15:01:16 <aguestuser> #startmeeting Application Teams Meeting 2022-03-14 15:01:16 <MeetBot> Meeting started Mon Mar 14 15:01:16 2022 UTC. The chair is aguestuser. Information about MeetBot at http://wiki.debian.org/MeetBot. 15:01:16 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic. 15:01:40 <aguestuser> here is pad: https://pad.riseup.net/p/tor-tbb-keep 15:01:55 <aguestuser> let's take a minute to fill in last week / this week etc... 15:02:06 <Jeremy_Rand_Talos_> hi! 15:03:28 <aguestuser> also open prompt for iscussion topics: any blockers, questions, or recent learnings you think might be relevant to the rest of the group? put 'em in the top of the pad! :) 15:07:25 <aguestuser> okay... 15:07:31 <aguestuser> seems like we're about ready to roll? 15:08:59 <aguestuser> okay 15:09:20 <aguestuser> so thought i'd start off with Jeremy_Rand_Talos_ 's scheduling question 15:09:22 <aguestuser> as it's easy 15:09:47 <aguestuser> org-wide policy is to keep all meetings at whatever UTC time they are scheduled no matter the daylight-savings time status 15:10:02 <aguestuser> anyone super upset about that? 15:10:21 <aguestuser> resolved! 15:10:23 <boklm> I'm fine with both 15:10:24 <PieroV> nope, it work for me 15:10:35 <PieroV> the release meeting is a bit worse, but I can make it work :) 15:10:37 <Jeremy_Rand_Talos_> aguestuser, is that a new policy? I like it, but it is different from what we did in the past. 15:10:38 <aguestuser> sorry. should have left a bit more space 15:11:14 <aguestuser> Jeremy_Rand_Talos_: i don't recall the IRC chat in which i saw this mentioned, but there was an email, etc... 15:11:23 <boklm> (in the past we usually changed it to stay the same local time in europe/america) 15:11:25 <aguestuser> also: people change time zones at different times all across the world 15:11:56 <Jeremy_Rand_Talos_> aguestuser, ok. I'm not on any internal mailing lists, so I will take your word for it. 15:12:05 <sysrqb> it is the new policy as of last ~November when we didn't adjust meeting times 15:12:27 <aguestuser> oh, okay! i just observed arma2 say this casually (i think on #tor-internal? and possibly also mentioned on mailing list) 15:12:28 <Jeremy_Rand_Talos_> sysrqb, I see, thanks for clarifying 15:12:31 <sysrqb> basically, some teams didn't want to move and agreeing on new times is hard 15:12:41 <sysrqb> so that forced most other teams to remain at their UTC time 15:13:03 <Jeremy_Rand_Talos_> Like I said, I like this new policy, just wanted to make sure everyone's on the same page 15:13:10 <sysrqb> yep :) 15:13:35 <aguestuser> okay, so: is everyone okay with observing the org-wide policy in this instance on this team? 15:13:42 <PieroV> yes 15:13:45 <boklm> yes 15:14:05 <aguestuser> hearing no objections... 15:14:18 <aguestuser> ... resolved! 15:14:26 <Jeremy_Rand_Talos_> Great! 15:14:56 <aguestuser> okay, other things on here are basically announcements 15:15:22 <aguestuser> i wanted to make sure folks (esp donuts!) caught the fact that 11.0.8 stable TBA release went out late friday 15:15:48 <aguestuser> special thanks to lavamind for staffing the blog rollout super-duper-late euro-time! 15:16:07 <boklm> I think if we do builds on wed, it's difficult to release on thursday 15:16:15 <aguestuser> maybe a small micro-retro there: expecting a from-scratch android build to run in ~5 hours on the build machine is overlly optimistic! 15:16:27 <boklm> with time for signing it will probably more likely on friday 15:16:35 <aguestuser> boklm: is this about the next release? 15:16:43 <aguestuser> i was doing a quick check-in on the last one 15:16:48 <boklm> about 11.0.9 15:17:06 <aguestuser> the only thing i wanted to touch on was PieroV: i think we leanred that you going ahead and doing the second build on your machine was a good idea! 15:17:18 <boklm> about the discussion topic saying "wed build, thu stable release (11.0.9) fri alpha release (11.5a8)" 15:17:39 <aguestuser> given the build machine took ~16hr to run the android build! 