16:01:41 <morganava> #startmeeting Applications Team Meeting 2025-06-23
16:01:41 <MeetBot> Meeting started Mon Jun 23 16:01:41 2025 UTC.  The chair is morganava. Information about MeetBot at http://wiki.debian.org/MeetBot.
16:01:41 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
16:01:42 <ma1> o/
16:02:00 <morganava> good morning i hope you all had a good weekend
16:03:16 <morganava> pad as usual https://pad.riseup.net/p/tor-tbb-keep
16:04:04 <morganava> this morning i've been poking around gitlab a bit and updating things here and there
16:04:22 <morganava> coolest discovery wtih the upgrade i've found so far is project+group level comment templates
16:04:58 <morganava> so basically in comments there's a little speech bubble that let's you fill-in commonly used messages
16:05:26 <morganava> i took the liberty of adding various commonly used things around releases and backporting etc
16:06:21 <morganava> i've *also* started farming out particular audit issues
16:06:55 <morganava> obviously if you get assigned something which is obviously not your wheel-house feel free to shoot it over to someone more appropriate
16:07:30 <morganava> and *more* immediately, i've also assigned out the release+dev notes review issues
16:08:01 <morganava> almost everyone should have 2 major Firefox releases to look at
16:08:22 * dan_b goes and looks
16:08:49 <dan_b> i have 4 triage changes and 2 dev+rel notes 👍
16:08:50 <morganava> the due date on those release note reviews should the same as the issue triage (July 16)
16:09:25 <morganava> oh yeah i also took advantage of the new multiple assignees and added in relevant people to the triage issues
16:11:10 <morganava> ok that's enough announcements from me, does anyone else have anything to announce/request?
16:11:35 <PieroV> I've written a commented range-diff and attached it to the deep review issue
16:11:55 <PieroV> I haven't finished with it yet, I'm at Android... which is always the hardest part
16:12:15 <morganava> ah yes
16:12:43 <PieroV> Hopefully it should help to review at least the diffs, even though checking the commits would also be a good idea... Even though probably not something we can do for the current due date (Jun 25)
16:13:49 <PieroV> Apart from this, I'd like to take the 140 rebase
16:14:21 <PieroV> As I should've used build1 and build2 tags for 140.0a1-15.0-2, but I used only build1
16:14:51 <PieroV> Has there been any update re git repo? I've seen Moz published a very close July 7th as the eol for gecko-dev
16:14:59 <PieroV> Much sooner than we anticipated...
16:15:08 <morganava> i updated the TPA issue with that news last week
16:15:23 <morganava> last i saw on the thread TPA is asking for more specifics
16:15:36 <morganava> but boklm was unblocked on pushing because of some LFS thing iirc
16:15:49 <morganava> (to gitlab in general, not the tor-browser.git repo)
16:15:57 <boklm> yes, I pushed some test repo on gitlab
16:16:10 <PieroV> Well, LFS is always problematic, esp with forks
16:16:18 <morganava> yeaaah i hates it
16:16:34 <morganava> probably becuase i just don't undertstand it but cest la vie
16:17:10 <PieroV> So, what's the current idea? To use gecko-dev/esr140 for the first rebase, or to wait at this point, since we'll need the new history very very soon?
16:19:08 <morganava> i don't see much harm in waiting presuming TPA can be unblocked
16:19:45 <PieroV> ack. Worst case scenario if I have time before the thing gets unblocked I can somehow do it locally
16:20:36 <morganava> yeah basically
16:20:42 <morganava> worst case scenario yolo and TPA is sad
16:21:41 <morganava> is there any likely fallout for our tor-browser CI jobs?
16:22:11 <PieroV> CI is broken anyway
16:22:15 <PieroV> In 140
16:22:27 <boklm> is it disabled?
16:22:29 <morganava> well fair, but i meant more with the fetching/cloning
16:22:43 <PieroV> We have a ocal clone in the runners
16:22:50 <PieroV> It'll need to update, but it's a single fetch
16:23:06 <morganava> do we have an answer for anarcat's q in https://gitlab.torproject.org/tpo/tpa/team/-/issues/42129#note_3212865 ?
16:23:06 <PieroV> And lints also do a single fetch now
16:23:21 * anarcat perks his ears
16:23:22 <dan_b> CI i think was double broken but I ran jobs for lavamind last week till at least the 128 runners were unbroken
16:23:35 <morganava> ie do we have a repo with merged history (guessing the one boklm pushed) and a list of repos the gc commadn will need to run (guessing all the tor-browesr/mulvad-browser forks?)
16:23:58 <PieroV> boklm's repo for GC
16:24:09 <PieroV> And Tor Browser forks... well, it depends
16:24:17 <PieroV> Only the ones that were born private
16:24:28 <PieroV> The rest should share a common pool with tor-browser.git
16:25:32 <morganava> is the gc command something needs to be run before or after the new branches are added?
