[02:02] chrisccoulson: still around? [04:23] ddecator: thunderbird 3.1 looks like it's building === micahg changed the topic of #ubuntu-mozillateam to: Welcome to the Ubuntu Mozilla Team: | ML: http://is.gd/83fnr | Firefox 3.6.6 in Hardy/Lucid / 3.6.7 in Maverick | Help test Firefox 3.6 and xulrunner 1.9.2 migrations in Jaunty/Karmic -- Caution is advised -- http://is.gd/dhTXP | Thunderbird 3.1 Coming Soon to Maverick/PPA | Help test Mozilla prerelease updates http://is.gd/dsudW | Next Meeting: TBD [06:59] micahg: huh what? team? [07:00] ddecator: PPA bug triage [07:00] micahg: ah, alright, sounds good [07:50] micahg: TB 3.1 failed due to missing files [07:50] ddecator: yees [07:50] right, you probably got an email about it, haha [07:50] I didn't upload it, that's an old build in the dailies [07:51] ah [07:51] ddecator: but that is the current pproblem as well [07:51] darn.. [07:52] micahg: i didn't get an update for XR 2.0/FF 4.0 tonight and it says they'll build in 21 hours..was there just not an update tonight? FF 4.0 is broke for me atm due to w/e reason y-ofel talked to you about before [07:52] ddecator: I'm getting them rescored, but the build is broke :( [07:53] micahg: :( [07:53] micahg: so anything for me to work on? [07:54] ddecator: oh the TB31 build is failing in teh exact same place :( [07:54] micahg: darn, i was hoping to use it soon... [07:55] ddecator: I'll try this weekend, I have to get something done first though [07:55] micahg: sure thing, just let me know if you want me to look into anything [07:55] ddecator: no, I think it's good ATM [07:56] chrisccoulson: if you get a chance, can you see why lp:firefox is installing the firefox-4.0 symlink in the firefox-3.7 package [07:57] ddecator: I would ask you, but I think it needs advanced package fu [07:57] micahg: no problem [08:07] ddecator: I have a task :), can you find the firefox PPA bugs and add a task for this new project and make the Ubuntu task invalid. Only move them if the version is later than what's in the archive [08:07] micahg: sure. what's the project? [08:09] micahg: nvm, just got the email [08:11] ddecator: here's the project: https://edge.launchpad.net/ubuntu-mozilla-ppa-bugs [08:12] ddecator: this weekend I'll fix the PPA apport hooks to point there instead of to Ubuntu [08:12] micahg: sounds good [08:13] * micahg probably should have made it more generic for that answers would fit as well, but oh well [08:13] now I must sleep for a short while [08:14] micahg: night === yofel_ is now known as yofel [10:55] hello guys.. why do i get this error when trying to installing firefox 4.0 ? actually, i already installed it, codename shiteroko is im not mistaken.. but after doing 'sudo apt-get upgrade', i cant run it. try to reinstall and got the error. http://pastebin.com/bU4xLMJb [10:57] the error when i try to running it is something like "Cannot execute child process. Files not found" [11:14] hotfloppy: that's known and being worked on [11:19] yofel: oic.. thanks :) [15:09] hello guys.. i just upgraded my system (apt-get upgrade) and now firefox change to Namoroka. before this, its mozilla firefox. and also now i cant open the profile manager.. is something happening on the PPA ? coz my thunderbird title also change to shredder (if im not mistaken the name).. [15:27] chrisccoulson: ah, for dh_link is that how it works? [15:29] hotfloppy: yofel: thanks to chrisccoulson, the install issue should be fixed in tonight's upload [15:30] :) [15:30] okay.. how bout my current issue ? is that already in consideration ? [15:30] micahg - yeah, that's how it works ;) [15:30] it creates the link in the first listed binary package [15:30] chrisccoulson: I'll have to remember that, otherwise, I need to use -p? [15:33] chrisccoulson: oh, BTW, there's a build2 for FF3..6.7 [15:34] micahg - ok, thanks. i'll look at that later [15:34] for now, my computer is broken ;) [15:34] chrisccoulson: k, they seem to be on target for release on Tuesday === hotfloppy1 is now known as hotfloppy [16:08] chrisccoulson: how should we treat ums bugs? Since the same version is in Maverick, is that considered 'in archive' or since the build is from a PPA we should use the PPA bugs