15:18:08 <aguestuser> boklm: gotcah -- i was starting above that at "Announcements" 15:18:19 <PieroV> aguestuser: it's okay for me, but I think that it took that long because Richard needed to build everything from 0 15:18:21 <aguestuser> (and then i suppose abusing the "Announcements" aspect to actually discuss the thing) 15:18:25 <boklm> ah yes, sorry 15:18:31 <PieroV> like the compilers as well 15:19:12 <aguestuser> PieroV: right, but if that's an antipatable delay (and we know now it can cause workday-lenght unexpected delays, just something to keep in mind if we've got a build we're trying to get out the door on a tight turnaround?) 15:19:33 <sysrqb> aguestuser: (to be fair, richard building on the build machine should be much faster next time now that their dependencies are built) 15:20:06 <aguestuser> upshot: it meant that tpo/web folx were having to singn off on MRs around midnight their time to meet the deadline rather than in the middle of their workday, so if i could anticipate and make adjustments to avoid that next time i'd like to 15:20:41 <aguestuser> PieroV: true true. in any case, sort of just a "let's keep this in mind next time" thing. (since i didn't know to keep it in mind the last time.) 15:20:42 <PieroV> I think lavamind is based in Canada, I don't know if any tpo/web is in Europe 15:20:43 <aguestuser> moving on! 15:21:08 * Jeremy_Rand_Talos_ notes that adding Cirrus as a builder (still WIP) might result in some improvement here 15:21:13 <aguestuser> unless anyone else has ideas for optimizing that? 15:21:24 <aguestuser> ... 15:21:37 <boklm> I think being able to merge things directly (like was the case in the past) without having to wait for someone from web team to do it would be nice 15:22:14 <aguestuser> boklm: +1. and lavamind mentioned that was on TPA discussion agenda for this week, so should be an optimization in play for future releases. c 15:22:19 <aguestuser> i can make a follow-up note! 15:23:36 <aguestuser> follo-wup note mad! 15:23:47 <aguestuser> okay... i think that's enough micro-retrospective! 15:24:21 <aguestuser> the build is out. the chemspill is mopped up. TBA v96 is live on all releases. huzzah 15:24:30 <aguestuser> PieroV: anything you want to flag about time off? 15:24:37 <sysrqb> \o/ 15:24:48 <aguestuser> \o/ indeed! :) 15:24:54 <boklm> \o/ 15:25:28 <PieroV> No, I wanted just to update that I am taking also next Monday (last week I told you about 24 and 25) 15:25:44 <aguestuser> so, just note PieroV out 24-Mar to 28-Mar. have lovely vaycay time! :):):) 15:26:00 <PieroV> Thanks! I will meet folks from Guardian in Vienna :( 15:26:03 <PieroV> * :) 15:26:04 <aguestuser> funnnn 15:26:19 <aguestuser> say hi to n8fr8 plz! :) 15:26:24 <PieroV> sure 15:26:36 <aguestuser> next... 15:26:39 <Jeremy_Rand_Talos_> PieroV, enjoy! Looking forward to next time I can travel for vacation.... 15:26:54 <PieroV> thank you! 15:27:01 <aguestuser> openssl-update-driven android+desktop release this week! 15:27:11 <aguestuser> boklm... 15:27:17 <aguestuser> (sorry for long digressions!) 15:27:23 <aguestuser> thoughts on timeline? 15:28:18 <aguestuser> i was working off what richard had in nc... 15:28:21 <boklm> I think if we do builds on wed, releasing on thu is too optimistic 15:28:25 <aguestuser> (which i am having a hard time generating link for) 15:28:49 <aguestuser> boklm: you are right i mis-read calendar 15:28:50 <boklm> signing/upload takes a lot of time 15:28:59 <aguestuser> can folks read this link? https://nc.torproject.net/remote.php/dav/calendars/aguestuser/a-test-calendar_shared_by_richard/ 15:29:03 <aguestuser> (i think that is wrong link) 15:29:13 <aguestuser> i can type out everything it says but will be faster if we can all look at it) 15:29:25 <PieroV> I can't 15:29:31 <aguestuser> it says 11.09 build on tue 15-Mar 15:29:41 <aguestuser> 11.0.9 15:29:46 <sysrqb> https://nc.torproject.net/apps/calendar/p/Dy5spytzmYoJPieT 15:29:51 <sysrqb> (linked on https://gitlab.torproject.org/tpo/applications/tor-browser/-/wikis/Release-Schedule) 15:29:55 <aguestuser> sysrqb: thx! 15:30:26 <PieroV> (adding an item in the pad, for #1758156) 15:30:55 <aguestuser> boklm: can you see that? (more importantly: does it look realistic to you?) 15:31:21 * boklm is still trying to load the calendar 15:31:28 <aguestuser> stable(11.0.9): build tue, sign wed, release thu 15:31:52 <aguestuser> alpha (11.5a8): build wed, sign thu, release fri 15:32:10 <aguestuser> (both are intended as desktop + android builds) 15:32:15 <boklm> I think stable release will be late thu or friday morning 15:32:42 <aguestuser> noted! 