15:57:50 #startmeeting network-health 15:57:50 Meeting started Mon Apr 12 15:57:50 2021 UTC. The chair is GeKo. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:57:50 Useful Commands: #action #agreed #help #info #idea #link #topic. 15:57:55 hello everyone! 15:58:01 o/ 15:58:08 it's been a while that we had network-health meetings 15:58:10 o/ 15:58:17 o/ 15:58:22 ah it is an irc mtg not a bbb mtg. ok 15:58:22 but now that things are back to more normal in tor land 15:58:24 (hi) 15:58:37 there is a realistic chance those meetings stay 15:58:44 so, get used to them ;) 15:58:51 we even have a pad (thanks gaba) 15:58:53 http://kfahv6wfkbezjyg4r6mlhpmieydbebr5vkok5r34ya464gqz6c44bnyd.onion/p/tor-nethealthteam-2021.1-keep 15:59:36 okay, it seems its structure follows our usual meeting structure 15:59:51 where we have discussion items added to the repsective part of the pad 16:00:09 and then update with things we worked on/plan to work on/need help with etc. 16:00:15 hi! 16:00:23 note: a large part of our work is s61 related 16:00:36 and we have a separate meeting within the network-team meeting for that 16:00:48 so, gaba, i geuss this sync is for all work but s61, right? 16:00:51 *guess 16:00:53 if you have anything to talk about in this meeting, please add it to the agenda 16:00:57 geko: right 16:01:02 great 16:01:12 so, let's update the pad i geuss 16:01:14 *guess 16:01:31 * irl loading the pad 16:02:38 o/ 16:03:03 are the reminders like '* Remember to "/me status: foo" at least once daily.' leftover from whatever team we took this text from, or is that intended? and if so where, etc 16:03:31 yes 16:03:46 i am fine doing the status thing or not 16:04:09 i recall it does/did not work too well for other teams 16:04:31 but we can try if folks are interested 16:04:52 removing it. i think for now we can start with this meetings 16:05:03 and add the planning work to the pad 16:05:46 sounds good 16:06:07 hmm, if it's only about non s61 work, i guess i should delete what i put in the pad 16:06:23 yeah, i guess so, was about asking :) 16:06:28 *about to ask 16:06:32 ok :) 16:06:52 irl: did you get the pad to load? 16:07:21 is there a network health scope somewhere? i was about to list something that i realized was probably network team 16:07:23 yes i did 16:07:44 dgoulet: do you have some spare time to help nickm help with the libressl thing? 16:08:06 arma2: tricky. 16:08:20 i think we never sat down to draw the scopes properly 16:08:27 arma2: if in doubt, just listed and then we see 16:08:58 there is some here: https://gitlab.torproject.org/tpo/network-health/team 16:09:07 GeKo: I have very little... at least we diagnosed the problem so I can help a bit but I have very little :S 16:09:09 the five areas of network health 16:10:20 dgoulet: k, no worries then 16:10:42 okay. 16:10:53 do we have anything folks want to discuss today 16:11:09 currently the discussion section is empty (which is fine) 16:11:33 but if there are urgent/less urgent things we should talk about while here, please speak up 16:11:38 i just wrote 'we have a relay operator volunteer who has volunteered to run the next relay meetup, and i think that email thread might have died without enough attention.' 16:11:43 which could use some remembering 16:12:03 arma2: which mailing list? 16:12:13 nope, just email 16:12:38 arma2: can you move that topic to discussion instead of status for you? 16:12:42 arma2: let's move this activity for end of may. i can reply to that email thread, but right now i'm overloaded with outreachy contribution period 16:12:45 the one where the author wrote "if roger thinks this is good then let's do it" 16:12:55 and everonye was waiting on this roger guy? ;) 16:13:02 gosh, really 16:13:09 well, somebody should pretend to be him so it can move forward i guess 16:13:17 k 16:13:22 sounds like yes, that one ;) 16:13:28 ggus: thanks 16:14:44 irl: is there anything we can help you with related to your tasks? 16:14:50 questions/concerns? 