15:02:38 <karsten> #startmeeting metrics team
15:02:38 <MeetBot> Meeting started Thu Sep 26 15:02:38 2019 UTC.  The chair is karsten. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:02:38 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
15:02:41 <irl> for the remaining changes to gitlab ci for the other codebases, you can make those without review if they are exactly the same as the metrics-web changes (excluding the exonerator line) which i think they are
15:02:50 <irl> acute should be here
15:03:17 <karsten> yes, sounds good, will make those changes. in theory, CI will tell me if I screw up.
15:03:29 <irl> indeed (:
15:04:12 <karsten> do you have more topics for the agenda?
15:04:40 <irl> nothing to add
15:04:57 <karsten> while we wait, is #31826 easy for you to do?
15:05:22 <irl> yes
15:06:17 <gaba> hi!
15:06:20 <karsten> hi!
15:06:42 <irl> #31826
15:06:47 <irl> done
15:06:50 <karsten> yay!
15:06:55 <karsten> thanks. :)
15:07:24 <karsten> okay, shall we start?
15:07:29 <gaba> sure
15:07:36 <karsten> Let's give a last review of the proposal (to be submitted this Friday)
15:07:42 <karsten> have there been any changes?
15:07:51 <gaba> Al cleaned it
15:07:52 <karsten> I didn't check for a few days, and I didn't hear anything.
15:07:57 <gaba> that has been the only difference
15:08:23 <karsten> okay, I can take a look tomorrow morning.
15:08:28 <gaba> ok
15:08:33 <karsten> is tomorrow a work day for you, irl?
15:08:40 <irl> yes
15:08:45 <karsten> great!
15:08:59 <karsten> the plan is to submit tomorrow afternoon PDT?
15:09:04 <gaba> You can leave a comment if anything needs to be changed.
15:09:07 <gaba> Yes
15:09:18 <karsten> okay.
15:09:52 <karsten> anything else on that topic, or moving on?
15:10:05 <irl> i'm stuck on time zones
15:10:11 <karsten> heh
15:10:21 <irl> does this mean that as long as the comments are done before the end of my work day they are fine?
15:10:28 <irl> because then pdf afternoon happens after i finish?
15:10:31 <irl> pdt even
15:10:33 <gaba> before of your work day tomorrow?
15:10:39 <gaba> before noon I think
15:10:52 <irl> well, after noon is my afternoon
15:10:56 <gaba> Right now is 8:11 am here in the PDT world :)
15:11:01 <karsten> plenty of time!
15:11:12 <gaba> ok
15:11:17 <irl> ok, so yes, i have until the end of my day to make the comments and then you still have hours of time
15:11:23 <gaba> ok
15:11:47 <karsten> okay, great.
15:11:52 <karsten> roadmap?
15:11:59 <irl> ok
15:12:34 <karsten> as discussed last week, I worked on a couple smaller issues.
15:12:56 <karsten> none of which being on the roadmap, but all of which being in our endless backlog.
15:13:01 <karsten> in terms of trac tickets.
15:13:23 <irl> i am on track to get the monitoring system finished off to close that ticket before monday
15:13:34 <gaba> #28322 ?
15:13:35 <irl> yeah
15:13:43 <gaba> ok
15:13:50 <irl> after that we can add other specific tickets but all the common problems we've hit will be covered
15:13:56 <irl> especially disk space related ones
15:14:58 <gaba> and next issues will be the cards at the top of the backlog for you irl
15:15:09 <gaba> next issues *to work on*
15:15:37 <irl> yes, onionoo hosts are up next
15:15:44 <karsten> I still didn't find anything particularly important/urgent for me to do in the backlog column. :/
15:16:07 <gaba> the next one that has your name on says "Write reference scripts for graphing key scaling experiment metrics"
15:16:16 <karsten> "Perform one off page load metrics for user perf metrics" is what? set up a selenium like thing?
15:16:17 <gaba> oh, scalability
15:16:28 <irl> yeah page load metrics are selenium
15:16:49 <karsten> I'm not sure what that would buy us at the moment.
15:17:02 <karsten> matt set up such a thing, and dennis set up something similar.
15:17:20 <irl> #31071 would be good to do, and might also inform the exit list work
15:17:50 <karsten> that's from the icebox, yes.
15:17:53 <karsten> moved.
15:17:58 <karsten> to backlog for the moment.
15:18:11 <karsten> but would it make sense to move selenium stuff to the icebox?
15:18:31 <karsten> I don't see myself working on that in the near future.
15:18:39 <irl> yeah i think we need to see the existing selenium work first
15:18:50 <irl> and when we run out of progress on that we can see about picking it up
15:19:11 <gaba> There is other 'sponsor 30' one without name or ticket "Set up Tor Metrics mirror of OONI metadata database"
15:19:29 <karsten> so, moving selenium to icebox for now.
15:19:52 <karsten> ah, that one has a sponsor?
