[00:00] asac: is firefox-qt still alive? :P [00:00] chrisccoulson: idr [00:00] sebner: install kmozillahelper :) [00:00] micahg - ok, i will carry on with that again tomorrow [00:00] but i'm starting to officially hate gjs :) [00:00] sebner: i think most understand that native qt doesnt give much over a good qt theme engine ;) ... but adds loads of maintenance effort [00:00] chrisccoulson: k, let me know if I need to look at it this weekend [00:01] micahg - ok, now worries [00:01] thanks [00:01] s/now/no [00:01] heh [00:02] asac: btw, have I ever told you that you work far too much? :P (assuming you are currently in the same timezone as I am) [00:02] yes [00:02] ;) [00:02] :D [00:42] weird... 3.7 build still seems broken. [00:42] DanaG: xul didn't bui;d [00:42] Dang. [00:42] DanaG: hopefully new 3.6 builds in about 12 hourss [00:43] idk if I can fix xul193 tonight [00:43] hmm, perhaps I should try wrapping mozilla-runtime with an sh script that sets LD_LIBRARY_PATH. [00:43] DanaG: already exists [00:44] Well, somehow mozilla-runtime isn't getting it, or something. [00:49] yay, making my own wrapper worked. [00:49] DanaG: ?? [00:49] I dpkg-diverted mozilla-runtime out of the way, and made a #!/bin/sh script with this: [00:49] export LD_LIBRARY_PATH=/usr/lib/xulrunner-1.9.3a4pre/:$LD_LIBRARY_PATH [00:49] env > /dev/stderr [00:49] exec /usr/lib/xulrunner-1.9.3a4pre/mozilla-runtime.distrib $@ [00:50] env for debug, of course. [00:51] interesting... it seems mozilla-runtime gets a very minimal environment. [00:52] It'd be nice if the "plugin has crashed" showed the logo of the plugin or of the creator. [00:52] Such as Flash / Adobe, or Silverlight / Microsoft. [00:53] http://blog.mozilla.com/dolske/2010/02/10/crashed-plugin-ui/ [00:56] i'd prefer a full-on kernel panic than have to worry about EVERY SINGLE PLUGIN [00:57] generic image = KISS [00:59] asac, fta. can we get umd build that is newer than changeset 68737c373aba? bugzilla 558690 was an ubuntu-only bug and it just got fixed. [01:00] *umd 3.7 build [01:00] LLStarks: I'll get to it when I can :) [01:00] LLStarks: first priority is 3.6 so we can get ipc testing [01:00] good call. i'm quite fond of oopp. [01:01] btw, how can i make my builds have proper hinting? [01:01] cairo is sucking hard. [01:01] LLStarks: if we knew that, I don't think we'd have an issue :) [01:02] LLStarks: there are some workarounds in the old bug for hinting [01:02] what is the heart of the problem that makes hinting break with official mozilla tarballs? [01:02] and how does umd circumvent that issue? [01:02] DanaG: I'll look into the wrapper and make sure we're setting everything we should [01:02] LLStarks: we don't know and how does it [01:02] Now my only issue is "This report does not apply to a packaged program". [01:03] ... since I overrode mozilla-runtime. [01:03] well, fta must know something that i don't. [01:03] :) [01:03] LLStarks: I'm mising something [01:03] DanaG: 'tis ok, I have enough info now [01:04] DanaG: you can attach wrapper and comments to bug 513887 [01:04] Launchpad bug 513887 in firefox "Flash, Java, etc. does not work with out of process plugins and causes Firefox to become unresponsive" [High,Triaged] https://launchpad.net/bugs/513887 [01:04] micahg, umd builds have proper hinting. i also think 3.6 in main does too. [01:05] LLStarks: ah, hinting is fixed, smoothing is not [01:06] LLStarks: for 3.6 we're using in source libs, for 3.7 we are not [01:07] LLStarks: I'll work on getting Firefox 3.7 to work like 3.6 after UDS [01:07] source libs? [01:07] LLStarks: libraries that mozilla ships vs system libs [01:08] aside from stray bugs, 3.7 works the same as 3.6 as far i'm concerned. no hinting errors. [01:09] oh, maybe latest cairo fixes it idk [01:09] 3.6 uses system cairo, right? [01:09] LLStarks: no [01:10] LLStarks: for