16:58:51 #startmeeting weekly network team meeting, 22 October 2018 16:58:51 Meeting started Mon Oct 22 16:58:51 2018 UTC. The chair is nickm. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:58:51 Useful Commands: #action #agreed #help #info #idea #link #topic. 16:58:57 o/ 16:59:11 good morning!! 16:59:29 o/ 16:59:34 hello hello 16:59:38 hello 16:59:41 mikeperry: ping? looks like we're missing an update from you? 17:00:07 hi ahf and dgoulet and gaba and asn! 17:01:14 so, let's start with our roadmap 17:01:19 * juga around 17:01:40 For the roadmap just a reminder that sponsor 8 is a priority 17:01:45 looks like last week we mainly did the sponsor 19 report and a bunch of bugfixes for 035 17:01:47 I'm trying to sort out the roadmap by priority 17:01:58 take the stuff at the top first *if possible* 17:02:02 yeah 17:02:03 hi juga! 17:02:55 last week was also bug fixing and mike working on the padding stuff 17:02:59 gaba: btw, me and dgoulet, identified a pretty important HSv3 issue (#27838). we added it to the roadmap because it's something to be done in 035 timeframe and it's not reflected in the roadmap. 17:03:13 so, looking at the roadmap for me, I see the first thing with my name on it is "merge existing memory patches", which I'll try to work on 17:03:14 yes, I saw it. I was going to ask about that one. 17:03:18 ok 17:03:27 nickm: is that the mmap() ones? 17:03:39 if for some reason anybody thing that priorities should be different please talk to me 17:03:44 ahf: mmap and maybe cleaning up #27359 17:03:50 ack 17:04:14 ahf: I don't know what the current status is there, so if you want to tell me what to do next on those patches (revise mine, update cyberpunks's, or merge the two) i'd appreciate that 17:04:23 also, there stuff related to the reports done, please add how much time it took you. It is ok to add it to actual points with your name on it 17:04:27 does everybody else have a clear next step on the roadmap? 17:04:43 nickm: i'm gonna take a look, yes 17:04:45 I do (expecting ahf to update me soon :) 17:04:59 dgoulet: wrt #25502 ? Sounds good 17:05:02 * asn added points to the s19 report 17:05:18 nickm: yes 17:05:39 one good place to start there is with a breakdown of the work and a plan, unless there is one already 17:05:55 nickm: that is what ahf is doing I believe 17:05:59 great 17:06:04 catalyst: are you around today? 17:06:05 ye 17:06:11 nickm: yes 17:06:37 i did an estimation for the hsv3 roadmap item, and it was replaced with ?????????. 17:06:40 mikeperry, catalyst, ahf, asn: do you have a clear next task on the roadmap? 17:06:59 nickm: yep 17:07:06 my big thing is row=87 which lacks a ticket. 17:07:14 aka "Review current WTF-PAD branch" 17:07:18 and also make a ticket for it 17:07:19 nickm: 035 bugs, then bootstrap reporting stuff 17:07:26 i'm gonna modify the #25502 entry in the roadmap tonight with the new tickets 17:07:27 the WTF_PAD branch is currently being revised by mike. 17:07:31 so we can try to do estimates for those 17:07:49 thanks ahf 17:08:01 right now #25502 is too meta for us to estimate anything there 17:08:27 catalyst: great-- and same note about a breakdown of work and a plan. You're listed as working with Teor there, but it's also okay to pull me in on this one; I don't have enough sponsor8 stuff right now. 17:08:40 nickm: ok, thanks 17:09:27 ok i made a ticket for the wtf-pad review/merge (#28142) 17:09:30 gaba: can you reach out to mikeperry about the stuff here? He was sick the last couple of mondays, I think, but we should be keeping in touch 17:09:33 asn: woo! 17:09:37 still related with the roadmap but i added it to the meeting discussions is who can be available to help with sbws 17:09:45 yes nickm 17:09:49 asn: would you like me to work along with you on that too, or review it once you're done? 17:10:00 gaba: okay; we'll get to that in discussions 17:10:09 if no more on the roadmap, next topic is reviews? 17:10:14 nickm: it's gonna be in merge_ready after the review phase finishes. 17:10:22 nickm: so at that point, i think you will need to take a look too... 17:10:39 asn: sounds fine! 17:10:46 i will probably be back-and-forth with mike before getting it to merge_ready 17:11:12 updated the ticket to be in needs_revision since mike has the mutex atm 17:11:12 ack; please feel free to pull me in if you think there's any discussion where i'd be helpful 17:11:26 ack 17:12:03 on reviews -- looks like almost everybody has 4-6. 17:12:20 btw i feel like we pumped up the estimation of #28142 in mexico but now it's back to 5-10 17:12:26 #28142 is a big thing. 17:12:35 i think we put it to 15 or so in mexico 17:12:41 I think teor4 may be blocked on the travis sanitizer ones; I'll ask them at the patch party or sooner if they want to offload them. 17:13:03 #28142 great; I was guessing wildly, but I'd be happy to have my guess be way off 17:13:08 this week's review round added 2-3 tickets to each person. 