16:58:48 #startmeeting network team meeting, 2 july 2018 16:58:48 Meeting started Mon Jul 2 16:58:48 2018 UTC. The chair is nickm. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:58:48 Useful Commands: #action #agreed #help #info #idea #link #topic. 16:58:51 Hi folks! 16:58:54 hi hi 16:59:00 Meeting pad is at https://pad.riseup.net/p/XIHRUTtCngRZ 16:59:16 who do we have here today? I know dgoulet is off for Canada Day... 16:59:17 hello! 16:59:24 * dmr waves 16:59:39 * arma2 is nearby if you need him 16:59:58 o/ 17:00:03 hi 17:00:08 * Phoul is here to watch again 17:00:22 isabela: ping 17:00:26 roadmap time! 17:01:25 Are we all working on stuff that's on the roadmap? Do we still think our chances are reasonable for doing the stuff there between now and September? 17:02:28 (crickets...) 17:02:56 some of the stuff there that I'm on, I can only make progress on in collaboration with the other people ther 17:03:26 so, if there's anything we're both on, please don't assume I'm too busy :) 17:03:43 we should get the memory profiling s8 work planned this week in which tasks we do and who does what 17:04:06 a bit of it is planned in the notes from the seattle session 17:04:13 but it was only david and i in that one 17:04:14 ahf: so I was assuming that you and/or dgoulet would be working on the "where did the memory go" part. Is that right? 17:04:21 i think so, yes 17:04:26 I'm hoping we can get memory numbers early, and then plan the rest. 17:04:30 and then we are gonna review that together with you 17:04:31 nickm: oi 17:04:32 I feel a little guilty about doing the src/common refactor stuff, but if y'all still think it's a good idea I'll keep going 17:04:38 hi! We're talking about the roadmap 17:04:41 it was a _REALLY_ good idea 17:04:43 and I figure you probably care :) 17:04:46 it's very nice work already 17:05:48 isis: I'm with you on NSS usage and on wide-create. How's the wide-create looking? 17:06:05 mikeperry: I'm with you on some of the wtf-pad stuff. I gather you're not poking that this week? 17:06:10 (fyi we got the extensiion for sponsor8 - so now the end of the contract is dec 31) 17:06:15 isis: we should also get the relay crypto proposal advanced 17:06:33 nickm: I might try to page some of it in as I write the PETS post, to mention it there 17:06:37 isabela: but the extension was primarily for the browser team, right? 17:07:00 isabela: I haven't heard anything from sbs or others about the mobile embedding stuff. Do you know if we've had any feedback there? 17:07:13 nickm: I think trying to wrap up the vanguards stuff into a scipt tht people can pip insgtall and writing a blog post is slightly higher priority. that is my current guess. really going to try to push that this week 17:07:14 ahf: yes 17:07:22 mikeperry: makes sense 17:07:31 im with catalyst on #22266. i'm not sure what's the current state of that ticket. catalyst, would you be able to summarize the situation there and what needs to be done in your view? 17:07:32 nickm: 'sbs'? 17:07:52 nickm: i need to ping them for feedback, i dont think we have done that yet 17:07:55 ok 17:07:58 what is sbs? 17:08:08 wrt #25502, i think that ahf is working on it, and im plannin to help along with reviews etc. 17:08:22 isabela: so we aim for same schedule as we have now in the network team 17:08:29 asn: this is what the process handling work i'm doing now is about 17:08:39 asn: trying to get it to work on windows too so we don't loose feature parity 17:08:41 isabela: whoops; I meant simone 17:08:54 ahf: yes, no need to change things in the roadmap but wanted to share the news just in case y'all need more time too 17:09:01 isabela: ack! 17:09:08 nickm: i think the only feedback there is from simone is on that ticket 17:09:32 https://trac.torproject.org/projects/tor/ticket/25510 -- where's the feedback? 17:09:43 asn: i'll need a bit to refresh my memory on that ticket 17:10:13 I don't see feedback on the child tickets eit 17:10:15 either 17:10:19 catalyst: cheers 17:11:15 nickm: ? 17:11:26 nickm: which ticket is the feedback on? 17:11:31 nickm: ok then i missundertood because i thought they were created based on feeback 17:12:13 nickm: i thought the tickets there were created based on feedback - is confusing since the master is 'collect feedback' 17:12:27 nickm: the wide create stuff hasn't seen much progress the last couple weeks because the meeting and then i was away and doing smaller tasks, but i'm going to start back on that this afternoon 17:12:48 isis: cool! In theory, we're working on it together -- what's the next step? 17:12:55 mikeperry: we are together on #7068. im not sure what needs to be done there and i need to read the paper. we should probably try to meet up at some point and figure this out. 17:13:02 nickm: so those child tickets are not based on any feedback, then they are based on what? our own testing? 