickm> hi! 17:02:56 <asn> oops forgot the transition thing 17:03:21 <gaba> hi! 17:04:38 <nickm> so for our "stuff to do every week" list, do we have any pending CI failure issues? 17:04:55 <catalyst> do we still have the intermittent stem timeout? 17:05:08 <nickm> I think we might have a different one 17:05:16 <nickm> with similar symptoms 17:05:43 <nickm> asn: you were on CI+coverity last week; did you run into anything? 17:06:00 <asn> yes coverity had some false reports: https://trac.torproject.org/projects/tor/ticket/30342#trac-add-comment 17:06:11 <asn> i opened a ticket but quickly found that coverity just needed a synch 17:06:30 <asn> the synch occured, and then we got two new reports: #30361 17:06:38 <asn> that's it 17:06:40 <nickm> ack 17:06:51 <catalyst> asn: there was the triple-underscore typo but that's also resolved i think 17:06:58 <nickm> right, that was a stale build 17:07:09 <nickm> our coverity setup is wonky, because coverity is bad at certificates 17:07:17 <asn> catalyst: ye nick fixed that on april 20th or so. 17:07:38 <nickm> [a security company making certificate mistakes, who could imagine] 17:07:53 <catalyst> [overspecialization strikes again] 17:08:16 <ahf> :-S 17:08:45 <nickm> if anybody runs into the stem thing, please make sure there's a ticket? I tracked down the last one, but it was really hard 17:08:58 <nickm> next up is 0.4.0. 0.4.0 is released. Woo! 17:09:09 <ahf> \o/ 17:09:19 <dgoulet> yes great success 17:09:22 <nickm> 0.4.1 freeze is supposed to be 15 May, in 9 days from now. 17:09:45 <nickm> So let's take what we can in 0.4.1 and start stabilizing 0.4.1 17:10:15 <nickm> question: does our practice of not merging anything into 0.4.2 for a month after 0.4.1 is frozen make sense? 17:10:32 <nickm> (I'm talking about the thing where we close our merge window for a month.) 17:11:24 <ahf> for doing bug fixing? 17:11:25 <dgoulet> I like it personally because makes us stabilize the features released, not the the new fancy ones 17:11:36 <nickm> I think it only works if we make it work though 17:11:39 <asn> is our roadmap accounting for this month gap? 17:11:52 <asn> im afraid that we might end up leaving too much stuff unmerged on trac, and bitrotting 17:11:53 <nickm> if folks are working on features for the next release instead of stabilizing this one, everything breaks down 17:13:18 <nickm> There are other ways we can do this. We can open up 0.4.2 for merging _existing code only_, but not merge anything that didn't exist on 15 May. 17:13:23 <nickm> (For example) 17:13:40 <nickm> that way we avoid bitrot while also trying to channel our energy to fixes 17:13:41 <dgoulet> hmmm asn is making me think... 17:13:48 <nickm> One important deadline is S29 17:14:07 <gaba> 29? 17:14:12 <catalyst> also S19 at end of May 17:14:22 <nickm> arg, 19 17:14:39 <gaba> 19 at the end of may, 31 end of november... 17:14:59 <ahf> yeah 17:15:15 <nickm> are there things targetted for tor that are in s19 and not implemented yet? 17:15:36 <ahf> that is what we are going to look at after this meeting i think 17:15:39 <gaba> yes 17:15:56 <nickm> dgoulet, asn: Any other ideas on how to get stabilization on 0.4.1 while avoiding bitrot on pending patches? 17:15:57 <gaba> there are a lot of things in the roadmap related to s19. we will go through them after this meeting 17:16:00 <nickm> ok 17:16:09 <catalyst> nickm: from me, a few more bootstrap reporting things. mostly continued incremental improvement, not nearly as big a change as 0.4.0 17:16:16 <dgoulet> nickm: not really, honestly both ways have pros/cons so 17:16:18 <nickm> ok 17:16:29 <nickm> catalyst: when do you anticipate those becoming needs_review? 17:16:32 <nickm> (or are they now?) 17:17:15 <catalyst> nickm: the pubsub one will be ready for review this week. the others i'm less sure about, and i might want to prioritize tickets that i haven't yet opened 17:17:35 <nickm> ok 17:18:29 <asn> nickm: i'm not sure tbh what should happen 17:18:38 <asn> i think we must heavily prioritize 041-must (?) tickets durin this period 17:18:57 <asn> but do we know of enough bugfixes that need to happen to keep us occupied for a month? 17:19:06 <nickm> So as a general question, to see if we're on the same page, is anybody _not_ okay with prioritizing 0.4.1 fixes and S19 wrapup from 15 May through 31 May, and 0.4.1 fixes through 15 June? 17:19:32 <nickm> asn: I don't know! In the past, we have never gotten all the xxx-must stuff fixed before our official release date. 17:19:42 <nickm> but maybe that's because of bad priorities later on during stabilization? 