16:59:53 #startmeeting weekly network-team meeting, 27 august 16:59:53 Meeting started Mon Aug 27 16:59:53 2018 UTC. The chair is nickm. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:59:53 Useful Commands: #action #agreed #help #info #idea #link #topic. 17:00:07 it's that time again! 17:00:21 the pad is still at https://pad.riseup.net/p/tor-netteam-2018.1-keep 17:00:22 hi 17:00:50 hey! 17:00:56 hello 17:01:08 ehlo 17:02:07 hello i am kind-of-around if people need things from me 17:02:59 reminders: we'-re still focusing on sponsored work, and within that we're focusing on sponsor8 17:03:06 * ahf nods 17:03:14 0.3.4.7-rc is out; with luck we won't find any tricky bugs that will require another rc 17:03:25 ! 17:03:51 let's have another look at the roadmap and ask ourselves -- are all of these things on track for the 035 feature freeze? 17:04:07 If not, please please let me and isabela know 17:04:59 reviewer assignments are also on the pad. looks like there are a nontrivial number this week. 17:05:09 reviwer assignmnets not 100% done yet 17:05:13 sorry about that 17:05:19 we are still load balancing 17:05:20 right dgoulet ? 17:05:36 yes 17:05:53 but mostly between me and asn right now so the rest should be what is expected 17:06:50 You need to move #26818; catalyst is swamped, and that branch is very big. 17:07:24 yess we are load balancing the NSS stuff 17:07:31 ahf: were you still interested in reviewing the NSS patches? 17:07:38 yep! 17:07:53 i think i'm already marked as reviewer on one or more of them 17:08:06 ahf: ok i'll hand off my state to you later today, then 17:08:33 ahf: how are you in terms of load? 17:08:40 ahf: would reviewing NSS stuff make you overloaded? 17:08:50 maybe we can split the NSS stuff between the two of us 17:08:59 you take the RSA one, I take the TLS one, or the opposite. 17:09:03 yeah, that might be a good idea, i don't have much overview of how big the task is right now 17:09:06 there are two NSS branch that are quite big... 17:09:12 ok, then let's split them 17:09:16 ok great 17:09:31 catalyst: any suggestions on how to split the NSS tickets? :) 17:09:37 have you seen the TLS one, or just the RSA one? 17:09:38 i am going over my old review queue as well today, hope to look at the NSS work tomorrow, then back to the s8 coding 17:10:04 asn: i've looked at the DH one, but not yet anything later than that 17:10:06 which one is bigger? o r are they approx the same? 17:10:33 there is a TLS, RSA, DH and ... hashing one? 17:10:41 i haven't looked at RSA or TLS yet 17:10:43 for review this week, there is TLS and RSA 17:10:46 ack 17:10:48 nickm: are they approx the same size? 17:10:54 the hashing one was included in DH 17:10:59 ahf: you have a preference for TLS or RSA? 17:11:13 i think i might prefer RSA just because it sounds more crypto, but I don't mind either 17:11:34 i'm fine with either 17:11:41 the TLS one is based on the RSA one; it's possible to treat it all as one big branch 17:11:44 I think it is starting to look likely that wtf-pad may miss 035 btw. I'm going to keep working on it since having something that can be discussed, used, tested, and experimented on before Mexico City will be v useful 17:11:58 but it's not so big a deal that it get into 035, I don't think 17:12:15 mikeperry: ok. could you note that on the spreadsheet ? 17:12:20 ahf: ok you take TLS, I take RSA? and we revise this week if one is too big? 17:12:26 yes 17:12:28 let's do that 17:12:30 ahf: basically, you can assume that the RSA stack has been reviewed 17:12:41 ahf: but let's be in contact 17:12:46 are you gonna update the reviewer sheet with this when you do reviewers for the entire week? 17:12:50 yes i will 17:12:54 yep, i'm probably gonna skim over the RSA one too then 17:13:06 k sounds good 17:13:25 the branches are both big, but there is a fairly big amount of code movement in each 17:13:46 fyi, there is a memory leak somewhere that I am still tracking down, so don't be alarmed if the tests report a memory leak. 17:13:51 ok 17:14:03 ok 17:14:10 i second nickm's recommendation for git diff --color-moved when reviewing some of these commits 17:14:41 ack 17:14:52 mikeperry: ack on the wtf-pad stuff 17:14:53 (and if your git doesn't have --color-moved, it's pretty easy to build a new git from source. Ask me for help!) 17:15:02 i have color-moved and i love it 17:15:07 we ready to move on to rotations? 17:15:09 i have that too 17:15:31 This week we have asn=triage, catalyst=community,dgoulet=meetings,mikeperry=ci/coverity. 