18:00:21 <GeKo> #startmeeting tor-browser
18:00:21 <MeetBot> Meeting started Mon Oct 10 18:00:21 2016 UTC.  The chair is GeKo. Information about MeetBot at http://wiki.debian.org/MeetBot.
18:00:21 <MeetBot> Useful Commands: #action #agreed #help #info #idea #link #topic.
18:00:27 <GeKo> hi all!
18:00:31 <mcs> hi
18:00:37 <boklm> hi
18:00:42 <GeKo> another week with a tor browser meeting
18:00:47 <GeKo> let's do some status updates
18:00:51 <GeKo> who wants to go first?
18:00:57 <femme> hi
18:01:01 <GeKo> o/
18:01:55 * mcs can go first
18:02:02 <mcs> Last week, Kathy and I tested the patch for #18753 and revised our patches for #20111.
18:02:10 <mcs> Along the way we found #20304 and created a patch for it.
18:02:17 <mcs> We also helped debug #20250, we helped with bug triage, and we completed a few code reviews.
18:02:26 <mcs> During the latter half of the week we made a lot of progress on #20185, which we plan to finish this week and then move on to MacOS sandboxing (#20121).
18:02:31 <mcs> That’s all for us.
18:03:27 <arthuredelstein> hi all :)
18:04:32 <GeKo> okay, here is what i did:
18:04:50 <GeKo> i finished the usual begin-of-the-month admin stuff, triaged tickets
18:05:02 <GeKo> and spent almost my whole time in the gitian engine room
18:05:38 <GeKo> i wrote patches for #20184, #20302 and am almost done investigating #20301
18:06:04 <GeKo> i started to look at yawning's sandbox code as well
18:06:34 <GeKo> i tried to get our nightly builds going again but that needs still some debugging it seems :(
18:07:05 <GeKo> this week i plan to conitue that and and try to work with the linux sandbox code
18:07:17 <GeKo> +  there are a bunch of code reviews waiting it seems
18:07:38 <GeKo> (thanks to everyone looking at some patches already)
18:07:46 <GeKo> that's it for me
18:09:37 <GeKo> who is next?
18:09:57 * boklm can go next
18:10:04 <boklm> This past week I made patches for #20183, #20182 and #20217. I Looked at why p7zip doesn't set permissions on #20210.
18:10:30 <boklm> This week I'm planning to continue on #20210, and get a first version of #19067.
18:10:37 <boklm> That's it for me
18:11:37 * arthuredelstein can go
18:11:46 <arthuredelstein> Last week I posted patches for #20244 and #18093 (the latter needs the former).
18:11:51 <arthuredelstein> I also opened #20317 and posted https://bugzil.la/1253006 (aka #6458) and https://bugzil.la/1308607 (aka #20317).
18:11:58 <arthuredelstein> I reviewed patches for #20111, #20184, https://bugzil.la/1264571, https://bugzil.la/1301523, https://bugzil.la/1264562
18:12:07 <arthuredelstein> And I started working again on #19459 but haven't tracked down the problem in Windows yet.
18:12:20 <arthuredelstein> This week I hope to finish #19459, #16622, and help with whatever SponsorU/sandboxing stuff is needed.
18:12:31 <arthuredelstein> That is all for me.
18:14:47 <GeKo> do we have anyone else for a status update?
18:16:05 <arlolra> i can update tm stuff
18:16:17 <arlolra> unless this is the wrong time / place
18:16:25 <GeKo> go ahead
18:16:54 <arlolra> we're switching to building on top of all the tor browser patches in https://github.com/TheTorProject/tor-messenger-build/pull/9
18:17:05 <arlolra> after naively trying it
18:17:08 <arlolra> i needed to revert
18:17:09 <arlolra> https://gitweb.torproject.org/tor-browser.git/commit/?h=tor-browser-45.4.0esr-6.0-1&id=5995aea1797320295fff1247bf4baac9ec511876
18:17:15 <arlolra> and
18:17:15 <arlolra> https://gitweb.torproject.org/tor-browser.git/commit/?h=tor-browser-45.4.0esr-6.0-1&id=9064546a5e54d94054d944dd910a5db3a67032dc
18:17:27 <arlolra> the first one looks like i just need to set a pref instead of reverting
18:17:41 <arlolra> the second one i haven't looked at yet
18:17:56 <arlolra> that's it for now
18:17:57 <arlolra> thanks
18:18:35 <arthuredelstein> arlolra: What problems were you running into for those two patches?
18:18:51 <arlolra> oh, and is there an open bug for the .dmg background not showing on Sierra?
