16:58:22 <micah> #startmeeting Network team meeting, 27 March 2023
16:58:22 <MeetBot> Meeting started Mon Mar 27 16:58:22 2023 UTC.  The chair is micah. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:58:22 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:58:33 <Diziet> o/
16:58:37 <nickm> hihi
16:58:48 <mbeth> hiya
16:58:49 <gabi> hello
16:58:50 <micah> hi all, @ahf is unable to make it today, so I'll try to fill his shoes.
16:59:07 <jnewsome> o/
16:59:31 <micah> The agenda will be slightly different from what is usually done, since he is out, so the meeting might be quick
16:59:41 <Diziet> quick meeting <3
16:59:56 <nickm> (keen)
17:00:09 <micah> so first item: @ahf asked that we see if there are any announcements or discussion points that should be raised
17:01:07 <nickm> qq for gabi and diziet : are we all going ot be at the arti convesation time tomorrow?
17:01:25 <gabi> sure
17:01:32 <nickm> I'm hoping to check in about status and next steps wrt onion service work, since I'm a little disoriented
17:01:40 <Diziet> nickm: Would it be possible to move it ?
17:01:50 <Diziet> I discover that the timezone change has caused me to have a clash...
17:02:04 <nickm> we should probably reschedule it in general then.  But for tomorrow I could do earlier
17:02:05 <Diziet> (Not a regular thing, just this week.)
17:02:08 <nickm> oh ah
17:02:21 <Diziet> Earlier would suit me nicely, but I'm generally flexible.
17:02:32 <nickm> any time from 1300-1600 UTC tomorrow works nice for me. Is there a good time for you two?
17:02:45 <nickm> err, 1300-1800 UTC
17:02:48 <Diziet> (timezone thing> and a bug in my phone's calendar app which caused all my meetings to lose their Icelandic status.)
17:03:19 <Diziet> 1500 ?  Is that in the way of anyone's breakfast/lunch ?
17:03:20 <gabi> I don't have a preference, any time between 1300-1800 UTC is ok
17:03:29 <gabi> sgtm
17:03:34 <nickm> ok. 1500 wfm
17:03:41 <nickm> see you then!
17:03:45 <Diziet> Cool thanks
17:03:48 <gabi> ty
17:03:57 <Diziet> Sorry to be last minute confusion
17:04:18 <micah> ok, nice - any other discussion points?
17:04:37 <micah> if not, we could move to mikeperry s61 sync
17:05:04 <micah> mikeperry: are you available to do ^ ?
17:05:08 <mikeperry> yah
17:05:30 <nickm> Diziet: np, glad I asked :)
17:05:38 <Diziet> nickm: Me too.
17:05:41 <Diziet> OK, I think as a non-s61 bod I will take my leave, thanks all :-).
17:05:45 <micah> ok, sounds like there is nothing else that we need to talk about right now, so lets move onto s61
17:05:49 <micah> o/ Diziet
17:06:10 <micah> take it away mikeperry
17:06:30 <mikeperry> so the main thing for the group is that it looks like we will not have the monthly sync next week, but we still need to do the quarterly report for what was done in the quarter (due at the end of april)
17:06:41 <mikeperry> so I am thinking we just do that async on the pad
17:07:59 <mikeperry> the stuff wrt the audit and the timelines for it are firming up. micah says that will start mid-april
17:08:55 <mikeperry> only one org is doing all of it, so we can have them look at stuff that is already stabilized first, so there is less churn there for stuff that may still have bugs being fixed
17:09:05 <micah> yeah, it looks like we are going to go with Radically Open Security for the audit, and their availability suggests April 17th as the first they can being
17:09:08 <micah> begin
17:09:37 <micah> +1 to prioritizing the stabilized parts
17:10:32 <mikeperry> so it might be good to start thinking where that line is for various objectives' work
17:11:20 <GeKo> didn't we say anyway the sync in march was the last bbb sync? ;)
17:11:29 <mikeperry> yeah
17:11:35 <GeKo> project over, case closed :p
17:11:36 <micah> that was before the NCE I believe?
17:13:00 <mikeperry> so for work on things right now, dgoulet and I are still harvesting protocol warns and BUG() macros and such things from shadow. the pattern seems to be "fix two bugs and one more emerges"
17:13:08 <mikeperry> so that is at least a downward trend :)
17:15:07 <mikeperry> juga+geko: how is the ratio analysis wrt onbrisca going? also any new issues?
17:15:16 <GeKo> https://gitlab.torproject.org/tpo/network-health/onbasca/-/issues/152#note_2888840 looks nicely done
17:15:23 <GeKo> if you want to have a look
17:15:32 <micah> It sounds like a plan to skip the next monthly s61 sync but to be sure that the quarterly update pad is tended for the report
17:15:40 <GeKo> we think we could try a ratio of 0.9 and are likely find
17:15:42 <GeKo> *fine
17:15:50 <GeKo> and no new issues
17:15:59 <GeKo> juga just tagged sbws 1.6.0
17:16:15 <GeKo> we believe we are essentially done here
17:16:17 <juga> yes, only missing the .deb package
17:16:27 <GeKo> \o/
17:16:32 <micah> yay
17:17:03 <mikeperry> ok cool. yeah that sounds great. my instinct would be to try a lower ratio and see if that improves the stability more. so yeah maybe just trying 0.5 and 0.9 both with this kind of analysis and verify that it does what we expect
17:17:25 <mikeperry> (0.5 should have less variance in bridges marked slow, 0.9 would have more change)
17:17:52 <mikeperry> also less total bridges marked, obviously with 0.5, and more with 0.9
17:18:11 <juga> should we try each ratio for something like a week?
17:18:22 <mikeperry> yeah that sounds good
17:18:26 <juga> ok
17:19:34 <mikeperry> so my next moves will be to stare at this latest new emergent bug in shadow, and see if I can add more loglines or whatever for it or sth
17:19:57 <mikeperry> and chat with dgoulet tomorrow
17:20:32 <mikeperry> the ddos also is somewhat intermittent, btw. we saw some effects of this on what looks like KIST+circewma handling on relays
17:20:53 <GeKo> yeah
17:21:06 <mikeperry> when there is a spike of circuits, CC performance degrades, and the KIST loop is our prime suspect
17:21:19 <mikeperry> this is not something we can really fix in C-Tor though. it will have to be an arti-relay thing
17:22:34 <mikeperry> ok any questions or other issues?
17:22:40 * juga is good
17:22:43 <GeKo> i am good
17:23:24 <mikeperry> ok I think we're done here then
17:23:28 <micah> great!
17:23:43 <micah> I think that can close out our meeting then, last chance for anyone to bring anything up
17:24:00 <micah> ahf will return as your regularly scheduled meeting wrangler next week.
17:24:28 <micah> With that, I bid you adieu, thanks everyone for the productive meeting :)
17:24:30 <micah> #endmeeting