15:32:45 <aguestuser> alpha? 15:32:47 <boklm> signing/upload takes 1.5 or 2 days 15:32:55 <Jeremy_Rand_Talos_> All dates in the calendar link from the wiki appear blank? Not sure if just taking ages to load due to disabled JIT. 15:33:17 <PieroV> It doesn't load to me either 15:33:23 <boklm> (also with tpo/tpa/team#40657 upolad takes a lot of time) 15:34:00 <aguestuser> boklm: so, if we wanted to get both out by friday, we'd realistically need to start building today... 15:34:07 <aguestuser> sysrqb: when is openssl bump? 15:34:14 <boklm> and for alpha I think release will be monday or tuesday 15:34:20 <boklm> openssl bump is tomorrow 15:34:27 <aguestuser> okay then! 15:34:34 <boklm> https://mta.openssl.org/pipermail/openssl-announce/2022-March/000216.html 15:34:37 <aguestuser> so a revised (more realistic) timeline: 15:35:02 <aguestuser> stable (11.0.9): build tue, sign wed, release fri 15:35:19 <aguestuser> alpha (11.5a8): build wed, sign thu, release mon 15:35:33 <boklm> ok 15:35:44 <aguestuser> /s/mon/mon->tue 15:36:10 <aguestuser> boklm: on process for the "joint release" aspect of this 15:37:06 <aguestuser> i think it is: 15:37:46 <aguestuser> (1) once you have built and pushed some unsigned artifacts somewhere 15:38:05 <aguestuser> (2) sysrqb signs the apks with the signing key and uploads them to people.tpo 15:38:17 <aguestuser> (3) i put apks on dist.tpo and playstore as normal 15:38:33 <aguestuser> (4) i defer on changelog updates / blog to richard? 15:38:40 <aguestuser> is that about right? 15:39:26 <sysrqb> i recommend you (aguestuser) only publish the apks on the play store 15:39:36 <boklm> I can do the upload of apks on dist.tpo at the same time I put the desktop files there 15:39:36 <sysrqb> and let richard/boklm publish everything on dist.tpo 15:39:44 <aguestuser> chill! 15:40:29 <aguestuser> okay, so we still need a touchpoint betweeen boklm finishes the build -> sysrqb signs the apks and sysrqb signs the apks -> aguestuser uploads apks to playstore 15:40:30 <aguestuser> but that's it 15:40:33 <aguestuser> yes? 15:40:52 <boklm> yes 15:41:14 <aguestuser> lovely. and do we need to do any work in this meeting to figure out who the builders are? (assuming richard and boklm?) 15:41:36 <boklm> and when we publish blog post so you publish on playstore around the same time 15:41:58 <PieroV> I can also build, if needed, just let me know 15:42:10 <aguestuser> boklm, PieroV: sounds good! 15:42:14 <boklm> I can do builds too 15:42:28 <aguestuser> okay, so i think that covers all of the things we wrote down 15:42:34 <PieroV> I added a point 15:42:40 <aguestuser> you did! sorry! 15:42:44 <aguestuser> PieroV: floor is yours! 15:42:57 <PieroV> 1758156 has been published 15:43:10 <aguestuser> ah yes, the hanger-on to last week's chemspill 15:43:16 <PieroV> So I think we could cherry-pick it for this stable, instead of waiting another week 15:43:26 <aguestuser> agreed! 15:43:34 <PieroV> https://hg.mozilla.org/releases/mozilla-esr91/rev/6b663b8c6eca 15:43:53 <PieroV> We have some time, since we are waiting for openssl anyway 15:44:30 <aguestuser> PieroV: if you land this in desktop, i will land it in TBA? 15:44:50 <boklm> desktop and TBA are using different branches 15:44:51 <PieroV> works for me, who is reviewing it? 15:45:22 <boklm> ah, I misread the question 15:45:50 <aguestuser> boklm: it's just cherry-picking this commit onto the relevant branches... which should be okay? 15:46:02 <boklm> yes 15:46:27 <PieroV> aguestuser: are we on v96 on Android, right? 