22:59:49 #startmeeting weekly network team meeting, 4 Sep 2019 22:59:49 Meeting started Wed Sep 4 22:59:49 2019 UTC. The chair is nickm. Information about MeetBot at http://wiki.debian.org/MeetBot. 22:59:49 Useful Commands: #action #agreed #help #info #idea #link #topic. 23:00:13 https://pad.riseup.net/p/tor-netteam-2019.1-keep is our pad 23:00:25 who is here today? 23:00:54 o/ 23:01:13 o/ 23:01:16 * teor4 is here 23:01:50 hello gaba, catalyst, and teor4! 23:02:33 asn and ahf are most likely asleep 23:02:40 mikeperry / dgoulet: you around? 23:03:33 I see mikeperry editing; I bet he'll be in soon 23:03:56 first thing: we have 041 released, and 042 coming up. 23:04:11 the 042 status page is created, but not yet useful: we/I need to do a bunch of triage on it. 23:04:35 big feature freeze is 8 Sep; feature freeze is 15 sep 23:04:56 sorry about all the places on the 042 page that still say 041 :/ 23:05:39 anything to talk about on 041 and 042? 23:05:56 if not, we're on to the kanban! 23:06:10 ok 23:06:54 (gitlab has a nice interface to look at milestones like for 042 https://dip.torproject.org/torproject/core/tor/-/milestones/1) :) 23:07:37 yah I am here; there are some small circuit padding fixes and reorgs that could go into 0.4.2, but maybe not 23:07:46 when is the next time origin/master will be open again? 23:07:57 mikeperry: let's add that to the discussion topics? 23:08:34 I moved issues into the 'next' column as we had them in the roadmapping session for s27 in sweden 23:08:49 teor: should 30901 be in "Doing" rather than "review"? 23:09:19 let me check 23:09:22 #30901 23:09:39 it's in needs_revision in trac? 23:09:40 mikeperry: are you working or planning to work on on #30992 ? 23:10:22 I haven't had time to set up my gitlab account or learn to use the kanban, because I have been overloaded 23:11:07 teor: on the bright side, it's pretty fast to do. 23:11:12 gaba: who can get teor an account? 23:11:26 gaba: I've put #29211 into "doing" since I'm currently working on a bunch of little #29211 subtickets 23:11:28 I have an account, I haven 23:11:32 gaba: that's one of the things, and no, not for 0.4.2 at this time 23:11:41 haven't had time to reset my password and learn the new system 23:11:54 nickm: I can get them an account. I will send you a msg later teor. 23:12:28 no really it's all on me, as soon as I get rid of urgent tasks, I'll do it 23:12:45 mikeperry: is something we can get into s2? or we just remove it and add it to the icebox? 23:12:50 ok teor4 23:13:47 gaba: I think icebox until some padding researcher realizes they need it. There are many things of this nature 23:14:20 gaba: plan for now is to document all of this kind of stuff in the dev doc (this one is already in there) 23:14:45 ack 23:14:54 there are a couple tobias found that need to be added to the doc 23:15:17 (actually he wrote a patch, that's why I asked about origin/master) 23:16:28 any more on the kanban? On to review status? 23:16:43 I'm currently all caught up with with my reviews, so if anybody is overloaded, I can take more on. 23:17:59 aha I was just writing that in the pad 23:18:08 ? 23:18:23 I can review patches from folks on circpad since I can do those quickly. I don't have time for anything else rn 23:19:53 It might be easier if I focus on #29211 reviews, and hand over my other reviews to other people 23:20:00 okay -- 23:20:07 just set me as reviewer on whatever, and I can take it on 23:20:20 sound ok? 23:20:52 sure I'll have a look at my reviews soon 23:21:13 thanks -- do you want me to set you as reviewer on all the #29211 stuff that I did today? 23:21:52 right now I only put you on #31625, since that's the one where I could use feedback. (It's design review, not code review.) 23:24:09 teor: (just let me know, we can do it however you want) 23:24:20 (no need to decide right now) 23:24:21 next up is discussion 23:24:32 first topic is that we have some stalled proposals that never made it into torspec 23:25:01 I'm okay writing a draft policy for when we do that, if folks would like me to. We can also discuss it here first. 23:25:14 I'm also okay with just adding 306 and 307 for now 23:26:20 nickm: sorry, I was looking at reviews. Yeah let me see what I can pick up this week. I want to pick them up one ticket at a time, so I don't block review assignment. 23:26:31 makes sense, take your time 23:26:47 does anybody have thoughts on what to do with proposal numbering backlog? 23:27:08 nickm: if you have a priority order, feel free to send it to me 23:27:14 ok, will do 23:28:25 ok, if no thoughts on backlog, I'll add 306 (happy eyeballs) and 307 (onion balance support), and then draft a policy 23:28:28 I don't mind what we do with proposals? Let's deal with the backlog, and write a policy for avoiding it in future? Or try out the new policy on the backlog? 