lucid we run dailies again [01:10] but its probably broken ;) [01:10] e.g. fails to build [01:11] h [01:11] *ah [03:05] asac, i'm getting striped pngs with 3.6 main. [03:07] http://img710.imageshack.us/img710/2558/stripe.png [03:07] libpng acting up? [03:14] micahg, ever see anything like this in 3.6 main? [03:14] http://img710.imageshack.us/img710/2558/stripe.png [03:14] certain pngs not rendering properly [03:15] renders normally in chromium [03:15] LLStarks: no, you have the link so I can try? [03:16] http://shinigamifaqs.com/Themes/default/images/theme/main_block.png [03:16] LLStarks: looks fine to me [03:16] LLStarks: what graphics driver? [03:16] intel [03:16] * micahg too [03:16] not all pngs do that [03:17] LLStarks: idk, are you running the latest kernel? [03:18] yeah [03:18] 21 [03:18] LLStarks: also, there have been a couple intel driver updates in the past couple days [03:18] if that was the case, why only certain pngs? [03:18] alpha channel? [03:19] for what it's worth, i am running xorg-edgers and alpha pngs are looking fine [03:19] apng too [03:20] LLStarks: could be there's a bug in xorg-edgers drivers, I'm running hte intel driver from archive lucid [03:20] hmm. [03:20] certain jpegs are also doing this when not zoomed [03:21] http://img191.imageshack.us/img191/9499/lovesgv.jpg [03:21] looks fine when zoomed [03:22] here's how i see it: http://img338.imageshack.us/img338/7241/lovescreenshot.png [03:22] LLStarks: try archive drivers [03:23] will do. [03:23] brb [07:48] is spidermonkey in lucid in any way? [07:58] hey is there an ubuntu ppa for lucid as they removed spidermonkey from there repositories [08:59] hi hi kangoro [08:59] hey BUGabundo_remote [11:52] chrisccoulson: you gave tbird back? [11:56] do you have the build failure log somewhere still? [11:56] i need that to get doko look into it [12:41] the bot works again :) [12:43] is firefox safe to use or is it still broken? [12:44] 3.7 == no browser 3.6 keeps locking up on me and others [12:49] gnomefreak: daily? [12:50] nope 3.6 is still broken [12:50] asac: yes [12:50] was there a daily today or yesterday? [12:50] last build before was in the middle of a the plugin transition [12:50] so we want to check latest today before going crazy [12:50] asac: i got updates a little while ago. sometime in the last 18 hours or so [12:51] asac: version 3.6.5~hg20100414r34076+nobinonly-0ubuntu1~umd2 [12:51] hmm [12:51] yeah [12:51] * asac upgrades too [12:51] to see [12:51] if there are hangups we should check [12:51] otherwise we wiill be hit hard by 3.6.4 [12:52] the last couple of weeks has had this issue [12:52] * gnomefreak tries running 3.7 from term [12:53] 3.7 o0utput from term is gnomefreak@Development:~$ firefox-3.7 [12:53] gnomefreak@Development:~$ ps aux |grep firefox [12:53] 1000 4262 0.0 0.2 1856 580 pts/1 S+ 07:52 0:00 grep firefox [12:53] not sure [12:53] * gnomefreak sick of using chromium it doesnt print correctly [13:13] hey asac, sorry i had to go out for an appointment [13:13] no, i didn't give thunderbird back [13:14] (i'm not allowed to for stuff in main) [13:25] ok firefox froze so i killed the process but why are there 4 processes, seems like a lot for just a browsere [13:25] s/browsere/browser [13:26] ^^^ maybe causes the "slow/freeze" not sure what word to use [13:29] chrisccoulson: yeah [13:29] chrisccoulson: i found out who by now ;) [13:29] chrisccoulson: how is the 3.6 build1 for security ppa project going? [13:42] asac - oh, micahg has already updated the branch for that? [13:43] excellent :) [13:44] asac - ok, i will get that in to the PPA this afternoon [13:44] we just go through the normal release process for that? (ie, tag it 3.6.4+build1+nobinonly-0ubuntu1) [13:48] * gnomefreak getting tired of getting up to change cd player. need to get a battery for remote [13:59] * gnomefreak