15:46:29 <aguestuser> and we could link the issue that i think richard already made to the release tickets for 11.0.8 and 11.5a8 to make sure it gets done before release? 15:46:33 <aguestuser> PieroV: correct! 15:47:02 <PieroV> then I think Moz didn't backport to 96, but it should be a trivial cherry-pick 15:47:12 <aguestuser> kk 15:47:39 * aguestuser digging for the MR for landing this fix... 15:48:05 * aguestuser err the issue.. 15:48:54 <aguestuser> well... i will find it after meeting 15:49:09 <aguestuser> (unless someone else has it handy. wanted to link it in discussion notes) 15:49:33 <PieroV> Alpha: https://gitlab.torproject.org/tpo/applications/tor-browser-build/-/issues/40452 15:49:47 <PieroV> 11.0.9 stable: https://gitlab.torproject.org/tpo/applications/tor-browser-build/-/issues/40453 15:49:51 <aguestuser> PieroV: i meant the (likely still confidential) issue about the fix 15:49:59 <aguestuser> from upstream 15:50:08 <PieroV> oh sorry 15:50:16 <aguestuser> nw! 15:50:49 <boklm> https://bugzilla.mozilla.org/show_bug.cgi?id=1758156 ? 15:51:28 <aguestuser> boklm: yep richard had made an issue about incorporating that fix once it landed upstream. thought it would be nice to link that issue to release prep issues 15:51:29 <boklm> (linked from the hg.mozilla.org commit page) 15:51:37 <aguestuser> don't need to take up meeting time with it tho! 15:51:59 <boklm> should we make an issue on our side? 15:52:13 <Jeremy_Rand_Talos_> boklm, that Bugzilla link is still restricted 15:52:21 <PieroV> boklm: we did it for the other issues 15:52:27 <boklm> ah, or richard made one already? 15:52:34 <aguestuser> boklm: i thiiiink so 15:52:41 <aguestuser> but i am having a hard time finding it 15:53:12 <PieroV> I can't find it either 15:53:16 <aguestuser> aha! 15:53:36 <PieroV> https://gitlab.torproject.org/tpo/applications/tor-browser/-/issues/40830 15:53:39 <PieroV> It's already closed 15:54:07 <aguestuser> k 15:54:28 <PieroV> (I think we could reopen it, and it doesn't need to be confidential - it just says to cherry pick the same commit) 15:54:37 <aguestuser> PieroV: cool! 15:54:47 <aguestuser> (fyi -- i get 404 trying to follow that link?) 15:54:56 <boklm> you need to be logged in 15:54:59 <aguestuser> (but signed in) 15:56:02 <aguestuser> ok, at any rate: PieroV: can you handle munging that issue and linking it to release prep tickets? 15:56:12 <PieroV> sure 15:56:16 <aguestuser> thank you! 15:56:21 <aguestuser> okay... closing in on time 15:56:42 <aguestuser> anyone have something pressing for the good of the group? 15:57:11 <PieroV> I don't 15:57:20 <aguestuser> hearing nothing: would anyone mind closing us out by telling us about something inspiring/amusing/not-apocalyptic that happened to them this week? 15:58:15 <aguestuser> how about: i went to a bar with a lot of smiling people not wearing masks like in the before-times! they had a pin-ball machine and i played it! 15:58:24 <aguestuser> check. :) 15:58:29 <aguestuser> alright. with that... 15:58:30 <PieroV> sorta 15:58:33 <Jeremy_Rand_Talos_> :) 15:58:34 <PieroV> I had a graduation party 15:58:35 <aguestuser> oh. tell! 15:58:38 <aguestuser> wee! 15:58:41 <PieroV> for a friend's of mine 15:58:47 <Jeremy_Rand_Talos_> oh nice! 15:58:56 <aguestuser> where? 15:59:04 <aguestuser> or rather: what kind of environment? 15:59:04 <PieroV> in Treviso, at a bar 15:59:11 <aguestuser> :) 15:59:13 <aguestuser> huzzah 15:59:16 <aguestuser> life returns! 15:59:23 <PieroV> and we didn't wear masks for long 15:59:27 <Jeremy_Rand_Talos_> PieroV, tell your friend that some random blockchain developer they've never heard of says congrats! 15:59:34 <PieroV> as we still can't eat with them :D 15:59:40 <aguestuser> hehe 15:59:41 <PieroV> lol okay 15:59:54 <aguestuser> okay. sadly our time has come to an end. 15:59:57 <aguestuser> have great weeks all! 16:00:00 <aguestuser> see you in #tor-dev 16:00:03 <PieroV> thanks, same to you! 16:00:04 <PieroV> o/ 16:00:06 <aguestuser> #endmeeting