18:19:04 <GeKo> no
18:19:16 <GeKo> care to file one?
18:19:18 <arlolra> ok, i'll open one
18:19:19 <arlolra> sure
18:19:32 <GeKo> i assume that affects tor browser and tor messenger alike?
18:19:37 <arlolra> yes
18:19:54 <GeKo> wow, that update broke quite some things it seems :/
18:19:55 <arlolra> arthuredelstein: the first one, the profile wasn't loading
18:20:23 <arlolra> arthuredelstein: the second one, startls in xmpp wasn't workgin
18:20:36 <arlolra> i didn't investigate too much though
18:21:21 <arthuredelstein> arlolra: I'm wondering if those failures indicate some intrinsic problems with those patches.
18:21:50 <arlolra> ah ... well, the first one, i think it's because we're using the certdb
18:22:01 <GeKo> makes sense
18:22:47 <arthuredelstein> So the first patch shouldn't have an error then, though, should it?
18:23:05 <arlolra> i just need to set nocert = false
18:23:19 <arlolra> security.nocertdb: false
18:23:25 <arlolra> i was going to try that now
18:23:32 <arlolra> and hopefully that fixes it
18:23:47 <arlolra> the second patch seems more problematic
18:24:12 <arlolra> but i can report back when i have more details
18:24:29 <GeKo> fwiw: bobnomnom reports #19910
18:24:31 <arthuredelstein> thanks!
18:24:53 <arlolra> interesting
18:24:55 <GeKo> and they might be right and your issue could be a point for it
18:25:28 <GeKo> i have seen at least one scary crash where this code was in the backtrace but it is hard to reproduce
18:26:10 <GeKo> might be coincidence or not, dunno yet
18:26:11 <arlolra> i didn't get any crashes, just xmpp stream closing
18:27:14 <GeKo> fwiw: i saw you asking for a git branch for comm-central in #build
18:27:25 <GeKo> i guess a better channel could be #releng
18:27:50 <arlolra> oh, the instantbird people directed me there
18:27:52 <arlolra> thanks
18:28:01 <GeKo> i could look up the bug where they added esrXX to git for us and you could ping the respective folks
18:28:04 <arlolra> i just did hg-fast-export.sh locally
18:28:31 <arlolra> i might create a tor-messenger repo on git infrastructure and keep that sync'd myself
18:28:44 <arlolra> GeKo: that would be great, thanks
18:30:46 <GeKo> okay, do we have something for discussion?
18:31:37 <GeKo> i tried hard to understand why poor stable osx users are having issues with the control socket and saw mcs/brade did as well
18:31:53 <mcs> Yes, that is a puzzl.er.
18:31:57 <GeKo> but i guess there is not much to do more on our side until we get more data
18:31:57 <mcs> puzzler
18:32:14 <GeKo> i.e. profiles at least
18:32:27 <mcs> It seems like people are somehow running the alpha Tor Launcher but it is unclear how that happens.
18:33:00 <GeKo> yes, especially as they seems to have the stable stuff otherwise (e.g. tor 0.2.8.7)
18:33:03 <GeKo> *seem
18:34:10 <mcs> Well, the extensions become “sticky” once they open the alpha (if the browser profile is shared between alpha and stable) but tor will be whatever tor is part of the app they open.
18:34:28 <GeKo> that's true
18:34:36 <mcs> Still, at least a couple of user say they never opened an alpha, right?
18:34:41 <mcs> users
18:34:42 <GeKo> yes
18:35:01 <mcs> OK. I think we need more data unless we see this ourselves.
18:35:05 <arthuredelstein> I guess ideally we would keep the browser profiles separate for alpha and stable, right?
18:35:28 <arthuredelstein> Prevent the stable browser from opening a profile from alpha and vice versa.
18:35:45 <arthuredelstein> I don't know how one would do this.
18:36:46 <mcs> We could do that by using a different TorBrowser-Data directory for alpha… it might be worth doing but might cause some confusion too.
18:36:49 <GeKo> maybe, i am not sure about that
18:36:53 <GeKo> yes
18:38:31 <arlolra> cough https://trac.torproject.org/projects/tor/ticket/20309
18:39:24 <GeKo> sort of :)
18:39:37 <GeKo> arlolra: https://bugzilla.mozilla.org/show_bug.cgi?id=1250618
18:39:51 <GeKo> hwine and bhearsum are probably the folks to bug
18:39:57 <arlolra> ty
18:40:36 <GeKo> okay, anything else?
18:41:22 <GeKo> thanks all then *baf*
18:41:26 <GeKo> #endmeeting