hates filing wishlist bugs [13:59] file a bug on liferea and i get the following output [13:59] OpenOffice path before fixup is '/usr/lib/openoffice' [13:59] OpenOffice path is '/usr/lib/openoffice' [14:00] and it never closes [15:15] i thought we were adding support to apport for PPA packages? [15:56] well ubuntu-bug fails but apport-collect bug# works [16:06] asac - is OOPP meant to be enabled in 3.6.4? [16:08] chrisccoulson: yes [16:08] chrisccoulson: thats why we need to get build1 up now [16:08] and call for testing [16:08] so we can tell them: no, that needs to stay disabled ... ;) [16:08] based on real feedback [16:08] asac - it's disabled in build1 [16:08] it is? [16:08] ok [16:08] well, disabled as in "dom.ipc.plugins.enabled" is set to false [16:08] then i didnt see that they changed their plans [16:08] but it's a resounding failure when i enable it :( [16:08] it just crashes and hangs [16:08] yeah [16:09] so they probably noticed it ;) [16:09] however, afaik the code landed [16:09] and they dnt know if it causes regressions even with disabled [16:09] iirc [16:09] so lets get it up ;) [16:09] ok, so we leave it disabled for now then, it's practically unusable with it enabled [16:10] chrisccoulson: we shouldnt enable without them [16:10] we ride whatever they do [16:11] but they landed that ode [16:22] asac - so, for the 3.6.4 update, i just tag it for release like i would with any other upload before i put it in the PPA? (sorry, i'm a bit unsure as I didn't upload the last security updates until after the mozilla release) [16:22] i don't want to mess up the version numbers in the ppa ;) [16:23] chrisccoulson: yes. you just use the _BUILD1 tag and use +build1 as the upstream version [16:23] and you document it like that [16:23] chrisccoulson: also you get the USN for that for this upload [16:23] because in case its good we dont need to upload again [16:23] i think there are other BUILD1 uploads a bit further down in the changelog [16:23] (otherwise look on stable branches like hardy) [16:25] asac - cool, that's ok then, i will get that sorted now [16:26] nice [16:26] chrisccoulson: also setup a blog... get on planet. blog about it. [16:27] tell me so i can forward that call for testing on my blog ;) [16:27] e.g. lets get started on getting more folks again on the ppa [16:27] chrisccoulson: http://www.asoftsite.org/s9y/archives/146-new-firefox-securitystability-upgrade-available-for-testing.html [16:27] asac - ok, no problem . the other thing i've been looking at today is bug 543064, but i'm not sure how that's meant to work on tb2 [16:27] Launchpad bug 543064 in thunderbird "ensure that x-www-browser is used if no http handler is found through gnome integration" [Medium,Triaged] https://launchpad.net/bugs/543064 [16:28] i can't find any preference for it, and if i remove the handler specified in gconf, then it fails to open any links just like in tb3 [16:30] i tried following the code in tb3, and i think we'd need a patch if we were to make it use x-www-browser as a fallback. i'm just wondering if i should fix bug 543060 as a priority instead, as that would make most of the issue go away [16:30] Launchpad bug 543060 in thunderbird "thunderbird - gnome integration should work even without -gnome-support" [Wishlist,Triaged] https://launchpad.net/bugs/543060 [16:46] asac or anyone, do you think i should make the chromium lang packs mandatory for everyone? or just keep it like it is [16:46] wrt bug 561624 [16:46] Launchpad bug 561624 in chromium-browser "Create meta-package with dependencyes like in google-chrome" [Undecided,New] https://launchpad.net/bugs/561624 [16:55] fta: no [16:56] fta: langpacks should get integrated in language selector [16:56] talk to arnegoetje [16:56] if we can do that for universe packages (e.g. if we see its installed also pull in