15:02:11 #startmeeting metrics team 15:02:11 Meeting started Mon Jun 10 15:02:11 2019 UTC. The chair is karsten. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:02:11 Useful Commands: #action #agreed #help #info #idea #link #topic. 15:02:22 sorry for the major confusion about meeting times... 15:02:38 glad it all worked out in the end and we're having a meeting now. :) 15:02:58 I just put a few topics on the agenda pad. 15:03:09 anything else you want to add? 15:03:18 nothing from me 15:03:32 np. is ok for me 15:03:36 cool! 15:03:43 let's start with: 15:03:44 Monthly report for April and May 15:03:58 this is a very quickly written list of what we did in april and may. 15:04:07 I might find more items after the meeting. 15:04:14 do you want to add any of the discussion on performance metrics? 15:04:23 could you please look through your notes/inboxes and add anything that's missing? 15:04:32 yes, will do that after the meeting 15:04:36 gaba: I have the updated/new graphs and the error codes on that. 15:04:43 ahh 15:04:43 ok 15:04:48 yes 15:04:49 or what else did you have in mind? 15:04:54 that is it 15:05:01 (2 months is quite a time, I might have overlooked a lot.) 15:05:07 okay. 15:05:14 thanks, irl! 15:05:25 quick question about #30219: 15:05:39 should we add an @source annotation to bandwidth files? 15:05:53 no need to discuss this now, but this could help me make progress over the week. 15:06:10 I also asked atagar via email to take a look, 15:06:22 because I vaguely recall irl saying that he had thoughts on such an annotation. 15:06:25 it is something that tor already has 15:06:38 source, downloaded-at, purpose and maybe others 15:06:40 yes, do you recall how it's being used? 15:06:48 is it specified somewhere? 15:06:49 https://gitweb.torproject.org/tor.git/tree/src/feature/dirclient/dirclient.c#n1816 15:06:58 it's not specified anywhere because it is only used internally 15:07:05 but it is an ip address 15:07:12 which would work, I guess. 15:07:32 we can also treat it as an opaque string 15:07:53 yes, totally. 15:08:10 I think metrics-lib would treat it as a string anyway, rather than parse it. 15:08:16 pretty sure. 15:08:23 yeah 15:08:31 okay, want to quickly comment on the ticket? 15:08:34 atagar didn't want to add it to stem because collector wasn't using it 15:08:44 if we did use it then atagar has reason to add it to stem 15:08:57 i will comment on the ticket, yes 15:09:04 awesome! 15:09:15 it's a decision to be made, but I think it could be useful here. 15:09:32 other than that, I see us adding a bunch of bandwidth files and losing important information while doing so. 15:09:49 okay, moving on. 15:10:00 OnionPerf log file availability 15:10:09 just the quick question: is it easy to make these available? 15:10:17 if not, never mind. 15:10:33 it is at least a few hours of figuring out how to log into the boxes again 15:10:42 :( 15:10:45 it has been a long time since i did and i don't even know which yubikey will do it 15:10:50 hehe 15:11:04 okay, I'll assume it's for post-all hands then. 15:11:36 next topic is all hands related, too. 15:11:59 let's briefly talk about what I can tell them that the metrics team could contribute. 15:12:11 I started a list on the pad. 15:12:42 this is mostly a list ordered by effort, not a commitment of any kind. 15:12:55 any commitments should happen in stockholm. 15:13:06 we could do things like deploying metrics tools on test networks to help with scaling experiments 15:13:11 but I thought it's useful to give an idea what we could do and how hard that would be. 15:13:12 #17036 15:13:56 hmm, tell us more? 15:14:15 there are test networks and we might want to do experiments in those that we are not comfortable running on the live network 15:14:25 we might need to do long running measurements in those networks 15:14:42 in that case, running onionperf or collector in those networks would probably be useful 15:14:54 I know of 1 test network. are there more? 15:14:59 it's just a vague class of things that we could do 15:15:04 i don't know 15:15:43 we're probably not going to sign up to integrate collector+metrics-web in chutney 15:15:44 added to the list. 15:15:58 not at all! :) 15:16:01 but we can help these tools be reused 15:16:21 ok cool 15:16:30 I think the test network already runs a version of collector and onionoo. 15:16:54 sounds good. 15:17:42 I guess we'll discuss this list in a few days at the all hands prep meeting. 15:17:59 if anything else comes up, please feel free to just add it. 15:18:13 ok 15:18:31 gaba: what else would we want to discuss regarding all hands? 15:18:39 I'm asking because you suggested this topic at the last meeting. 15:18:44 did you have anything else in mind? 15:19:01 not really. I think you got everything covered here. 15:19:06 perfect! 15:19:17 I think we're out of topics then. 15:19:40 next meeting would be in, err, 2.5 weeks or so. 15:19:41 yes 15:20:08 alright! 15:20:17 june 20th? you are going to be in the all hands 15:20:40 I'll be at all hands then. 15:20:48 ok 15:20:49 in a weirdly different time zone. 15:20:57 in my timezone ;P 15:21:00 yes, that! :D 15:21:34 so you will do the meeting from the meeting? 15:21:47 ah, no, I wasn't planning to. 15:21:55 june 20th is in 1.5 weeks. 15:22:04 so 27th? 15:22:07 june 27th would be the next meeting. 15:22:07 if the weekly meeting is at 7 or 8 then it will be before your all hands meetings 15:22:15 ah. 15:22:30 okay, maybe we should have it then. 15:22:31 7/8 am 15:22:38 to give an update. 15:22:39 i will be vacation still on the 20th 15:22:40 ok 15:22:42 ahh 15:22:43 ah, right. 15:22:48 ok, then let's do 27th 15:22:51 cool 15:22:52 yes. 15:22:53 and we can update on signal 15:23:00 or mailing list 15:23:02 sounds good 15:23:03 yep. 15:23:27 cool! let's end the meeting before we can't say anymore that it was a good, short meeting. 15:23:37 have a great time. bye! o/ 15:23:44 bye! o/ 15:23:52 enjoy holidays and trip to canada 15:23:59 thanks! 15:24:18 #endmeeting