project? [16:09] micahg - we should treat them as ubuntu bugs, as the u-m-s PPA is about as official as it gets for PPA's (and those packages are destined for the archive anyway) [16:09] chrisccoulson: k, thanks [16:10] chrisccoulson: i'm not sure how I can do that in the apport hook though ;) [16:12] yeah, that's a pain. what logic are you using? [16:12] chrisccoulson: apport has a 3rd party option that I'll use to redirect to the new ppa bugs project [17:43] micahg - for the FF4.0 branding, should we use the name of the current branding "Mozilla Developer Preview" in the desktop file rather than Shiretoko? [17:44] i was just thinking about it, because i need to change the StartupWMClass to that anyway so that it works with docky [17:58] chrisccoulson: I think the codename is still minefield at the moment, and I made that change yesterday [17:59] chrisccoulson: but last night's builds won't work because of the issue you fixed this morning [17:59] chrisccoulson: so if the minefield desktop file has StartupWMClass, then it should be fine tomorrow [18:00] * micahg is off...commuting :) [18:00] micahg - which branding are we building it with now then? [18:00] chrisccoulson: unofficial branding/minefield [18:00] micahg - ah, the unofficial branding isn't minefield though is it? [18:01] chrisccoulson: it is for trunk [18:01] chrisccoulson: when they branch, they'll probably give it a codename [18:02] micahg - ah, ok. so, we should be building it with the nightly branding, which is minefield [18:03] that's how i understand it now anyway ;) [18:03] chrisccoulson: right, which is the change I made yesterday as to which .desktop file to use, the build was already using that, just the .desktop was wrong [18:04] k, really off now [19:14] jdstrand, is there anything else i need to do to get the seamonkey security update in to lucid? i'm still not completely clear how we handle these in universe [19:15] chrisccoulson: you tell me when it is ready, and I copy it [19:15] iirc there was some more testing that was desired (did micahg say that)? [19:16] chrisccoulson: does it require the newer nss? [19:18] jdstrand, it's ready to go for lucid [19:18] we haven't had much testing for < karmic, so we should leave those for now (and they're all still on the 1.x version anyway) [19:18] k [19:22] jdstrand - you're at the spring next week aren't you? [19:22] s/spring/sprint [19:22] heh [19:23] chrisccoulson: yep [19:23] chrisccoulson: seamonkey is now published for lucid [19:23] jdstrand, excellent, thanks [19:24] jdstrand, there's another round of updates coming on tuesday [19:24] sure! thanks for the update :) [19:24] chrisccoulson: for seamonkey? [19:24] jdstrand, for seamonkey, firefox and tb [19:24] :\ [19:24] https://wiki.mozilla.org/Releases/ [19:25] chrisccoulson: will you have releases prepared for firefox then? [19:25] jdstrand, we have build1 in the PPA and in maverick already [19:25] i'm just about to do build2 [19:25] and thunderbird is done already too [19:26] i just need to reupload SM2.0.6, then that's done too [19:26] chrisccoulson: k. all signs point to 'yes' for openjdk. we just need euca testing [19:26] chrisccoulson: so hopefully jaunty and karmic can get the new firefox and the transition will be complete [19:27] jdstrand, cool, that's good :) [19:27] i will be glad for it to all be finished ;) [19:28] yeah, totally [19:28] jdstrand, i may as well do the 3.6.7 update for karmic and jaunty shouldn't i? (i think it's unlikely we'll push the updates out before 3.6.7 is released now) [19:29] chrisccoulson: yes [19:41] jdstrand, have you been using the FF4 beta at all? [19:41] chrisccoulson: no, not at all. hat is the eta for that? [19:42] jdstrand, probably november. it's quite nice though [19:42] but it's even nicer on windows 7! [19:42] * chrisccoulson runs [19:42] heh [19:44] i don't know if we will get a client-side-decorated version on linux, with the menu button in the titlebar [19:44] i hope we do :) [19:48] windows 7? never heard of it [19:48] lol [19:48] :) [19:49] i installed it in virtualbox just so i could try FF4.0 on it ;) [19:50] chrisccoulson: so did they achieve the speed improvements? am I remembering that they are doing more of