langpacks) [16:57] i guess it's too late for lucid [16:57] * micahg thought universe packages ship their own translations [17:01] the thing is that those langpacks *must* be in sync with ch. otherwise, it's all messed up [17:01] good point to note down [17:01] fta: in that case we probably have to ship all langpacks in the chromium package itself [17:02] asac: why not just do a binary depends on the lang package deb? [17:02] asac, "depends" or drop the all-arch package? [17:06] ok, uploading like it is [17:08] asac: do you want to file the bug to drop xul191 from Lucid? [17:08] chrisccoulson, uploaded [17:08] fta - excellent, thanks :) [17:09] chrisccoulson: was my fix good enough for thunderbird? people still complain that they can't start (I guess asac was right and no one reads the NEWS) I added the text to the master bug [17:09] micahg - i've not had any issues with it at all [17:10] chrisccoulson: right, because you followed the instructions I"m sure :) [17:10] asac: chrisccoulson: does someone want to file the drop xul191 from Lucid bug? [17:11] what is still depending on it now? (other than sugar, that we're going to drop) [17:11] it's probably best to add xul191 plus the remaining rdepends to the same bug [17:11] chrisccoulson: only gjs, but we already have a bug to fix that [17:12] yeah, i need to try and fix that [17:19] chrisccoulson: I just got a package I needed for libjdic-java sync'd so I"ll upload the diff for it a little later [17:52] micahg - do you know which of the dailies needs fixing most urgently? [17:54] chrisccoulson: he is off [17:54] chrisccoulson: everything is red ;) [17:54] asac - oh, i didn't notice he'd gone [17:54] let me check [17:54] asac - am i looking in the wrong PPA? [17:54] https://edge.launchpad.net/~ubuntu-mozilla-daily/+archive/ppa/+packages [17:55] i see quite a bit of green too ;) [17:55] so firefox 3.6 on hardy is borken [17:55] then ffox 3.7/1.9.3 [17:55] ok, i had a quick glance at ff3.6 on hardy. i suspect our cairo patch is what breaks that [17:56] sorry i get picked up [17:56] have to run down [17:56] chrisccoulson: could be [17:56] i'll have a look at that then and see what i can do to make it build on hardy [18:50] asac - so, ff3.6.4 is building in the u-m-s PPA now. i'm just testing xulrunner too, and will upload that in a bit [19:16] bdrung: does the eclipse package with the .ini have a binary depends on xul192? [19:17] bdrung: nm, it does seem to [19:38] chrisccoulson: what's going on? [19:38] chrisccoulson: why are you release xul194? [19:38] micahg - xul1.9.2.4? [19:39] chrisccoulson: yes [19:39] micahg - asac wanted to get it in to the u-m-s PPA so we can get people testing it [19:39] chrisccoulson: it's not ready yet [19:39] chrisccoulson: we still have a bug with OOPP [19:39] what isn't ready? [19:39] micahg - OOPP is disabled isn't it? [19:40] (it seems to be disabled here anyway) [19:40] chrisccoulson: no, not entirely [19:40] i tried enabling it ;) [19:40] it didn't work very well [19:40] chrisccoulson: right, and it should be shipping by default in 3.6.4 [19:41] I think that someone might have stumbled on the key [19:41] so there's going to be another upload for final 3.6.4 anyway? [19:41] chrisccoulson: not before release [19:41] you mean lucid release? [19:41] * micahg is confused [19:41] me too :) [19:42] OOPP is shipping upstrea, by default in 3.6.4 most likely [19:42] * micahg does not want 3.6.4 in Lucid release [19:42] micahg - ok, i get that bit. [19:42] don't worry, it's not in lucid release [19:42] it's targetted for the lucid-security pocket, and in the u-m-s PPA [19:42] chrisccoulson: there's info in bug 513887 if you want to look into it now [19:42] Launchpad bug 513887 in firefox "Flash, Java, etc. does not work with out of process plugins and causes Firefox to become unresponsive" [High,Triaged] https://launchpad.net/bugs/513887 [19:43] chrisccoulson: I think it's a packaging issue [19:43] micahg - but you say that OOPP will be enabled for the final release? (which implies that we are going to do another upload for the security release anyway) [19:43] chrisccoulson: I was going to look into it this weekend [19:43] chrisccoulson: it is for flash by default [19:44] it's not in the current release though is it? (i tried youtube and didn't see it run in another process) [19:44] chrisccoulson, micahg: so this next ff release will be the first with the process separation for plugins? [19:44] jdstrand: yes [19:44] jdstrand, yeah, that's right [19:44] cool. I am a fan of that btw [19:44] chrisccoulson: yeah, I saw that also [19:44] but it's disabled at the moment [19:44] chrisccoulson: I need to discuss with upstream, but there will definitely be a build 2 [19:45] chrisccoulson, micahg: we should be sure to test the apparmor profile with that configuration [19:45] micahg - ok, that's fine. i'm happy with people testing build1 right now === dpm is now known as dpm-afk [19:45] jdstrand - i'm wondering if that's not the cause of the current issue ;) [19:45] hopefully it'll all just work, but we might need some finetuning [19:45] it doesn't work at all atm [19:45] chrisccoulson: look in kern.log [19:45] firefox just hangs [19:45] chrisccoulson: I'm not, because security PPA uploads IMHO should be free of known issues [19:45] 1 second [19:45] chrisccoulson: but the profile is disabled by default [19:45] so if that is a fresh install, then it shouldn't b [19:45] e [19:46] jdstrand, oh, i've not actually enabled it anyway [19:46] chrisccoulson: you can be super sure by looking at the output of aa-status [19:47] (run with sudo) [19:47] chrisccoulson: once you have a build you are starting to feel comfortable about, we should test the the profile on [19:47] jdstrand: can you make the apparmor notification system show blocks with the profile in complain mode or is that a bad idea? [19:48] micahg: apparmor-notify doesn't care if it is enforce or not [19:48] while testing in complain mode is valid, enforce mode still needs to be done [19:48] jdstrand: it seems to [19:48] jdstrand, i don't see the FF profile when i run aa-status [19:48] chrisccoulson: k, then it is disabled [19:49] micahg: oh, maybe the string is different... [19:49] jdstrand: should I file a bug? [19:49] micahg: yes please. it may not get fixed in lucid, but it might [19:49] (it should be a teeny change) [19:50] micahg: in the bug, please give dmesg output that should have been caught [19:50] jdstrand: k, I was also thinking, can we get an abstraction for feed readers in apparmor? [19:50] jdstrand: k [19:50] micahg: you mean like liferea? [19:50] jdstrand: liferea, akregator, and the like [19:51] jdstrand: akregator is blocked, that's what spurred the thought [19:51] chrisccoulson: I was planning on trying to fix the OOPP issue in firefox over the weekend [19:51] micahg: yes, we can, but not for lucid. we can add akregator to the profile [19:51] jdstrand: k, I'll file 2 bugs then [19:52] micahg: if another ff upload is planned for lucid, we can put it in there [19:52] and then abstract it out in maverick [19:52] jdstrand: not planned before release [19:52] AFAIK [19:53] jdstrand - barring any major show-stopper, there won't be another upload for release [19:53] micahg - do you already have an idea what causes the hang? [19:53] chrisccoulson: seems to be LD_LIBRARY_PATH [19:55] chrisccoulson: it's set in the run-mozilla.sh script IIRC, but something seems to not be right in our current env [20:10] damn, chromium still doesn't appear in the ubuntu software center as a result for chrome [20:11] chrisccoulson, ^^ do you happen to know what the search engine of that stuff is looking for? [20:12] fta - i'm not too sure, but i