17:13:08 mikeperry: im also gonna see you at pets which is gonna make things easier, but thats in 3 week or so. 17:13:16 isabela: our own testing, our own complaints about the code, etc :) 17:13:21 ok 17:13:29 yeah is a bit confusing 17:14:08 isabela: I guess basically, asking whether it's working, and if not, what we can do to help, would be awesome 17:14:20 asn: #7068? 17:14:24 nickm: yay! so the next step is probably the buf_t storage thing, i think i didn't quite get that far in the decoding logic 17:14:28 are we advancing enough on the roadmap stuff? Do we have time to talk about the review asignments? 17:14:37 mikeperry: eeh seems like roadmap entry is wrong 17:14:38 #25651 iirc 17:14:39 nickm: yes i will do that, i will ping all the people we did the survey with 17:14:40 asn: is that a typo? or #7028? 17:14:58 mikeperry: "Handle histogram issues in WTF-pad" 17:15:04 nickm: also i haven't done the sending logic at all 17:15:17 isis: ok. Is that based on the last review, or is is more advancing? I'm hoping we can get what we have cleaned and merged (if disabled) before we get too far ahead. 17:15:31 (the buf_t thing) 17:15:39 asn: ok. yeah, let's first try to get this vanguards post up, and while I write the research draft we can do a bit of chatting about it 17:15:45 also if you wanna work on something at the same time, I'd be happy to take a break from all this refactoring 17:15:51 mikeperry: ok. 17:15:58 mikeperry: what's the plan with the vanguard post? 17:16:03 mikeperry: is that for before or after pets? 17:16:12 asn: before 17:16:17 ok 17:16:18 catalyst: would you be okay sending komlo a link to our "big patches" WIP? 17:16:26 nickm: maybe we should do a high-level pass to implement stub function signatures, and maybe structs, and then go back and fill them in? 17:16:29 nickm: sure, i can do that 17:16:34 catalyst: ty 17:17:07 oh right, PETS 17:17:13 nickm: done 17:17:15 asn: I really hope we can get it done this week, but I have to double-check that the various tor patches for the control port CIRC stats are all in a current tor-alpha release (which one of them may not be yet?) 17:17:16 * dmr is sorry for the connection blip - back 17:17:18 ty 17:17:39 on to code review assignments? 17:17:43 asn: I think one of them might have only made it into master but may still be waiting on an 0.3.4 backport 17:18:20 isis: You've got the most left over from last week. I hope you can get to it all, but if you need to prioritize, my suggestion would be #26449 since it's CI 17:18:27 up to you though, of course 17:18:50 everybody comfortable with their assigned code reviews? 17:18:59 and if you're done with looking at those, please check out rotations :) 17:19:09 david told me there were not many needs_review this week 17:19:11 asn: thanks for doing the review assignments again btw 17:19:19 it was david actually 17:19:20 nickm: when do you think we can backport #26214 to 0.3.4? 17:19:27 asn: oh. Thanks, dgoulet! 17:19:39 mikeperry: dunno. Do you think it's ready? 17:19:40 there were very few of them so he just went for it 17:19:46 it's mostly the stuff from last week iiuc 17:20:16 nickm: I think so. you marked it as "wait tol see if anything explodes after merging to master" 17:20:27 ok 17:20:31 will do then 17:21:18 fwiw, appveyor has been broken iiuc and it's blocking stuff like #26437 17:21:26 im trying to see if it's still broken but it says "queued" 17:21:46 asn: It's probably busted because of the issue with the infinite loop in strcasecmp 17:21:50 we fixed that this morning, thoug 17:21:52 though 17:22:06 ( #26594 ) 17:22:06 also #26535 17:22:24 oh, most of these reviews look pretty small, it should be manageable 17:22:29 test.exe had a test-case that would take forever until #26594 was fixed 17:23:08 mikeperry: are you going to have time for code reviews this week, btw? 17:23:11 asn: do we need to cancel some Appveyor jobs that won't complete due to infinite loops? 17:23:21 catalyst: they time out after 60 min. 17:23:21 ha could be 17:23:23 I think 17:23:33 doesn't appveyor have a timeout? o.O 17:23:39 would be crazy if they just hang there forever lol 17:24:41 nickm: yeah. sorry I dropped that. have been feeling scattered 17:24:52 NP; we're all scattered sometimes 17:25:26 rotations: ahf==bug triage, asn==community, catalyst==coverity, dgoulet==ci 17:25:28 ack 17:26:03 also 0.3.1 is now EOL :) 17:26:22 we're down to 5 supported series -- it feels like a breath of fresh air somehow :) 17:26:36 ack 17:27:21 re Appveyor, i think everything is backed up behind waiting for other half of the job that timed out to finius 17:27:27 s/finius/finish/ 17:27:36 I'd like to hold off looking at 034-proposed and 035-proposed right now , and have that happen as part of the triage work I'm hoping to do this week. 17:28:25 Speakng of that ... 