17:59:28 #startmeeting weekly network team meeting, 5 mar 17:59:28 Meeting started Mon Mar 5 17:59:28 2018 UTC. The chair is nickm. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:59:28 Useful Commands: #action #agreed #help #info #idea #link #topic. 17:59:42 hello! 17:59:43 welcome! I hope everybody's doing well. 17:59:50 our pad is at https://pad.riseup.net/p/Ax0sWNWjEFbt 18:00:05 who do we have here this week? 18:00:27 buenas 18:00:33 o/ 18:00:46 hi 18:00:49 hihi 18:01:52 Hello everyone. 18:02:10 welcome, kushal ! 18:02:39 so, we have a whole bunch of stuff on the pad this week ( https://pad.riseup.net/p/Ax0sWNWjEFbt ) 18:02:57 I see updates from isabela, catalyst, ahf , pastly, and me 18:04:02 o/ 18:04:15 \o 18:04:43 so I think our biggest category of stuff for this meeting is getting ready for rome, and the roadmapping we have to do 18:04:56 including making sure that our near-term stuff (0.3.3) really gets done 18:05:13 we have 23 033-must tickets; if they're all really "must".... 18:05:28 also, 034 freezes in May. :) 18:05:46 so let's everybody remember to pick up code review from review-group-34 18:06:46 catalyst: I see your DST discussion on the discussion list. how would you feel about having meeting time planning be one of the things we do in rome when we're (almost) all in one place? 18:07:31 is this the problem where we might end up colliding with the browser team meetings here on irc? 18:07:47 ahf: that would be one factor to consider 18:07:50 if we do DST and they don't, that could happen 18:08:05 nickm: sure, we could do it in Rome or have it over email if people prefer 18:08:55 i would mostly like us to not forget to think about it 18:09:01 maybe this time we can bend timespace and find a time that works for teor, europe, and the NA west coast :) 18:09:04 catalyst: +1 18:09:18 that would be very nice 18:10:18 ahf: thanks for tagging the wakeup/timeout tickets I opened last week, btw 18:10:24 isabela: i'm gonna look at the incoming s8 onion service logs before i travel to rome and send it to you 18:10:36 nickm: huh? i didn't - i completely forgot that 18:10:37 have a thought about whether there are any you super-want to do, and we can divide them up. 18:10:44 we can divide them up! 18:10:51 ahf: I thought you put them under s8-something 18:11:12 no, i forgot about that friday. let me do that after the meeting 18:11:19 ok thanks 18:12:00 also, everybody please look at the "THINGS TO DO" section there. isabela and I have tried to get together the stuff that would be most helpful on. isabela's items are probably more time-sensitive than mine, since they're planning-related 18:12:04 nickm: did you see my stat reporter thing? it expanded a bit for being an idea i had to get some info out from my pocket phone because i noticed that i didn't use tor on the phone in a normal way when i had a debugger attached 18:12:20 ahf: thanks 18:12:25 hm. that could be pretty helpful. I hadn't seen it though 18:12:26 and i think it might be useful for relay operators too who like to do those shiny dashboards that have become hip in the cloud computing world 18:12:49 it's pretty simple: you can set an option in tor with a set of IP/port + a protocol and tor will connect to the service and report stats periodically 18:12:53 like graphite and statsd 18:13:16 which are time series DB's 18:13:31 could be worthwhile if we can keep it focused for now to something that helps us do our near-term objectives 18:13:50 could be a bad idea if it sprawls and prevents us from actually improving the things we're measuring :) 18:14:20 I guess "just dump it to a text file" is not going to work in phoneland? 18:14:24 yes, the patch is not super big, i'm gonna get it into a state where it can be reviewed by other people before rome and then i think want to talk a bit with people on the team in rome about it if it could be useful for them for internal stats of the tor process 18:15:03 just dump it in a text file is fine and what i do now (i use our logs and parse those) 18:15:18 btw, under the category of "let's plan for planning" -- folks should know that isabela and I are conspiring for this round of releases to actually go smoothly and as-planned. 18:15:27 ahf, Means instead of deploying our own tools, we can just depend on this to report the stats? 18:15:43 that's going to mean kicking way more stuff out in advance, though, and making more people upset about the stuff we say "no" to in advance 18:15:46 kushal: you can, yeah 18:16:02 ahf, Okay. 18:16:24 nickm: ack! 18:16:26 (fyi i left a note for a sync on the 9th at a lobby in one of the hotels to finish the roadmap prep work will email with mroe info once jon tells me a good spot to do that0 18:16:34 ^^ optional too 18:16:40 ahf: make sure you talk to dgoulet about how ( & whether) this fits into the trace ideas 18:17:02 nickm: i will keep that in mind with htis -i wont spend much time on it other than getting it into a shape where others can look at it, then we can decide if we want it not 18:17:21 nickm: i discussed it with dgoulet the other day - he is interested in it for