13:01:07 <hiro> #startmeeting 2026-06-01 network-health
13:01:07 <MeetBot> Meeting started Mon Jun  1 13:01:07 2026 UTC.  The chair is hiro. Information about MeetBot at https://wiki.debian.org/MeetBot.
13:01:07 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
13:01:18 <hiro> #link https://pad.riseup.net/p/tor-nethealthteam-2026-keep
13:01:33 <Rohithh[mds]> Exams are done and now having summer holiday so I can now totally focus on working with tor \o/
13:01:55 <hiro> don't forget to have some fun Rohithh and congrats on being through with your exams
13:02:04 <juga> +1
13:02:45 <hiro> who would like to go first?
13:02:47 <Rohithh[mds]> Thanks :)
13:02:49 <sarthikg[mds]> i can go first. i had a bunch of fixes going on for aggregator, like adding a few new columns, fixing a few. apart from it, things on nsa are going, including new charts, and deployment! this week was also experimenting with the monitoring service, and it seems to be going good
13:03:09 <juga> \o/
13:03:09 <hiro> nice thanks sarthikg
13:03:56 <GeKo> sarthikg[mds]: what is the timeframe for getting the latest statuses done by aggregator?
13:04:09 <GeKo> + the family_ids included?
13:04:19 <GeKo> we need that for happy family transition analysis
13:04:47 <sarthikg[mds]> relay pipeline is kinda stuck at data issue. i will be raising a MR for family_id's and exit_addresses in a few mins
13:05:07 <GeKo> how can we unstuck the data issue?
13:06:04 <sarthikg[mds]> i should be modifying the pipeline to skip if there is data issue, i'm a bit confused as to what all scenarios it should skip in, and what all scenarios it should error out
13:06:27 <GeKo> what does "data issue" mean here to begin with?
13:06:32 <GeKo> are we missing data?
13:06:41 <GeKo> or is it corrupted? or?
13:06:47 <hiro> sarthikg: we could sync about this but if you are missing the consensus in the database you should just skip that status
13:06:51 <sarthikg[mds]> yeah, we are missing some consensus data
13:07:28 <sarthikg[mds]> yeah, i'll update aggregator to skip that
13:07:35 <GeKo> so, we have the consensus on collector but it's not in the db?
13:07:39 <juga> sarthikg[mds]: the consensuses at collector-rs#62
13:07:53 <hiro> we can log the missed statuses so we know
13:08:18 <hiro> GeKo (IRC): I think we have the consensus because we created the network status record in the table, but we didn't create the entries
13:08:42 <sarthikg[mds]> juga: this one we skipped. there's a few one where only a single consensus is in the db. microdesc or the unflavored one
13:08:42 <GeKo> so this is a parsing issue then?
13:08:51 <hiro> it could be that the service was restarted
13:08:58 <hiro> I have to reprocess the data
13:09:19 <juga> sarthikg[mds]: ic, thanks
13:09:45 <GeKo> okay
13:10:02 <GeKo> do we have some notes/tickets where we collect what we need to reprocess over time?
13:10:21 <GeKo> it seems there are things piling up and it's hard to keep track on that
13:10:30 <sarthikg[mds]> GeKo (IRC): give or take, if the MR is merged today, i think taking a day to process a month of data, it should take about a week (considering it needs to process from Sept for family stuff)
13:10:47 <GeKo> yeah, that's cool. thanks
13:11:06 <sarthikg[mds]> GeKo: should we open tickets for that in aggregator or parser?
13:11:28 <hiro> there is a ticket for the missing consensus on the parser side
13:11:34 <hiro> *consensuses...
13:11:47 <hiro> if this is about reprocessing with aggregator it should be on aggregator side
13:12:00 <GeKo> yeah. where the issue is basically
13:12:12 <sarthikg[mds]> hiro: i see, i'll raise a ticket
13:13:14 <sarthikg[mds]> i think that's it from my side 😌
13:13:53 <hiro> thanks!
13:13:56 <juga> from my side:
13:13:56 <juga> last week i was jumping a bit from one to another open fronts
13:13:57 <juga> this week i plan to focus on collector and maybe sbws issue
13:14:31 <hiro> thanks @juga do let me know if you want to go through some of those collector issues with an extra pair of eyes
13:14:44 <juga> hiro: yeah, maybe, thanks
13:15:20 <GeKo> i've been doing mostly anomaly analyis last week (apart from bad-relay work)
13:15:45 <GeKo> focusing on finding issues in our collection/parsing and software
13:15:55 <GeKo> very useful and we got some fish
13:16:01 <juga> heheh
13:16:17 <GeKo> this week i try to finish the remaining bits for p183 O3.2 and O3.3
13:16:24 <GeKo> that's it from me
13:18:10 <hiro> @geko btw I am planning on addressing the issue in datastore#11 1 by 1 so it will be easier to understand what we have in comparison to metrics.tpo and what we are missing
13:18:38 <GeKo> sgtm
13:18:45 <hiro> I need to deploy onionoo with the fixes that were merged last week and fix some of the pipeline/parser issues that have come out
13:18:50 <hiro> everything else is in the pad
13:19:13 <hiro> I am also feeling a bit on the sick side today and might be afk for a bit after this sync
13:19:35 <juga> hiro: take care
13:19:42 <sarthikg[mds]> take care hiro!
13:20:14 <Rohithh[mds]> hope you feel better soon hiro
13:20:57 <hiro> thanks, is there anything else for today's sync?
13:21:19 <GeKo> not from my side, thanks!
13:21:31 * juga is good
13:21:40 <Rohithh[mds]> I'm groot
13:22:09 <sarthikg[mds]> i'm groot too
13:22:21 * hiro is groot!
13:22:26 <hiro> #endmeeting