17:28:41 To answer teor's question , I think we need to do 0.3.5 triage now 17:28:50 I don't have the pad where I wrote the triage instructions any more 17:29:02 :( 17:29:05 I can try to reconstruct it from the tag list though 17:29:56 oh! It's at https://trac.torproject.org/projects/tor/wiki/org/teams/NetworkTeam/TicketTriage 17:29:56 was this it: https://trac.torproject.org/projects/tor/wiki/org/teams/NetworkTeam/TicketTriage ? 17:30:02 yes, thanks! 17:30:24 I foulnd that while looking for the weekly bug triage instructions and was briefly confused :) 17:30:30 ahf: I'd like to do phase 1 this week. Are you still okay with helping? 17:30:54 everybody: please make sure that all must-fix stuff in 034 is marked with 034-must 17:31:41 when is the next 0.3.4 point release? 17:31:46 nickm: yes 17:32:00 mikeperry: next week maybe? 17:32:05 what is phase 1 though? 17:32:22 i assume we go over everything in all the phases or? 17:32:26 ahf: see the trac link above, under 17:32:34 looking 17:32:38 "Phase One" 17:32:38 t 17:32:50 Phase 1 is "identify all the stuff we must do and mark it" 17:33:01 phase 2 is "mark the stuff we must do in order to do the stuff we marked" 17:33:03 ah, cool 17:33:17 Phase 3 is "mark the stuff we're planning to remove and solicit comment" 17:33:24 Ideally we could do 1...3 17:33:30 so people could comment soon 17:33:33 but that might be ambitious 17:34:10 let's see 17:34:39 I don't think we're going to get a volunteer for maintaining old stuff and handling backports this week, are we? :) 17:36:16 As for the remaining 034 bugs: After we sort them to "must" and "don't need to" -- could we have a volunteer to make sure every 034 must bug has somebody working on it? 17:36:19 or is that my job? :) 17:37:17 i can do that 17:37:43 the tag i'm going to look for is "034-must", correct? 17:37:45 cool! ahf: let's remember to ping isis once we've been through 0.3.4 17:37:47 isis: yea 17:37:50 yeah 17:37:52 yeah 17:37:53 ack 17:38:02 make sure they're actually in the milestone, though, and feel free to disagree about what's "must" 17:38:21 ok :) 17:39:38 remaining discussion question is ahf about logistics and mexico 17:39:49 yeah, is our dates there fixed? 17:39:50 I think it's right that we have no meetings before the team day 17:39:54 anybody think I'm wrong? 17:39:54 good 17:40:18 i'm gonna ask for travel approval then 17:40:24 Also, if you haven't sent Jon the info he asked for when he sent the meeting invite, please do so TODAY if not earlier. 17:40:57 They're serious about needing everyone to get booked ASAP. 17:41:23 do we have any other discussion stuff for this week?? 17:41:44 sorry, I only meant one of those question marks 17:42:38 oh one thing 17:42:47 go for it! 17:43:06 erinn said we could buy lattepanda boards for testing windows stuff 17:43:32 i'm not sure if it's cheaper to buy them en masse or separately, but i can look into it 17:43:47 how many people roughly would want one? 17:43:50 ok, it would be good to find that out, and figure out who needs one. 17:44:00 I think we might need a couple of days to get back to you though 17:44:00 im interested! 17:44:13 I'm interested, but I don't know if I'd use it or not -- should find out 17:44:23 yeah, no worries, i don't need to know soon 17:44:27 like, figure out how hard the setup is 17:44:48 anything else for this week? 17:45:32 i've just switched to windows 10 on my desktop, gonna be fun to try after so many years 8) 17:45:50 nickm: isabela: we have a 1:1:1 tomorrow iirc at 10am PDT/1pm EDT 17:46:02 oh! Yes! 17:46:10 isabela: does that time still work for you? It works for me... 17:46:16 ahf: O_O oh no lol 17:46:31 ahf: as the main OS? 17:46:41 asn: yeah, still have qubes on my laptop 17:46:47 love it 17:46:47 nickm: one sec i need to do some utc calculation with other meetings i have 17:46:48 and can boot into debian too 17:46:54 was too annoying with a VM 17:47:09 windows 10 is neat. can ssh into the windows thing from linux. type debian in powershell and get a debian environment, lol 17:47:13 brave new world 17:47:40 ok. Let's call the meeting over then if there are no more topics :) 17:47:42 wait,, you just type "debian" and then you have debian 17:47:44 thanks everybody! 17:47:46 #endmeeting