a chromium thing with a confined rendering process? [19:53] jdstrand, i don't think there is anything like that yet, but it's meant to be getting a new (faster) JS engine [20:00] micahg - have you had any thoughts on how we should split xulrunner? [20:19] chrisccoulson: well, if we can split the components that don't need X, that would be good, I don't know offhand what they are though [20:20] micahg - that's the difficult part, i'm not sure either [20:20] chrisccoulson: it might just be mozjs, idk [20:20] chrisccoulson: i can try to look this weekend [20:21] micahg, i think that's the only thing too, so we would end up with something like xulrunner-1.9.2-mozjs for the couchdb guys and xulrunner-1.9.2 for everyone else [20:26] chrisccoulson: k, and the with the xul version in there, it will make it easier to recognize the need for porting [20:27] yeah, hopefully [20:27] well, if we think that's a sensible split, then i can do that maybe tomorrow [20:34] chrisccoulson: maybe we should call it xulrunner-1.9.2-nox, just so people don't think we have mozjs like Debian [20:51] chrisccoulson: package mozjs completely seperately if you plan to do that [20:51] its a subtree that can be build indepentently [20:53] also couchdb folks need to go for javascriptcore. in the end i think shipping mozjs will send the wrong signal [20:54] or produce a couchdb with in-source js shipped there and get security team blessing to do that [20:55] and throw out everything else that uses mozjs in archive [20:56] asac: everything else we can use a wrapper around I think (except maybe mongodb), the issue with couchdb is the space on the CD, it's the only thing that needs xulrunner ATM [20:57] * micahg needs to make gnome-shell use a wrapper this weekend [21:27] micahg - heh, i just had a read of lkcl's rant on debian-devel [21:28] chrisccoulson_: on debian-devel, you can take a swipe at someone :) === chrisccoulson__ is now known as chrisccoulson [21:38] micahg - can you get any mozillateam branches atm? [21:38] i get this error when i try and get a mozillateam branch: [21:38] bzr: ERROR: Not a branch: "bzr+ssh://bazaar.launchpad.net/~mozillateam/firefox/firefox-3.7.head/" [21:38] but i can access other branches :-/ [21:38] chrisccoulson: that doesn't exist anymore, it's lp:firefox now [21:38] or firefox-4.0.head [21:39] micahg - yeah, but i'm not trying to access that branch [21:39] i get that error when trying to access any mozillateam branch [21:39] (the one i want is lp:~mozillateam/firefox/firefox-3.6.lucid) [21:40] chrisccoulson that shouldn't be stacked on 3.7 :( [21:41] chrisccoulson: ugh, that's in branch format 7 [21:41] chrisccoulson: you'll have to branch it anew [21:42] hmmm, what happened there then? [21:42] chrisccoulson: it's stacked wrong for some reason [22:25] micahg - branch fixed now :) [22:25] * chrisccoulson wipes brow [22:44] sorry gueys ...i killed lp:firefox yesterday [22:44] when i tried to fix the stacking [22:44] you have to re-set that up [22:46] asac - no worries. i had a branch stacked against lp:~mozillateam/firefox/firefox-3.7.head for some reason [22:47] so, i just recreated that branch and then unstacked mine ;) [22:47] chrisccoulson: is it still in right format? [22:47] ;) [22:47] chrisccoulson: actually why i found out about the brokenness was that i wanted to commit something to ffox 3.6 and 4.0 [22:47] asac - it seems that it was already in the wrong format (firefox-3.6.lucid) [22:47] i dropped the ball ... let me check :) ... one sec [22:48] chrisccoulson: :((((( [22:48] how much i hate launchpad for that ;) [22:48] yeah, me too ;) [22:49] chrisccoulson: if i commit something to 3.6 branch can you remember to pull that to the 4.0 branch? [22:49] ;) [22:49] asac - yeah, can do. i've got quite a few things to push to the 4.0 branch now [22:50] ok ... i will let you know. its nifty stuff ;) [22:50] asac - is now a good time to clean up the transitional packages (seeing as we're LTS+1), or do we still need to keep them for the daily builds? [22:52] chrisccoulson: yes. drop them ... for daily builds we can make a fake transition as we did in the past [22:52] e.g. create a source package with just transitional packages and