can ask mvo when he's around again [20:12] would be nice [20:28] fta: I think software-center's data is cached [20:29] micahg, "more info" seems up-to-date [20:29] fta: maybe it's just the .desktop info that's cached [20:29] thunderbird shows up under mail now \o/ [20:31] firefox is not in web browser though [20:31] abrowser is [20:32] fta: that's not good :( [20:32] we might need one more upload then [20:32] :( [20:33] they look the same [20:36] fta: it's because kubuntu-firefox-installer is replacing the .desktop file in app-install-date [20:37] bad [20:37] fta: yep, I need to talk to mvo about what the fix should be [21:09] chrisccoulson: can I add the flash OOPP bug to the title in the channel as I forsee a lot of bugs from the security upload? [21:38] micahg - feel free, but are people going to enable it though? [21:38] chrisccoulson: that's what I was trying to tell you before, flash is enabled by default [21:39] hmmmm. flash is working fine here (and i don't get it running in a new process) [21:39] chrisccoulson: k, I'll test this weekend and see what I get...weird [21:40] * micahg is wondering if it's the people sharing profiles with 3.6 and 3.7 that had issues [21:41] micahg - if i change "dom.ipc.plugins.enabled" to true, then flash does fail badly [21:41] but that is set to false by default, and flash seems to be working normally [21:41] chrisccoulson: do you hvae a dom.ipc.plugins.flash? [21:41] micahg - dom.ipc.plugins.enabled.libflashplayer.so [21:41] which is set to true [21:41] hmmmm [22:02] asac: micahg: chrisccoulson: something is fishy in FF 3.7... all new links tend to open in a new window, not tab! :( [22:02] specially from google reader [22:02] BUGabundo: any other site? [22:03] BUGabundo: BTW, ff37 is an old build ATM [22:03] Mozilla/5.0 (X11; U; Linux x86_64; en-US; rv:1.9.3a4pre) Gecko/20100407 Ubuntu/10.04 (lucid) Minefield/3.7a4pre [22:03] micahg: not sure! I almost don't use FF this days [22:04] only for greader cause of one addon: nosquint [22:04] nothing like it for chromium yet [22:44] hi. On Lucid Linx Beta, after adding the PPA and running a 'safe-upgrade', I got an unbranded Firefox version, named Namoroka (and using a blue icon) [22:45] but the Thunderbird version is still using the official brand (is that expected too?) [22:45] I've been told to ask micha, but it seems that user isn't logged in [22:54] micahg - ok, i got OOPP working now [22:54] ah [22:54] he already left [22:54] weird... icedtea-plugin makes firefox segfault even in safe mode. [22:57] "The Adobe Flash Plugin has crashed" [22:57] rock on! [22:58] Now it just needs to show either the Adobe logo, or the Flash logo. [22:59] here's my icedtea crash: http://pastebin.com/ERTF7BE1 [23:00] looks like an openjdk issue to me [23:02] hi gents Can I ask here about thunderbird ? [23:03] I have a fresh 10.4 (64bit) with thuderbird and lightning-extension installed, but lightning plugin does not appear for some reason [23:03] any ideas whet I missed ? [23:04] *what [23:06] mitya: is it compatible with that thunderbird version? [23:07] JanC: hmm, not sure, how can I check it ? [23:07] I thought that if it is in the standard repo, than it is [23:08] mitya: well, 10.04 is still beta, so you might want to file a bug about it... [23:09] JanC: ok, I'll take a deeper look.. thanks [23:09] the version in the archive is not supported by the latest thunderbird [23:10] mitya: you could also try to uninstall the package and install a version from mozilla's addin site (or wherever it can be found) [23:11] JanC: ok, I will try, but I read somewhere that I need to compile it for 64bit, and I was happy when I saw in within the standard repo... it seems that I can't avoid the compilation [23:12] chrisccoulson: thx [23:12] try to find an other way [23:16] mitya: compilation? [23:16] should be just a .xpi