16:25:33 <boklm> so I think gc needs to be run on tor-browser.git and mullvad-browser.git, which should automatically apply to the forks if they share a common pool
16:25:43 <boklm> it's after the new branches are added
16:25:55 <morganava> ok, i'll follow up on the issue with this info
16:26:42 <morganava> alright, is there anything esle?
16:26:49 <boklm> maybe we can also run gc locally before pushing the new branch, which might improve things (although I'm not sure)
16:26:49 <PieroV> We need to relprep
16:27:02 <PieroV> 14.5.something going out tomorrow
16:27:31 <morganava> argh ack and legacy
16:27:37 <morganava> i can grab one if someone else will grab the other
16:27:38 <PieroV> Yes
16:27:42 <PieroV> Rebases have been made
16:27:46 <PieroV> Backports not yet
16:28:00 <PieroV> (I'd be up for not backporting to 13.5.x at this point :P)
16:28:07 <PieroV> If Mozilla didn't do it
16:28:16 <anarcat> (ack tx)
16:29:03 <morganava> ma1: are backports in your queue for today?
16:29:11 <ma1> yep
16:29:17 <dan_b> did we reach out and ask for stats on the 13.5 download info?
16:29:26 <ma1> and I can release and sign afterwards (I did a rebase with a fix)
16:30:08 <morganava> dan_b: no that fell down my todo queue
16:30:16 <morganava> in theory we should be able ot just look in grafana
16:30:18 <PieroV> I did open an issue at a certain point
16:30:42 <dan_b> if one knew that existed or where it was?
16:31:07 <dan_b> does anyone here know that?
16:31:12 <PieroV> I opened an issue saying we would like to do that
16:31:17 <PieroV> I don't think it's possible right now
16:31:24 <PieroV> I think we have only per-platform data
16:31:36 <morganava> ma1: ack, i'll optimistically release prep starting with 13.5 after this meeting and if someone else can grab 14.5 in parallel great otherwise i can get it too
16:32:01 <morganava> PieroV: no we definitely can based on mar download stats
16:32:49 <PieroV> morganava: I know, but I think we don't store this data
16:32:49 * dan_b is very curious
16:32:54 <dan_b> where is grafana?
16:33:05 <PieroV> I think we parse it and then we add to grafana only the per-platform data
16:33:28 <morganava> OH
16:33:40 <morganava> i thougoht we just had all those stats lying around in a db somewhere
16:33:45 <morganava> (we being metrics)
16:33:57 <morganava> dan_b: I can dig it up and send to you after this meeting
16:34:06 <dan_b> woo
16:34:40 <dan_b> sounds like we'd maybe want to be tracking version info too?
16:34:55 <dan_b> and platform is just windows? no granularity of 7, 8, 10 or 11?
16:35:16 <morganava> well we don't have any way of detecting such a thing
16:35:21 <PieroV> Yes, no minor OS version
16:35:23 <morganava> becuase you know, fingerprinting :p
16:35:24 <PieroV> We don't want that data
16:35:29 <dan_b> cus my next questions are to we want to discuss changing tracking, like at least add version?
16:36:01 <morganava> i would think no
16:36:02 <dan_b> makes sense, it's just off a single user-agent fetch call right?
16:36:08 <boklm> I think we only have the string in the update URL
16:36:14 <dan_b> no tor-browser version, not os version
16:36:20 <morganava> oooh right
16:36:31 <PieroV> dan_b: we have the browser version
16:36:31 <boklm> we have the current tor-browser version the user is running
16:36:45 <dan_b> i only care about that as a proxy for 13.5 users 😛
16:36:48 <PieroV> browser version + OS
16:36:51 <morganava> that we do necessarily get since the client needs to know whcih mar to download (ie which incremental vs full)
16:37:12 <dan_b> yes. so do we want to store that if we arent now
16:37:28 <PieroV> Yes. I opened an issue about that back months ago
16:37:35 <PieroV> Maybe 1 year ago or more
16:37:36 <morganava> hmm
16:37:37 <PieroV> I just can't find it
16:37:43 <morganava> PieroV do you have link?
16:37:45 <dan_b> cool. link? 👀
16:37:54 <morganava> my grafana credentials don't work anymore
16:38:41 <dan_b> haha who would one talk to about getting them or renewing them?
16:38:58 <dan_b> and where does grafana live? what team owns all this?
16:40:15 <PieroV> dan_b: you can probably guess it ;)
16:40:15 <morganava> metrics team does
16:40:20 <morganava> but i can go dig it up offline
16:40:22 <PieroV> httpS://grafana.torproject.org
16:40:52 <PieroV> Credentials are shared afaik, but I don't know the new ones
16:41:50 <dan_b> sounds like no one does, so yeah, morganava to fetch and distribute
16:41:51 <dan_b> thanks
16:42:08 <dan_b> but then i guess step two is push to store tb version
16:43:10 <morganava> ok folks
16:43:12 <morganava> have a good week
16:43:20 <morganava> and see you on the internet o/
16:43:28 <ma1> grazie, ciao o/
16:43:33 <morganava> #endmeeting