12:08:45 <hiro> #startmeeting network-health 2024-11-04
12:08:45 <MeetBot> Meeting started Mon Nov  4 12:08:45 2024 UTC.  The chair is hiro. Information about MeetBot at http://wiki.debian.org/MeetBot.
12:08:45 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
12:08:59 <hiro> yeah here too and I was taking water out of things hehe
12:09:08 <juga> oh shit
12:09:25 <hiro> it is fine the sun is out now and lots of things were full of water in the patio
12:09:28 <juga> hope not too bad
12:09:37 <juga> ok
12:09:41 <hiro> lost track of time sorry
12:10:03 <hiro> how is everyone doing? :)
12:10:36 <GeKo> i am doing :)
12:11:05 <juga> no worries
12:11:12 <juga> doing :)
12:11:26 <sarthikg[m]> doing good!
12:11:28 <hiro> my update is that last week I spent (or lost) a lot of time with collector... and my patch actually makes things worse
12:11:52 <hiro> and this week I am going to ask tpa for some ideas and maybe buy us some time bumping resources
12:12:43 <GeKo> are we good again due to being at the begin of a new month?
12:13:08 <juga> i almost have a layered upload version that works, but focusing now on soto
12:13:18 <GeKo> nice
12:13:19 <juga> or family, let's see
12:13:39 <hiro> I think we should be still a little slow because we might still get descriptors that need to get in the old month tarball but things should be back to "normal"
12:13:40 * juga still filling the pad
12:15:47 <hiro> @juga @GeKo @sarthikg[m] any issues that you'd like to discuss?
12:16:24 <GeKo> did you have to kick the db to pick ingesting data again? or what happened?
12:16:31 <GeKo> *pick up
12:17:43 <hiro> I kicked collector
12:17:50 <juga> hiro: maybe later a a sync about soto
12:17:59 <GeKo> okay
12:18:04 <hiro> sure @juga
12:18:18 <GeKo> nothing else from my side
12:19:06 <hiro> @sarthikg[m] everything ok on your side?
12:19:48 <sarthikg[m]> yeah, completed all the setup around db normalisation, except the data :(
12:20:31 <sarthikg[m]> by any chance, is the metricsdb limited on the egress speed? it seems slow
12:21:06 <hiro> uhm I don't think it is limited as in capped. I think we do not have a lot of resources on that machine
12:21:20 <hiro> let me check
12:22:11 <sarthikg[m]> ahh, i checked the cpu, it seems to be bottlenecking
12:23:26 <hiro> we have ram but we swap a loiit
12:23:56 <hiro> we have a lot of services running there
12:24:04 <hiro> maybe we should start to split things up
12:24:22 <hiro> like taking tagtor and the api away from the db
12:25:42 <sarthikg[m]> yeah, that should help maybe. anyways, i'm writing a script for mock data on local, if the dump keeps failing, that might just work as the backup
12:26:30 <hiro> I see that's great thanks!
12:27:11 <hiro> ok if that's all we can call the meeting and continue discussing on our new channel (#tor-network-health)
12:27:22 * hiro is groot
12:27:30 <GeKo> me too
12:27:47 <sarthikg[m]> me too
12:27:58 <hiro> #endmeeting