17:13:13 * catalyst can be backup for looking over heavy linker stuff 17:13:19 Does #28142 get us a version of wtf-pad that actually pads? 17:13:29 it does pad, but not the way that mikeperry wants it to pad. 17:13:37 and thats why we have the extra 3 tickets 17:13:40 to be put on top of it 17:13:48 ack 17:13:52 but still #28142... is lots of work. 17:13:58 yeah, I hear you 17:14:25 can/would you like to share it around at all? 17:14:35 share the work? 17:14:38 yes 17:14:44 yes it can be shared for sure 17:14:49 but sharing review is weird 17:15:01 but im up for doing it if we figure out a way to do it 17:15:15 maybe we can split reviewing and testing 17:15:21 and one person tests the branch while the other person does review? 17:15:23 not sure 17:15:38 hm. that might make sense. Anybody interested in testing wtf-pad after its next round of revisions? 17:15:54 i think that mike found some bugs during testing last week, and now he is fixing them 17:16:01 ack 17:16:10 there is also no real testing framework for it 17:16:14 did mikeperry give you a planned ETA for his revisions? 17:16:17 this week 17:16:24 ok 17:16:38 it's on #tor-dev from friday i think 17:16:43 but my backlog does not reach that far back... 17:17:00 ack 17:17:12 next topic is rotations... 17:17:20 oh one last thing: 17:17:49 catalyst: (thanks for the offer on the asan stuff! let's talk to teor4 when we see them) 17:18:20 my offer to help out with reviews that folks are stuck with goes for everybody, so feel free to grab me or reassign hard review stuff to me without asking. 17:18:35 rotations: we have ahf on triage and teor4 on CI 17:19:04 gaba: we're about to ratify the rotations changes, right? 17:19:09 yes 17:19:19 we have one more week to comment/edit or ask questions about it 17:19:27 proposal: https://pad.riseup.net/p/team_rotation 17:19:31 ack 17:19:36 It is a small change of reducing the rotating roles 17:19:40 summary: the rotations become triage and CI, and everything else is "stuff everybody should spend a little time on" 17:19:50 yes 17:20:15 the triage and CI rotations now have specified minimum versions, so folks don't think that CI means "fix everything that is remotely broken by yourself" 17:20:16 just a note: there is no concrete proposal on that pad. 17:20:21 I will do the work on team observer that was proposed and we still have stable roles as the assigner of reviews and stable release mantainer. 17:20:26 the summary that nickm just gave is a good proposal fort me tho. 17:21:05 Can somebody turn what we have there into a consensus proposal? 17:21:06 do we want triaging to look at coverity and create tickets for some of them if they can be clustered? we have had some weeks where there suddenly was 20 new coverity issues opened, but most weeks there is not that much 17:21:31 sure, unless it's obvious who should fix them 17:21:36 ack 17:21:37 I an convert it into a proposal adding soem text at the top but I assumed it was fine :) 17:22:06 it looks like a bunch of info 17:22:07 gaba: I don't think I understand what is actually the proposal either; when I summarized it above, I was secretly trying to confirm my understanding :) 17:22:16 oh 17:22:27 ok. I will try to summarize it 17:22:31 and send it back to the mailing list 17:22:31 thx 17:23:18 speaking of, resolving our current sekrit thread about committers and the stable role of stable release maintainer would be great and in turn expose it to net team for comments :) 17:23:39 +1 17:24:03 that is the other thing we have on the list for today. The idea is to have dgoulet merge code into master for a few subsystems. 17:24:06 gaba: I tried to write a summary at the top for you, but there's still a lot of internal discussion that could use cleaning up for the proposed version 17:24:15 yes. I will clean it up. 17:24:16 thanks! 17:24:20 np; thanks to you! 17:24:24 gaba: and latest response from asn also for him to imo :) 17:25:24 yep replied to the thread today. sorry for taking me a while :) 17:25:44 dgoulet, asn: I'm fine with the stuff that you wrote on the thread; if somebody turns it into a concrete proposal and sends it to network-team, let's agree to it? 17:25:46 nickm, gaba: I'm happy to volunteer to start the stable maintainer discussion on the net team if you like, I'm also happy to sit out, as you wish. I personally just want to get the ball rolling on that role :D 17:25:52 nickm: +1 17:26:06 sounds good. 17:26:10 yes please dgoulet. 17:26:13 dgoulet: you want to write the Concrete Proposal? 17:26:18 * asn concrete proposal is the phrase of the day 17:26:26 ok so I'll do both, the "master merger" and the stable ... 