17:15:40 ack 17:15:57 can we drop community for this week? 17:16:16 I'll take it on. 17:16:18 k 17:16:27 nickm: thanks 17:16:30 I'll probably do a minimal version, and just refer people to other stuff 17:17:16 I think we aren't adding anything to the 0.3.5 milestone this week either :) 17:17:16 feel free to refer to me there, i think you took quite some rotation duties for me while i was gone, nickm 17:17:55 thanks ahf 17:18:15 everyone make sure you read the announcements stuff -- looks like teor has some things there. 17:18:45 on to discussion -- we want anybody else at the team meeting day? I'm okay with the suggestions there, though we should manage Alex's expectations just to make sure he isn't bored 17:19:19 it's Alex who was in montreal? 17:19:38 Alex Crichton. Is that the same one? 17:19:45 yeah i think it's the same Alex 17:19:46 yes, that is the one who was in montreal 17:19:55 cool, he was super nice. i was worried we were too boring for him 17:20:42 If you've got any more people to invite, we should know about them ASAP, so contact me and isa. 17:20:49 juga? 17:21:07 juga was invited already 17:21:11 juga meets those criteria, yeah! If we can get here there, I think we should. 17:21:14 umm 17:21:14 great 17:21:20 we should double check team day tho 17:21:53 very nice 17:21:55 cool! 17:22:00 ok; and we should do so soon. I'm presuming it's pretty late if she's already got her flight, but we should ask her & jon 17:22:21 yep, i can do it 17:22:24 next discussion thing is PRs and CI and github. 17:23:00 Should we require there to be Github PRs for external contributors? 17:23:02 i think juga is coming guys 17:23:07 i read their Re: [network-team] [tor-meeting] Fwd: [IMPORTANT] creating the agenda for Mexico City 17:23:23 yeah, i think they are coming too - this was just about this specific meeting 17:23:27 its implied they are coming, and ive been assuming that they will be coming (right juga ?) 17:23:32 *they 17:23:41 The question is about the team day 17:23:44 ah sorry 17:24:26 i think we should invite juga to team day (and make sure they know they're invited) 17:24:31 agreed 17:24:38 * isabela pinging them via email 17:26:57 Any thoughts on the PR thing? 17:28:25 i'm not sure yet whether we want to make it an outright requirement to submit a PR against our repo. maybe we should consider setting the expectation that review of such contributions could be delayed while we do stuff they haven't done? 17:28:51 agreed we should not make it a requirement for volunteers 17:29:21 perhaps we can politely ask them , or tell them that it might get delayed. worst case we can make a PR of our own, so that we get the travis. 17:29:35 yeah, we can always make PRs ourselves, and add them to the ticket 17:30:05 I wonder if some time in the future it would make sense not to do this as some rotating duty. 17:30:29 s/not// 17:30:48 almost seems like community hero work 17:30:52 asn: +1 17:30:58 yeah, i think that fits well under the community hero work 17:33:07 ok 17:33:26 but how does the community hero know which tickets should be PRed is a different deal 17:33:37 perhaps we can figure this out in the future as we understand more about this problem 17:34:07 makes sense. as a first guess, I'd go "anything that has nontrivial code, and is in needs_review" 17:34:12 does that sound reasonable? 17:34:18 sounds good 17:34:33 is it something that can be done as part of review assignments? 17:35:18 that is also a plausibility 17:36:28 dgoulet: what u think? 17:36:47 maybe, though the earlier it gets done, the better for contributors. It's a good thing to check while assigning, at least. 17:37:05 ack 17:37:10 ok let's have it in mind and see how it goes 17:38:43 nickm, isabela: what's involved in the community hero role? I can try to help out with some of that this week. I'm checking IRC scrollback decently regularly again these days, so I can help out with some of the IRC communication / getting people started. 17:39:03 only one other discussion thing on the list -- but please read everybody's notes and see if there's more to discuss! -- 17:39:14 (sorry, still reading scrollback from the meeting) 17:39:44 how do we feel about Sep 10 for our tenative 0.3.4.8 release target? That gives us 2 weeks to notice if there's any criticial bugs left 0.3.4.7 and if they're hard to fix 17:40:37 ok! 17:42:02 we don't want it to be after the 0.3.5 freeze? 17:42:04 ok. that's it from me. anything else? if not, let's all take a couple of minutes to read all the updates, if we haven't already... 17:42:13 ahf: we wanted it to be on 15 august 17:42:18 ack! 17:42:48 it will slip again if there are bugs we can't fix trivially, but that's life :/ 17:43:22 ack 17:45:17 anything else for this week? 17:46:46 ok, thanks everybody! 17:46:49 o/ 17:46:55 just a reminder: I'm away on thursday. 17:47:01 but I should be here the rest of the week 17:47:16 peace all! 17:47:17 #endmeeting