17:00:21 #startmeeting anti-censorship weekly checkin 2019-09-05 17:00:21 Meeting started Thu Sep 5 17:00:21 2019 UTC. The chair is phw. Information about MeetBot at http://wiki.debian.org/MeetBot. 17:00:21 Useful Commands: #action #agreed #help #info #idea #link #topic. 17:00:22 * Samdney is watching during other work. 17:00:33 here's our meeting pad: https://pad.riseup.net/p/tor-censorship-2019-keep 17:01:51 gaba: is the bug smash fund announcement yours? 17:02:16 i wonder if we should really tag *all* non-sponsor tickets with BugSmashFund? 17:02:30 hi! 17:02:33 yes 17:02:47 we should tag all the tickets we are working on 17:02:50 that are non-sponsor 17:02:55 all bugs if possible :) 17:03:23 can we do this with a trac query? that is, select all tickets that don't have a sponsor and slap this tag on them? 17:04:06 that could be but it would be good to get bugs that we are working in the near future 17:04:09 so there is some progress 17:04:17 the idea is to report some progress in future blogposts 17:04:37 is it just for ones categorizes as "defect"? 17:04:46 yes 17:04:59 cool 17:05:28 ok, i'll keep that in mind 17:06:01 ok 17:06:33 next item on our list is #19332 17:06:43 I also added that one 17:06:53 is just checking as we do not have that one in the roadmap 17:07:28 i see. i think it should be on the roadmap -- and it won't require a ton of work from our side 17:07:51 basically, we need to be doing some work with logrotate and apache on our side to expose bridgedb's metrics to the metrics team 17:08:32 yes 17:08:33 ok 17:08:37 should we add it then? 17:08:38 i'll try to get this done in the next few days, so karsten can work on it next week 17:08:42 ok 17:08:49 sure, let's add it 17:08:54 no rush as karsten is busy with other things net week 17:10:03 i just added it 17:10:08 thanks gaba 17:10:17 any more discussion items before we move on to the review section? 17:11:22 a brief announcement from my side: so far, we got ~40 new obfs4 bridges because of our bridge campaign 17:11:43 and a few people helped us find bugs in our setup guide. and somebody even contributed a new guide for fedora 17:12:00 nice \o/ 17:12:07 yay \o/ 17:12:25 phw: approximately how many bridges are misconfigured where you have to follow up with them? 17:12:53 that would be interesting to keep track of 17:12:54 nice! 17:14:16 cohosh: only a handful. i remember somebody following a third-party guide that didn't work, and somebody set up relays (!) that ran obfs4, which was also unexpected 17:14:55 woah 17:15:21 i think our guides are in really good shape now and we should keep it that way 17:15:34 ...and discourage people from publishing their own guides, and instead just point to ours 17:15:35 nice 17:16:29 ok, let's move on to reviews 17:16:42 #31522 has no owner and no reviewer but is on review state 17:17:02 i only have one (#28533) and took the liberty of making cohosh the reviewer 17:17:17 I'll take #31522 out of review. 17:17:18 phw: cool 17:17:42 ok 17:17:59 whoops, thanks dcf 17:19:08 i'd like some help with #31425 but i really don't know what we need there 17:19:35 i've followed some debugging steps but the problems are intermittent and don't seem to have anything to do with the actual broker code 17:19:44 I don't know there. I haven't experienced any slowness myself, other than what looks like high network latency. 17:19:55 Your idea about VM deprioritization may be correct. 17:20:22 We had a sysmon outage detected yesterday again when cypherpunks reported having problems 17:20:31 but i didn't catch it early enough to do more debugging 17:21:27 One option is to go ahead with moving the broker to the Amsterdam data center, which we were going to do for IPv6 anyway (#29258) and see if that fixes things. 17:21:48 i like that option since we are doing it anyway 17:22:18 The migration in #29258 is stalled because of a deadlock with the decision whether to move the broker to a TPA machine (can't find the ticket just now), but I think that's an anti-pattern. 17:22:23 No reason we can't do one now and the other later. 17:22:49 what is the cost of moving data centres? 