17:20:05 <catalyst> nickm: so to be clear you're ok with me working on s19 stuff between May 15 and May 31? 17:20:10 <gaba> that prioritization means not doing some of the planned s27 or s31 17:20:11 <gaba> ? 17:20:30 <nickm> catalyst: Yes, since S19 does not exist after 31 may. 17:20:33 <nickm> there's no other time to do it 17:20:40 <nickm> (if we can't get it done by 15 may) 17:21:17 <nickm> gaba: It means that we might have to be slower on that stuff to get the more urgent (041 fix and s19) stuff done. 17:21:20 <gaba> +1 on focusing on s19 and maybe leaving on a side s31 in may but I think it would be important for asn and dgoulet to continue with the work on s27 17:21:24 <gaba> ok 17:21:27 <ahf> the general question: i think it sounds ok 17:21:45 <nickm> gaba: I am not proposing that anybody should stop work on all other sponsors: 17:22:00 <nickm> ... only that we should prioritize S19 and 041 bugs. 17:22:13 <gaba> sounds good 17:22:13 <nickm> If we can do that without a window freeze, that's fine. 17:22:22 <nickm> If we can do that with a window freeze, that's fine too (IMO) 17:23:07 <nickm> ok 17:23:17 <nickm> mikeperry: this okay with you? 17:23:38 <dgoulet> I'm ok with the general question from nickm 17:23:47 <nickm> meanwhile, everybody please load the kanban 17:24:17 <mikeperry> I think so 17:24:50 <mikeperry> (my immediate concern is getting the remaining sponsor2 items fixed in 041) 17:25:04 <nickm> mikeperry: I don't think that's landing this week. Let's target 0.4.2. 17:25:27 <nickm> asn: Am I wrong there? Are we much closer than I'd thought? 17:25:38 <asn> which task are we talking about? 17:25:45 <nickm> good question 17:25:51 <mikeperry> #28634 and children 17:26:34 <asn> so mainly #28780 17:26:44 <asn> because all the rest are in a finished form of sorts 17:26:49 <mikeperry> that is the highest risk one, I think, yeah 17:27:22 <asn> i think if we manage to have #28780 in needs_review and in a way that nickm is satisfied, and its before the 15th of may, i think it's fair game? 17:27:39 <asn> the rest is pretty much done 17:28:08 <nickm> ok 17:29:01 <nickm> let's try to figure out a working #28780 approach then 17:29:09 <nickm> on to the kanban? 17:29:41 <nickm> I'm moving my s31 stuff from "in progress" to "backlog", since I won't be workign on it this week 17:29:48 <nickm> (That's #29209 and #29218) 17:30:17 <nickm> I've added myself to #29024 , but only the part where we do chutney support for pluggable transports: i'm not planning to do snowflake support specfically. 17:30:38 <nickm> (Since that's S19) 17:30:43 <gaba> \o/ 17:31:07 <nickm> any other updates on the roadmap? 17:32:26 <nickm> if not, on to reviews! 17:32:30 <nickm> not too many this week I guess 17:33:16 <nickm> Does anybody have anything blocker-like there? 17:34:21 <ahf> i have none 17:35:03 <nickm> next up is rotations: ahf takes over triage, catalyst takes over ci + coverity 17:35:11 <nickm> any hand-off stuff there? 17:35:12 <ahf> yep 17:37:46 <nickm> okay, not hearing handoff issues... 17:37:55 <nickm> we're on to discussion, I believe. 17:38:00 <nickm> First though, one last reminder: 17:38:09 <nickm> BOOK YOUR STOCKHOLM TRAVEL IF YOU HAVE NOT ALREADY. 17:38:36 <nickm> (get a price for your flight, send email to travel@, get approval, book it.) 17:38:51 <nickm> The rate lock deadline is Friday 17:39:32 <nickm> And I know Isabela is really serious about this. So please, let's be the team that gets this right. :) 17:39:43 <ahf> :-D 17:39:48 <nickm> Any discussion topics for this week? 17:40:01 <gaba> Remember to fill in the doodle for monthly retrospective https://doodle.com/poll/8xhmkm3zqmzdg8pe 17:40:23 <nickm> looking at boldface I see asn is telling mikeperry about an issue he had with #29034. 17:40:33 <nickm> I need to keep talking about #28780 with mikeperry and asn 17:42:04 <mikeperry> oh shoot 17:42:10 <mikeperry> typo in that byug bumber 17:42:21 <asn> im not gonna be around after this meeting, but im up for it this week. 17:42:32 <asn> i also left a comment in the ticket today suggesting invariants we can check 17:42:35 <nickm> asn, dgoulet: I've been following along with #14398 discussion, let's talk a moment to talk about it for a bit this week too 17:42:50 <nickm> err, 17:42:53 <nickm> #14389 17:43:43 <nickm> asn: thanks, I owe you a read on that. 17:44:02 <nickm> anything else for this week? If not, let's call the meeting done, and forge ahead! 17:45:05 <nickm> ok, thanks everybody! 17:45:06 <nickm> #endmeeting