tracing and for relay measurements too 18:17:37 the malloc stats i added was a demo for that 18:17:51 catalyst: I think I dropped you a note a day or two ago about the s8-errors work on one of isabela's panning pads. I hope you saw that? 18:17:51 isabela: cool. i think that sounds like a useful thing to attend 18:18:26 i took a look at ahf's reporter thing, it was surprisingly minimal and looked pretty awesome it terms of how fast we'd be able to hook in whatever random measurements we wanted 18:18:27 isis: since you won't be able to be at rome, also, we should make sure that the stuff you're most concerned about makes it on the roadmap, and you wind up on it 18:18:53 isabela: any thoughts on how we can make sure isis winds up on the roadmap in a useful way? 18:19:02 isis: :-) 18:19:45 nickm: if you mean the brief message you sent on 3/3, i got it and haven't had a chance to write anything on the pad yet (i have read through parts of it though) 18:19:50 ok 18:19:51 i could write up a draft roadmap of the stuff i'm planning to do between now and may? (or longer… not sure if we were planning to roadmap again in may) 18:19:56 winds up == ? :) like helping out ? 18:20:11 like, we should make sure that isis's work winds up on the roadmap, with isis doing it 18:20:26 isis: please add to the pad the stuff you think we should be doing for the sponsors 18:20:39 also i'm happy to be on video/call/whatever while the roadmapping is happening, also for anything else 18:20:49 what is it? is that proposal you added that i moved to v? 18:20:55 isis: I wonder if there's a way for us to do that collaboratively, to make sure that we're collaborating 18:21:06 we could have a jitsi for sure 18:21:12 it was helpful for me at leasst 18:21:15 when i couldnt come 18:21:27 as i think about it, roadmapping is less "planning" and more "planning with a feedback loop" 18:22:10 where we ask each other annoying questions like "you can't possibly do that in a month" and "that's cool but we actually need to deliver on our obligations" 18:22:25 the goal now for me at least is to have that pad organized and hopefully with tickets for the things folks are listing 18:22:41 i added a lot of info there from reports 18:22:49 but is more for reference 18:23:01 isabela: i will update the pad today, thanks! 18:23:07 great 18:23:08 thanks 18:23:19 catalyst: it would be great to have your additions there. I'm pretty much at sea myself when it comes to the state of the s8-errors work, so we should try to figure out where we are, where we are going, and how we get there 18:23:43 once i have the tickets there my second move is to have owners for them 18:23:47 And if we don't know, or don't know scompletely, we should be aware of _that_ so we can fix it 18:23:50 *completely 18:23:57 then we will do another exercise of what goes on what release 18:24:07 cool 18:24:12 cuz this will determ where in the roadmap it will be 18:24:24 we have 3 releases freezing or coming out between now and september 18:24:31 033 releases in April 18:24:35 034 freezes in may 18:24:39 035 freezes in september 18:25:33 yes, so once we have a list of what we want to do for a sponsor, tickets for it, and owners, we will place it as part of a release so we know when it will be done 18:25:34 mm 18:27:49 anything more for today? An additional thought. There are something like 8 of us working full time on the network team. There are also 39 tickets in review-group-34 and 23 tickets in 033-must 18:28:18 that means that if everybody reviews 5 items in and fixes 3, we'll be done with those handily 18:28:29 three each of 033-must? 18:28:32 sounds good 18:28:39 assuming everybody does it 18:28:50 and assuming everybody is equally able to work on all 23 18:29:13 if we wind up at 10 April and we still have must-fix items, it will be hard :) 18:29:38 isabela: do you have anything else for this week, or shall we say "see you in rome!" to everyone and "see you online!" to isis and call the meeting done for today? 18:29:54 when is our next meeting btw? i need to ditch some classes during rome and would like to catch up on those when i get home and i might want to use the monday after rome for that unless we have a meeting there 18:30:07 monday the 19th that is 18:30:14 i think i should email the team and explain the 2 questions for the future vision roadmap hehe 18:30:19 whatever we want to call it :) 18:30:21 isabela: good idea 18:30:30 will do that! 18:30:35 ahf: maybe we can decide in rome? there are fewer of us here than there will be there 18:30:37 and if folks has questions i am happy to clarify 18:30:42 yes, agreed 18:30:44 my expectation is that on the 19th we might be better off skipping 18:31:04 * isabela will be flying 18:31:04 ack 18:31:05 :) 18:31:20 nickm: skipping 19th sounds reasonable. maybe we should announce that? 18:31:36 catalyst: let's discuss it on the 10th/11th, and we'll have time to announce it then 18:31:50 ok 18:32:09 ok. peace, all! Let's try to come out of this next meeting more organized, with plans we can actually follow, and increased hackish camaraderie ! 18:32:22 #endmeeting