17:23:04 if it's not a big effort, i think we should go ahead and move it now 17:23:05 Just the time it takes to set up a new machine. 17:23:24 Also a small bit of log file overlap that we'll have to figure out what to do with. 17:23:41 i think the vague decision of making it a TPA machine was: hold off on doing that until we have a really good reason to 17:24:13 #31232 was the ticket 17:24:55 Didn't find it at https://trac.torproject.org/projects/tor/wiki/doc/Snowflake#Tickets 'cuz it was missing a snowflake keyword :) 17:25:28 thanks for fixing that :) 17:25:56 Okay, I'll comment on #31425 that the idea is to try the broker migration we were going to do for IPv6 and see if that helps. I'll try to get a bigger VM this time too. 17:26:05 yeah we keep sort of discussing this and then leaving it for later >.< 17:26:16 dcf1: sounds good, thanks! 17:27:16 dcf1: and since you run the current infrastructure, feel free to weigh in on moving it to TPA if you no longer wish to do that 17:27:47 i think we're on this "we don't run the network" vs. "what happens if we cant get ahold of you" trade off right now 17:29:04 hiro: you do not have any review that needs to be added/discussed here? 17:29:09 This is also something you could usurp from me at any time if I were to disappear -- you would need to set up your own machine and push out a new release with new hostnames, but it wouldn't be broken permanently. 17:29:20 true 17:29:51 gaba: i put reviewing the metrics stuff for gettor on my tasks this week 17:29:59 (i think it's gitlab ticket 10 or something) 17:30:00 ah, ok. thanks cohosh 17:31:39 gaba: I am doing the ansible configuration and the docs at the moment 17:32:11 ok 17:32:35 I think I will need help reviewing the info on the website too 17:32:46 before chaging what we have on gettor.tp.o 17:33:36 but I am not writing any code atm 17:34:50 dcf1: i'll send a reminder to sina for #31455 17:35:27 If you do, Cc arma, who offered to get in touch last meeting. 17:35:33 gotcha 17:37:13 ok, are we done for today? 17:38:04 can people quickly look at the roadmap and see that what you are working on and planning to work on is reflected there? 17:38:32 * phw nods 17:38:37 and there are some stuff in the 'needs help with' section of each person. Not sure if you went through it phw 17:38:39 gaba: thanks for the reminder, doing that now 17:39:00 thanks 17:40:07 hiro: i haven't heard from arma2 yet regarding archive.org 17:40:43 ok 17:40:46 we can go ahead with uploading but let's not make gettor point people to archive.org just yet 17:40:49 I can start setting up the account 17:40:56 and prepare the upload scripts yes 17:41:10 and regarding gettor-web: do you need a review for the entire website, or just some changes? 17:41:46 I have used the styleguide for the website.. I think I will need review for just the content of the main page which is the user doc 17:41:47 looks like it's the entire website (i haven't looked at the repo yet) 17:42:01 the style is the same we use everywhere 17:42:10 ok, i'll review it 17:42:25 hiro: do you know the https://archive.org/details/@torproject account? That is what boklm plans to use for archiving archive.tpo. But for gettor maybe you want to structure the item/items differently. 17:42:54 I know we ahve an account but I wasn't sure we should use the same or another one 17:43:13 hiro: yes, gettor is somewhat different and maybe should use a separate account. 17:43:29 phw: please wait on the review tomorrow I will send some more update 17:43:36 hiro: gotcha 17:43:37 I'll send a status update tomorrow 17:43:47 Especially as new Tor Browser releases are not yet uploaded automatically under @torproject. 17:45:15 dcf1 will tor browser releases be updated to archive.org? 17:45:27 because that's all we need for gettor 17:45:36 so we might use the same account if that's the case? 17:45:41 hiro: I think that's the long-term plan but I don't think there are any plans to do it quickly, so better not to wait on it. 17:45:48 ok 17:47:02 cohosh: do we have a reviewer for #29206? 17:47:16 phw: yep dcf1 is reviewing it 17:47:35 oh, good. did we miss any reviews? 17:48:31 i don't think so. let's end the meeting 17:48:33 #endmeeting