put that in the ppa [22:52] asac - ok, cool, i'll do that before the next update then [22:53] chrisccoulson: check out what i committed ;) [22:53] not sure if i forgot to install something after enabling crashreporting [22:53] Committed revision 622. (on 3.6 brach) [22:54] asac - nice. what do we do with the symbols? [22:54] i will setup an automatic upload to mozilla for now [22:55] and then we should disable our crash reporting [22:55] at best that push would ahppen where we currently sort out the dbgsym on the build farm [22:55] until then we can manually upload [22:56] i will send an email to ted and CC you and if you want you can do the uploads of the .zip in there [22:56] ... we can do that for all biulds: archive + daily [22:56] they want everything [22:56] asac - ok, cool. so, the symbols package is not the sort of thing that users would normally install is it? [22:57] (ie, they don't need it for submitting crash reports?) [22:57] right. no user wants that [22:57] they only see the crashreporter and get asked if they want to upload that (which then goes to mozilla) [22:57] cool [22:57] and we can use their crash db [22:57] i'll talk to the LP guys tomorrow or next week, and see how we can make it more automated [22:57] also mozilla will consider us a first class citizen so if they see top crashes from our users (like on dailies) they now can hold back release etc [22:57] yeah cool. [22:57] i will connect you with ted [22:58] thanks [22:58] if you dont see a mail by tomorrow lunch poke me [22:58] chrisccoulson: it might be that for now we have to switch from dwarf to "stabs++" debug symbols. but most likely we just need to pull the trunk script for stripping the symbols to the fffox 3.6 branch to get it right [22:59] breakpad now understands dwarf. but lets talk that over [22:59] with a beer next week ;) [23:00] cool :) [23:00] asac - for disabling apport, we just need to blacklist firefox don't we? [23:01] is that blacklisting on client side? [23:01] i dont want the "submit crash" etc. to pop up at all [23:01] at best not even a core dump to be created [23:01] but if thats still done on development releases its fine [23:01] asac - yeah, it's possible to blacklist applications on the client side, so that apport will ignore crashes [23:01] our new crashreporter works without core [23:01] right. then thats the way forward [23:01] or maybe some apport hoook hint? [23:01] there's a text file for doing that. in fact, we might even be able to ship the blacklist file with firefox [23:01] (we still want to keep apport hook for normal bug reports) [23:02] yeah, i think the blacklisting is only for the crash reporting [23:02] right. i would hope we can ship it in firefox [23:02] but lets first get the crash reports working ;) [23:03] asac - /etc/apport/blacklist.d :) [23:03] so, we can easily do that, which is good [23:04] guys, you should seriously think about taking over the dailies. i'm considering shutting the bot down completely for several reasons [23:05] yay ... finally after years of pain we might be able to get rid of crash bugs [23:05] fta - i can probably do that [23:05] fta: other reasons than a) often ftbfs ... and b) not enough ppa builders? [23:05] i should be able to spend more time on these things now i'm not doing backporting for other releases ;) [23:06] asac, a/ + b/ + c/ canonical is advocating its lp dailies so there's no need for me to continue spending my time on this + d/ [23:06] i doubt we can produce lp dailies for mozilla etc. actually for many stuff it wont work in the future [23:06] asac, d/ canonical is also advertising the use of upstream binaries, like chrome vs chromium, ff instead of our builds, etc [23:07] all in one, that's too much for my taste [23:07] when did canonical do that? [23:07] read mpt and jorge's blogs [23:07] i am sure we are not really advocating use of mozilla binaries [23:07] -advertising+advocating [23:08] i will talk to jorge face to face about this next week [23:08] i doubt that d) is true in general. [23:08] and e/ i'm getting all the user complains. i forward some to micahg, but i'm just a middle man now [23:08] fta: right. assuming it goes to chrisccoulson i doubt it matters. but