16:15:08 either way, thanks for stepping up and helping out, much appreciated 16:15:18 not yet, the tasks are just getting started and everything is familiar just now 16:15:38 i had a call with anarcat today to go over some things related to future maintenance and monitoring 16:15:43 nice, if not don't hesitate to ping folks :) 16:16:09 that's for the collector monitoring item? 16:16:32 yes, many metrics services are monitored by ad-hoc cron jobs 16:16:56 collector is the first one that i’ll tackle, so that we have a unified system that collects all the alerts 16:17:33 the documentation will then need to cover how serious each alert is and how to handle them 16:18:14 right now there are just cron jobs that send raw grep output from logs and unless you wrote the code you’re not likely to know how to handle that 16:18:42 where are they sending that output too right now? 16:18:51 acute’s email inbox 16:19:02 and ahf's 16:19:09 acute: hi! 16:19:11 :) 16:19:15 o/ 16:19:24 okay, we have two separate folks getting that at least, okay 16:20:10 i am planning to pipe them into a matrix room, which will fit nicely with the current prometheus setup and desire to migrate to matrix 16:20:26 we can still have individual people getting emails too 16:20:55 another thing for the future meetings: if there is an item in the meeting pad you like to talk about please mark it as bold 16:21:33 irl: well, i was more worried about it only being one person right now 16:21:51 so, having two different people is okay-ish i guess 16:22:05 as a stopgap until we have something more robust 16:22:18 sure 16:23:23 anything else we have for today? 16:23:34 i'll throw one more in: 16:23:39 i'm finding some exits that use opendns 16:23:47 we used to badexit them and contact them 16:23:55 lately a bunch have been in hong kong 16:24:04 i think there is some hong kong isp that has 10 dns servers set and 1 of them is opendns 16:24:10 GeKo: yes, i added one item on discussion agenda 16:24:15 and when i tell the relay operators they say they're not using opendns 16:24:26 i've started ignoring that and not badexiting those relays and not contacting them 16:24:38 maybe that's fine. but if you have a better idea please say so :) 16:25:06 are they running the harmful opendns version? 16:25:11 that is censoring stuff? 16:25:16 yes 16:25:21 then we should badexit them 16:25:22 there is only one opendns and it censors, i believe 16:25:34 i think there are different opendns settings 16:25:38 but either way 16:26:22 trying to get into contact with the relay op + badexiting the relay if things are not getting fixed is the way i think we should go 16:26:39 i can do that part if you pipe in the scan results :) 16:26:51 otherwise i can scan, too, i guess 16:26:53 ok. i find them because they resolve blender.io to 146.112.61.108 16:27:28 $ host 146.112.61.108 16:27:28 108.61.112.146.in-addr.arpa domain name pointer hit-phish.opendns.com. 16:27:31 so, pretty simple 16:27:40 i'll tell you a bunch after the meeting and then we'll try to keep up 16:27:46 thanks 16:27:52 ggus: okay, you are up 16:28:08 ok 16:28:43 roger wrote a document about relay operator expectations, and i will share with tor-relays@ mailing list & community 16:28:49 lemme find the ticket 16:28:59 https://gitlab.torproject.org/tpo/community/relays/-/issues/18 16:29:24 last week gaba and i moved this issue to that repo, so relay operators can use anon-ticket to give feedback 16:29:32 https://anonticket.onionize.space/ 16:30:05 this public consultation will be open till first week of may 16:30:41 sounds good 16:30:46 what are the next steps after that? 16:30:49 seems ready to me for consultation that text 16:30:51 so today or tomorrow i will send this email inviting relay ops. 16:31:16 GeKo: next step: review feedback, merge suggestions, and put on community.torproject.org/relay 16:31:37 great 16:32:28 do we have anything else for today? 16:32:48 ggus: oh, do you need any help with that or are you fine? 16:33:27 GeKo: i'm fine 16:33:48 k 16:34:39 alright it seems we are done with our first meeting after the long break 16:34:47 thanks everyone and have a nice week 16:34:51 #endmeeting