15:19:56 <irl> https://github.com/ooni/sysadmin/blob/master/docs/metadb-sharing.md is the instructions for that one
15:20:03 <gaba> Sponsor 30 is officially starting (finally signed the contract) and we are setting up monthly meetings to coordinate work
15:20:18 <irl> which i think is going to be on the metrics-web host
15:20:35 <irl> and then we have the vanilla tor data available to produce mashup visualisations
15:21:20 <irl> there is also going to be bridge reachability metrics, and part of sponsor 30 is to add these tests to mobile
15:21:24 <irl> so there will be a lot more data points
15:21:33 <karsten> okay. I'll look into that.
15:22:00 <irl> even if you just look at it to make an assessment of what we need to do, that could be really useful
15:22:06 <karsten> right.
15:22:14 <irl> so we can be in sync with anti-censorship and ooni before suddenly we have to do things
15:22:38 <gaba> yes, i just forward a mail about the meeting to both of you
15:22:43 <karsten> what about those other three cards with my initials in backlog?
15:23:18 <karsten> I remember waiting for input from mike on such things, but that didn't happen yet.
15:23:27 <gaba> I'm wondering if we should re-assess this roadmap. The scalability issues may need to go to icebox
15:23:40 <karsten> yes, that's what I was about to suggest now.
15:24:02 <irl> i guess they are blocking on mikeperry a lot
15:24:16 <gaba> yes and mikeperry will not be taking them any time soon
15:24:25 <karsten> and then all those cards starting with 314* were part of the flags simulation that we're not going to do in the near future.
15:24:32 <gaba> and we have a specific proposal that may be funded for this starting in december
15:24:37 <gaba> yes
15:24:53 <karsten> can I move cards?
15:24:56 <gaba> let's focus in october and setpember in all the other stuff (tordnsel and sponsor 30)
15:24:59 <gaba> yes please
15:25:37 <irl> #31172 we might be able to say is wontfix
15:26:11 <gaba> ok
15:26:14 <karsten> where is that?
15:26:41 <irl> not a roadmap one, just looking at tickets
15:26:44 <karsten> ah okay.
15:26:55 <karsten> so, wontfix or fix by killing the jenkins job?
15:26:58 <karsten> or jobs
15:27:19 <karsten> not doing anything means that some jenkins job will remain unhappy.
15:27:34 <irl> ah yes, fix by killing the jenkins job
15:27:50 <karsten> okay. let me write a comment there.
15:29:39 <karsten> so, regarding the roadmap, I think I'm going to move my two cards from backlog to in progress.
15:30:36 <irl> if we add onionperf data to collector, does that automatically appear in metrics-web?
15:30:43 <irl> from a new instance?
15:30:56 <gaba> karsten: sounds good
15:30:57 <karsten> fine question.
15:31:15 <irl> we might need to "enable" instances, so we don't end up with all the experimental ones being visualised
15:31:22 <karsten> if it does we can still put in a filter in metrics-web.
15:31:29 <karsten> right.
15:31:34 <irl> that would be a roadmap task that we don't have yet
15:31:37 <karsten> we should be able to figure that out quickly.
15:31:46 <karsten> the filtering?
15:31:49 <irl> yeah
15:32:04 <karsten> or a comment on "Add experimental OnionPerf results to CollecTor".
15:32:09 <karsten> it's probably a few lines of code.
15:32:14 <irl> ah ok
15:32:16 <irl> that's fine then
15:33:38 <karsten> okay. anything else on the roadmap topic?
15:33:39 <gaba> Anything else related to roadmap?
15:33:41 <karsten> :)
15:33:42 <gaba> hehe
15:33:43 <gaba> ok
15:34:07 <karsten> "Disk space: 1 TB" on the ooni metadb sharing page. uh oh.
15:34:16 <irl> nothing else from me
15:34:19 <irl> karsten: we don't need all the tables
15:34:32 <irl> i /think/ we can just have the ones we want
15:34:34 <karsten> ah! would be great if we can have a subset.
15:34:42 <irl> the bulk of the data is the http response bodies
15:34:57 <karsten> okay.
15:35:11 <irl> saying this we will have all the tor logs for the bootstraps
15:35:20 <irl> and when we have a ton more data points for that it might get big
15:35:28 <irl> but we can ask ooni to give us a summary table then (:
15:35:44 <karsten> it would be good to start with a summary table, yes.
15:36:07 <karsten> anyway.
15:36:16 <karsten> out of topics?
15:36:29 <gaba> yes
15:36:33 <gaba> it seems we are done for today
15:36:49 <karsten> great!
15:36:54 <karsten> next meeting next week?
15:37:06 <irl> yep
15:37:09 <gaba> yes
15:37:20 <karsten> alright!
15:37:26 <karsten> bye, irl and gaba! o/
15:37:31 <gaba> bye!
15:37:37 <irl> bye!
15:37:42 <karsten> #endmeeting