you could use a different email etc. for changelogs [23:09] or/and a separate bot acccount [23:10] asac, it doesn't make sense for me to continue doing that, at least for umd [23:10] agreed ... just work on a seemless transition with chrisccoulson. thanks for letting us know [23:11] if you could keep nmt enabled that would be fantastic [23:11] until we find someone else willing to run that (might happen soon) [23:11] c/ is the new buzz jorge and daniel are spreading, so use that instead [23:12] well. there is always technology targetting the broad masses ... and stuff that really works ;) [23:12] the ftabot is the latter ;) [23:13] s/targetting/trying to target/ [23:14] chrisccoulson: has ffox 4.0 webgl enabled on linux? [23:14] i know that upstream doesnt think its ready. just wonder if we could make a special build available [23:14] asac - if it needs explicitly enabling, then it's not been enabled yet [23:14] but i don't mind doing that if you think it's worthwhile [23:14] heh. i figured that much ;) [23:15] i had it working a while ago, not sure it's still there [23:15] chrisccoulson: we cant change defaults for official builds for release. so i think it would need to be a special build. i guess i can also steal that for now ;) [23:15] i'll have a look at that in a bit, and merge some changes from FF3.6 as well [23:15] fta: really? that would be amazing. what graphics chipset are you on? [23:15] (i'm just doing the 3.6.7+build2 updates atm) [23:15] nvidia [23:16] with nvidia driver? do they support glsl shader language? nice [23:18] hm, i dropped the 3.7 transitional packages, now 4.0 is broken [23:19] what do you mean? [23:19] in the branch? or locally? [23:19] lol, everyting is in the transitional deb [23:19] heh [23:19] that would explain it ;) [23:20] root@ix:~ # dpkg -c /var/cache/apt/archives/firefox-3.7_4.0~b2~hg20100712r47341+nobinonly-0ubuntu1~umd1_i386.deb | grep usr/bin [23:20] drwxr-xr-x root/root 0 2010-07-13 11:29 ./usr/bin/ [23:20] lrwxrwxrwx root/root 0 2010-07-13 11:29 ./usr/bin/firefox-4.0 -> ../lib/firefox-4.0b2pre/firefox.sh [23:20] root@ix:~ # dpkg -c /var/cache/apt/archives/firefox-4.0_4.0~b2~hg20100712r47341+nobinonly-0ubuntu1~umd1_i386.deb | grep usr/bin [23:20] drwxr-xr-x root/root 0 2010-07-13 11:29 ./usr/bin/ [23:20] lrwxrwxrwx root/root 0 2010-07-13 11:29 ./usr/bin/abrowser-4.0 -> firefox-4.0 [23:20] micahg: ^^ [23:20] FAIL [23:21] guess just a few bzr move [23:21] didn't he already fix that? [23:21] * asac updates his whole system [23:21] could be that its not yet built or fta hasnt updated [23:22] "Need to get 679MB of archives." [23:22] thats ok [23:22] 0 upgraded, 0 newly installed, 0 to remove and 0 not upgraded. [23:22] so probably the build failed since 12th? [23:23] lp sucks, as usual [23:23] Show details firefox-3.7 - 3.7~a6~hg20100629r46385+nobinonly-0ubuntu1~umd1~lucid [23:23] python rebuild [23:23] no, the ppa builders are so behind that the builds were superseded before being built [23:23] Show details firefox-4.0 - 4.0~b2~hg20100715r47651+nobinonly-0ubuntu1~umd1 [23:23] yeah. its waiting for cycles :) [23:24] you could go for lpia ... lol [23:24] i complained this morning, and said it's unusable [23:24] i also said i'll kill the bot [23:24] i think htere is a problem [23:24] i didn't, but i should [23:24] with the scheduler again [23:24] well, they said that the builders just come and go (in #launchpad) - wonder if the found out what's wrong by now... [23:24] lots of builders are not utilized [23:24] * asac goes hunting in dark territory [23:27] lets see seems hour whole soyuz team isnt even there ;) [23:27] our [23:27] confirmed, webgl works here with ff4.0 [23:27] and chromium [23:27] cool [23:27] http://khronos.org/webgl/wiki/Demo_Repository [23:28] wonder if mozilla already published there simulater [23:29] poof [23:29] _usr_lib_firefox-4.0b2pre_firefox-4.0.1000.crash [23:29] heh [23:30] crashes on http://html5test.com/ [23:31] shame [23:32] corrupted stack, useless === fta_ is now known as fta [23:38] fta - i pushed a fix this morning for the symlinks being shipped in the transitional package [23:39] yep, it's not fixed in the ppa package, but using the bzr branch works fine