17:26:27 two suggestions for the proposal: 17:26:29 keep it simple 17:26:32 +1 17:26:33 +1 17:26:50 make sure that it's written so that if one of you is overloaded, the world doesn't block. 17:27:05 sounds good, I'll get the ball rolling! 17:27:12 we all have a lot of other stuff on our plates and the coordination issues here might take longer than we expect 17:27:18 ok, thanks dgoulet ! 17:27:50 first discussion topic is sbws ; juga nees an additional person 17:27:57 juga: what kind of help do you need exactly? 17:28:08 the idea is to have somebody else other than teor 17:28:19 that can help juga with any issue related to what she is working on 17:28:32 nickm, gaba, i think 1 person could be enough 17:28:42 not sure teor has time for it (or who else) 17:28:44 not specifically right now but in general with this 17:28:59 mainly i think someone should be reviewing code or 17:29:13 some main questions like last week 17:29:37 any volunteers? 17:29:41 i've added in the pad, what about 1h/week? for someone to review bwauth stuff? 17:30:08 code review shouldn't be too hard; I don't know if I know what kind of questions you have in mind 17:30:20 the code in question is sbws code, and that's in python? 17:30:28 yes 17:30:33 yeah, is it python code in sbws or is it C code in tor? 17:30:35 ok 17:30:39 i hope to get to tor code soon too 17:30:50 ahf: correct 17:30:58 i mean, both 17:31:06 i can try to be helpful with the tor code, i don't think right now is a good time for me to learn a new codebase so close to the end of s8 17:31:07 i can review Python code, but i'm not sure i have enough context to review the design 17:31:35 (nor do i really have the time to acquire the context before sponsor8 ends) 17:31:51 sounds good, let's see what teor says 17:31:58 ok, thanks 17:32:30 next topic? hackweek preparation 17:33:14 Just a reminder that we are starting to prepare it and please comment on it when you have a chance. https://pad.riseup.net/p/tor-netteam-hackweek-2019.1-keep 17:33:16 \o/ 17:33:37 It seems that everybody is ok with the week before fosdem 17:33:57 (ugh, I messed up my alarm for this meeting... sorry) 17:35:26 gaba: assuming it's gonna be in brussels i think 17:35:38 or close by. that is the other thing we are looking at 17:35:40 there are people who needs to go to fosdem if we aren't doing it in brussels :-P 17:35:40 big hub for travelling 17:35:43 yep! 17:36:02 if you have suggestions on where to do it close to brussels please add it to the pad 17:36:21 gotta find someplace with cheap & easy travel and lodging... 17:36:25 yes 17:36:41 do we want to do the wilmington thing where a lot of us are in a big airbnb and the rest takes a hotel nearby? 17:36:42 brussels may be easy travel but not so much about cheap on lodging 17:36:45 we might be able to get kuleuven to offer meeting space 17:36:50 there are many hotels there because of the EU hq 17:37:00 kuleuven ? 17:37:14 nice idea 17:37:20 the university in leuven, which is a suburb of brussels. they do anonymity research. 17:37:30 oh, i see 17:37:31 cool 17:37:37 perhaps an airbnb there would also be cheaper than brussels city? 17:37:39 https://www.kuleuven.be/english/ 17:37:48 yes 17:37:54 then we can move into brussel city on the friday for those staying for fosdem 17:37:58 http://homes.esat.kuleuven.be/~cdiaz/ 17:38:18 I'm going to talk with jon this week and start looking at possiblites around this 17:38:28 (last link: https://www.esat.kuleuven.be/cosic/ ) 17:39:05 that sounds like a good idea, arma1 17:39:13 it's a cheap train ride leuven<->brussels 17:39:19 yah 17:39:20 gaba: are you keeping metrics in mind here too if they wanna do a parallel meeting with us? 17:39:30 both karsten and irl expressed interest in that 17:39:35 i suggested bruges because it's nice, but i think leuven is realistic and we might get a free meeting space. ;) 17:39:44 I'm going to talk with them ahf 17:39:50 as you suggested that today 17:40:24 cool! 17:40:34 thanks! 17:40:52 let's see, any "want help with"s for today? 17:41:15 I've been letting people know I'm not on enough roadmap stuffed for right now so you should feel free to pull me in 17:41:58 I want suggestions on that publish/subscribe API proposal. Mainly, should I build it, or revise it? We can revise it post-build or post-merge, but now would be easiest -- so please have a look. 17:42:11 anybody else need help or have a discussion topic? 17:43:28 i already brought everything i had to discuss 17:43:36 * dgoulet is good 17:44:31 if nothing else -- let's call the meeting, and see each other online! 17:44:45 * nickm waits another 30 sec :) 17:44:48 bye! o/ 17:44:51 thanks! :) 17:44:59 o/ 17:45:05 thanks 17:45:24 cheers all! 17:45:25 #endmeeting