15:59:57 <GeKo> #startmeeting network-health 10/10/2022
15:59:57 <MeetBot> Meeting started Mon Oct 10 15:59:57 2022 UTC.  The chair is GeKo. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:59:57 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:00:03 <GeKo> let's do some sync for the week
16:00:46 <GeKo> it seems we don't have our susual pad available due to issues
16:00:57 <GeKo> as the riseup pad is down
16:01:06 <hiro> my updates are here https://gitlab.torproject.org/-/snippets/167
16:01:08 <GeKo> but we can chat a bit anyway
16:01:18 <GeKo> smart :)
16:01:41 <hiro> and here for last week log https://gitlab.torproject.org/-/snippets/166
16:02:04 <GeKo> hiro: did you see: https://gitlab.torproject.org/tpo/network-health/team/-/issues/272
16:02:19 <GeKo> is there something that needs kicking forom our side?
16:02:19 <hiro> yeah
16:02:25 <GeKo> *from
16:02:31 <hiro> the deployment to finish
16:02:38 <hiro> that's what we need
16:02:40 <hiro> :)
16:03:09 <hiro> during the weekend we had two succesful run of the update app
16:03:29 <hiro> so today I am reprocessing the missing user data for september
16:03:39 <GeKo> you mean the website deployment?
16:03:41 <hiro> and tomorrow should have new data for the dates we were missing in october
16:03:45 <hiro> yes
16:03:53 <GeKo> okay
16:04:00 <GeKo> could you give an update on the ticket?
16:04:05 <hiro> sure
16:04:10 <GeKo> so ggus knows what's up with that
16:04:14 <GeKo> and we have a record for later
16:04:17 <GeKo> thanks
16:05:09 <GeKo> does anyone of you have some topcis we should chat about today?
16:05:19 * juga is good
16:06:34 * ggus arrived now
16:06:36 <ggus> o/
16:06:48 <GeKo> ggus: hey!
16:06:57 <GeKo> anything we should chat about?
16:07:04 <ggus> GeKo: tor relay meetup agenda
16:07:14 <ggus> i didn't have time to send the invitation
16:07:15 <ggus> but
16:07:31 <ggus> 'no agenda, no attenda.'
16:07:43 <GeKo> heh
16:07:55 <ggus> GeKo: i thought about adding the state of ddos thing
16:08:07 <ggus> some relay ops are reaching out to me in private
16:08:12 <GeKo> yes
16:08:19 <GeKo> eol
16:08:24 <GeKo> is another one
16:08:41 <GeKo> just to give an update where we are
16:08:41 <ggus> https://twitter.com/applied_privacy/status/1579162537480056832
16:08:44 <GeKo> and resoning
16:08:50 <GeKo> *reasoning
16:09:57 <ggus> GeKo: we should talk about the new project with the community
16:10:04 <GeKo> i wonder how we should communicate that better but, sure,
16:10:29 <GeKo> i am open to all improvement ideas
16:10:34 <GeKo> right
16:10:40 <GeKo> that's another big thing
16:11:03 <GeKo> i'd ask you to start a pad if the riseup pad server wern't down :)
16:11:10 <GeKo> *weren't
16:11:22 <ggus> i'll create a ticket and tag you
16:11:25 <GeKo> but we are logging the meeting so nothing gets lost
16:11:37 <GeKo> that might be a good solution here too, thanks
16:12:56 <ggus> GeKo: last time we met at 19utc, is still a good time for you?
16:13:18 <GeKo> yep
16:13:38 <GeKo> it's getting better in the coming 6 months anyway as dst will go
16:14:23 <ggus> https://gitlab.torproject.org/tpo/community/relays/-/issues/50
16:14:30 <GeKo> ack
16:15:39 <GeKo> ggus: anything else for today (apart from our s112 part maybe)?
16:17:05 <ggus> yes
16:17:29 <GeKo> great. you still have the mic :)
16:17:37 <ggus> i don't think that MR fixed the snowflake spikes
16:17:43 <ggus> https://metrics.torproject.org/userstats-bridge-combined.html?start=2022-07-12&end=2022-10-10&country=af
16:18:33 <GeKo> the website update is not deployed yet fwiw
16:18:41 <ggus> i'm bringing this because journalists and volunteers are asking me about metrics and snowflake
16:18:53 <GeKo> yeah, i know
16:19:04 <GeKo> i think we are close, though, hiro?
16:19:40 <hiro> Yep
16:20:19 <hiro> ggus: the mr has run just twice yesterday
16:20:53 <hiro> The spikes are caused, I think, by the update app not finishing in time for the other run
16:21:05 <hiro> I am trying to optimize the queries
16:21:19 <hiro> So that’s why I am reprocessing the data
16:21:52 <ggus> i see
16:22:13 <hiro> The idea is that we have been seeing spikes in “secondary” countries in terms of traffic where counting only a fractions of descriptors twice could make a difference in terms of %
16:22:39 <hiro> Now that I reprocessed the last few days worth of data a few times we are seeing spikes also in the general graph
16:24:16 <hiro> The bad news is that restoring data is going to take a few days still
16:24:27 <hiro> All our processes are very slow
16:24:45 <hiro> But we are already testing the next pipeline
16:25:12 <hiro> So hopefully we are not that far away from a future where all these processes are more agile
16:25:59 <ggus> great! :)
16:26:46 <GeKo> okay
16:27:08 <GeKo> ggus: you are good so far?
16:27:37 <ggus> yes
16:27:41 <GeKo> great
16:27:46 <GeKo> so for the s112 part
16:28:04 <GeKo> i can report that i've been going over tickets making sure aeverythin is tagged properly
16:28:10 <GeKo> *everything
16:28:21 <GeKo> i am not done yet but will finish that tomorrow
16:28:30 <GeKo> mikeperry started tagging stuff as well
16:28:42 <GeKo> and what we have right now you can find at https://gitlab.torproject.org/groups/tpo/network-health/-/issues/?sort=created_date&state=opened&label_name%5B%5D=Sponsor%20112
16:28:57 <GeKo> it's already some tickets :)
16:29:23 <GeKo> if there are things mislabel or stuff is missing feel free to correct that
16:29:29 <GeKo> otherwise just ping me
16:29:43 <GeKo> do we have any other s112 update for this week?
16:30:44 <ggus> gaba and i started to create the community team tickets for s112. we plan to finish this week
16:31:11 <GeKo> nice
16:32:03 <GeKo> anything else for s112? or are we good here for this week?
16:32:41 <GeKo> assuming everyone is fine :)
16:32:51 <GeKo> thanks for this week and happy hacking
16:32:55 <GeKo> #endmeeting