13:04:46 #startmeeting Network Health 2026-01-12 13:04:46 Meeting started Mon Jan 12 13:04:46 2026 UTC. The chair is hiro. Information about MeetBot at https://wiki.debian.org/MeetBot. 13:04:46 Useful Commands: #action #agreed #help #info #idea #link #topic. 13:07:33 i guess i can start 13:07:45 i did the backlog thing after back from the break 13:08:04 and spent a bunch of time in bad-relay land and on the anomaly-report 13:08:16 both will be my focus for this week as well 13:08:21 I think I have managed: https://pad.riseup.net/p/tor-nethealthteam-2026-keep 13:08:34 ok welcome back everyone! 13:09:04 hi 13:09:06 who does want to start with the weekly updates? ok great 13:09:07 and that's it for me 13:09:12 oh 13:09:31 sarthikg[mds]: i am still eager to get access to relay statuses 13:09:43 do you have another eta update for that? 13:10:48 GeKo: hey. so the pipeline was ready, but fetching the older statuses for relays was very slow with the current clickhouse tables. the pipeline works fine for bridges currently 13:11:57 i discussed with hiro, and seems like the changes to insert jsons in clickhouse, paired with some work on calculating the last_changed_or_address_port would get this back online 13:12:11 I have the change ready, and I'll try to deploy today 13:12:17 so we can start testing 13:12:26 kk 13:13:55 GeKo (IRC): if you want to get started with bridge_statuses, I can individually start that pipeline for now 13:14:26 sarthikg[mds]: no need to for. i need the relay ones :) 13:14:29 but thanks 13:14:42 sure 13:16:15 oook! sarthikg do you want to go next? 13:16:40 my update: i had error-handling implemented for datamodel, aggregator & nsa. I implemented a few graphs in nsa, and am currently working on getting the aggregator working asap. 13:18:25 Thanks! 13:18:34 i finished collector-rs mvp and this week i'll start parsing descriptors and continue with the implementation for Tor anomalies 13:18:48 yay! thanks juga 13:19:02 np 13:19:08 regarding the issues with the firewall on metricsdb let me know if I can help w tpa 13:19:45 hiro: ok, currently investigating if the request arrives to web server (doesn't look like) 13:20:27 but the onionperfs expose the file via https, so that shouldn't be blocked right? 13:20:33 *files 13:20:33 i don't need to open a port, but to see whether there's some restriction blocking the request 13:20:38 right 13:20:47 in theory at least 13:20:58 but for some reason is not arriving from rust and metricsdb-03 13:21:03 and the onionperf you are check does have files right? because maybe we have an issue with the instance? 13:21:19 it arrives from other machine and same machine with curl 13:21:33 no, the instance is fine 13:21:33 oh oooook! weird 13:21:40 exactly 13:21:41 it's one of those bugs I guess 13:23:58 ok, my update is working through backlogs, helping Sarthik get all the data he needs and work on metrics.tpo bugs 13:24:33 PSA: this week the p183 sync is at 6PM UTC on Tue (tomorrow) 13:24:59 i won't be able to attend it, but will leave written down what i did 13:25:06 yep sounds good 13:25:29 sarthikg: if that's too late for you, you do not have to attend.. I can give your update 13:26:05 i should be able to attend this one, no worries 13:26:13 ok! if everyone is groot, we can go back to chat async, unless ofc someone has something to bring up for the meeting 13:26:28 * juga is fine 13:27:51 From my side I'm currently working on nsa bandwidth route for relays 13:28:06 i am fine 13:28:15 i am good too 13:28:38 Thanks Rohithh let us know if you need anything. and thanks for contributing 13:28:52 Sure ma'am 13:29:04 I am groot as well 13:30:07 oook thank you everyone! 13:30:11 #endmeeting