23:28:39 ok 23:28:48 fine w me 23:28:49 nickm: 306 is draft, we asked neel for revisions, and they never happened 23:29:00 IMO it still should go into the repo as a draft. 23:29:07 else we've got a numbering gap 23:29:22 sure, it's obviously not getting revisions any time soon 23:29:38 we can revise it when it's funded or when someone has time 23:30:03 ack 23:30:46 similar on second question: I'd like to advance your draft policy proposal about moving design proposals towards "open" and "accepted". We don't have to do it urgently, though. 23:31:12 would you like to a) come back to it yourself soon; b) come back to it yourself later; c) have me take a whack at a draft; or d) other? 23:31:13 oh I had forgotten about that one. 23:31:30 feel free to re-draft, I have a lot of other policies on my list 23:31:37 ok 23:31:55 next on to mike's question about when the merge window opens 23:32:08 in theory our policy is that it stays close for a month 23:32:09 (or we can do prop 306 when some browser realises they really want it) 23:32:24 last time, we tried opening it early, once we had critical bugs fixes in 0.4. 23:32:27 *0.4.1 23:33:42 In theory, the reason for keeping the merge window closed was to give us time to _just_ work on stability 23:34:06 but last time, the 041Status page seems to have worked a lot better 23:34:25 at least, it did to me. what do others think? 23:35:10 yes having that single page to look over was v helpful 23:35:43 yes, I like the single page, when we move to gitlab we should have a release column in the kanban 23:35:52 * if we move 23:36:34 two problems with opening the merge window early: 23:36:52 people get confused about where to merge things, so we should document that clearly somewhere 23:37:35 people have conflicting priorities on bug fixes vs new work, so we should document that clearly somewhere, and check in to make sure we expect similar things 23:37:58 both of these problems exist even if we don't open the merge window :-) 23:40:15 So to be clear: I think we can open the merge window early if it would be helpful to people. But let's make sure we document things and add check-ins to our process, send out an email as a reminder, then open merges. 23:40:30 How do people feel about that? 23:41:41 I'm okay with that 23:43:59 let's think about what our minimum on bugfixes is before we can open 043. 23:44:10 (doesn't have to be right now) 23:45:22 mikeperry has some work to offload 23:45:31 catalyst would like feedback on jenkins failures 23:45:52 catalyst: okay if I put some notes about them on the pad? 23:46:02 nickm: sure 23:46:04 or we could make a new pad or something 23:46:31 mikeperry: are any of your at-risk tasks things I can or should take on? 23:47:05 nickm: or you could update the CI failures wiki page with your observations 23:47:20 oh! where is that? 23:47:29 https://trac.torproject.org/projects/tor/wiki/org/teams/NetworkTeam/CIFailures 23:47:32 (and is there anything else for today?) 23:48:03 there probably should be a section for "CI failures we're still diagnosing and don't have tickets yet" 23:48:18 ack 23:49:07 I am not sure.. the big thing crushing me is having to pick up the firefox esr work while still finishing up sponsor2 and also ramping up scalability and performance funding proposals, and follow up with researchers.. each of those things are my responsibility and I don't want to give them up to do the esr review.. 23:49:28 wow that's tough, sounds like 3 people's work 23:49:33 I can add you to the thread about the circuit padding doc and you can see what you think about Tobias's work (which is v good and thorough, but needs to be merged with the main doc or left to sand as a hacking doc 23:50:15 I can integrate two documents together if you point me towards them, and that's all I need to do 23:50:32 I don't think I understand well enough to resolve conflicts, but I could proabably be helpful 23:51:12 yeah, that would be helpful. there's a lot of odds and ends there 23:51:25 hopefully after reading both docs you'll understand everything you need. that's thier job :) 23:51:28 ok. send me an email with what I should integrate together? 23:51:37 and what I should do with it when I'm done 23:51:44 if they are still failing at that then we need to stepback and re-evaluate it 23:51:44 and i'll see what I can do 23:53:11 ok great, thanks a lot 23:53:40 np 23:53:45 do we have anything else for today? 23:54:38 not to discuss. Please read announcements as I added something about the bug smash fund campaign :) 23:57:13 okay. Thanks, everybody! I'll see you all online! 23:57:16 #endmeeting