[00:00] heh, i found another branch with a stacking problem [00:01] i've added firefox-3.7.head back until i've fixed any more that i find [00:02] right, thats the 3.6.7build2 updates done on all releases [00:03] fta: that problem is fixed in lp:firefox [00:03] chrisccoulson: I wonder how the branches got stacked like that [00:04] micahg - i'm not too sure. i've added it back for now, and i'm just running bzr reconfigure --unstacked on the other branches that are stacked against it [00:04] we can remove it again once we've fixed them [00:04] chrisccoulson: k [00:10] asac - you added your changelog entry in to the version that's already uploaded ;) [00:13] micahg - our debian/control is quite a bit smaller now ;) [00:13] chrisccoulson: for which version? [00:13] micahg - firefox-3.6.head [00:14] micahg: oh, before i move more reports to the ppa project, is the idea that all bugs filed by someone using the ppa be moved, or that just bugs affecting a release in the ppa and not the stable release be moved? [00:15] chrisccoulson: well, I wouldn't have removed all of the transitional packages, since .head is used for the dailies, we still need them until Hardy, Jaunty, and Karmic is EOL [00:15] b'ah, debcommit duplicated asac's changelog entry in my commit because i moved it around [00:15] micahg - i discussed it with asac briefly earlier, he suggested introducing a source package in to PPA's to provide transitional packages [00:16] ddecator: any report about a version in a PPA that's not the mozilla-security PPA, i.e. anything with a ~ in it [00:16] chrisccoulson: ok, that's probably a better idea :) [00:16] chrisccoulson: I'll try to do that over the weekend then [00:16] micahg: alright, just wanted to make sure i moved the right ones. i'll work on that when i get time [00:16] micahg - cool, thanks [00:16] fta: would you be able to make me a member of the umd team? [00:18] chrisccoulson: omg [00:18] man i am too much out of this ;) [00:18] should i uncommit ? [00:18] asac - no worries, i fixed it in the commit i just did [00:18] i missed it up completely ;) [00:19] even the add/update lines were collapsed [00:19] heh, no worries ;) [00:19] * asac runs away in shame [00:20] chrisccoulson: committed changelog syntax fix ;) [00:20] after all i need a few commits ;) [00:22] chrisccoulson: oh ... the patch i committed i should upstream ;) [00:22] bug me if you dont see me changing the patch to contain the mozilla version soon [00:22] micahg, i'm not the owner, asac is [00:22] asac - ok, thanks :) [00:23] micahg: what about (owner?) [00:23] asac: umd [00:23] what do you need? [00:23] asac: I just wanted to be added so I can upload [00:23] hmm. from what i understand chris wanted to take over the whole bot ;) [00:24] i think its smarter as he might have the powers to move that to a data center machine at some point ;) [00:24] heh [00:24] awesome, the hardy daily builds work again :) [00:33] jdstrand: would you be able to look at my package set addition request tomorrow? [00:37] fta_: can you reenable umd? === fta_ is now known as fta [00:37] asac: until then, can you add me so that I can upload? [00:40] i also said i'll kill the bot [00:40] i didn't, but i should [00:40] micahg, ^^ [00:41] +have [00:41] fta: ah, ok :), thanks, I'll get the builds scored up again tonight, chromium-daily as well :) [00:43] micahg, i'm not evil, i'm just angry [00:43] fta: I know, I'm trying to help where I can at the moment [00:45] micahg, if the lp thing was a new problem, i wouldn't make such a fuss, but it's a recurring issue, cycle after cycle, they don't care [00:46] seems soren is angry about the same thing [00:46] fta: indeed, well, LP seemed to be running on autopilot for the most part for a couple days which didn't help [00:47] by cycle, i meant, maverick, lucid, ..; [00:47] not just days [00:47] fta: oh, I know, I remember this last cycle [00:47] the build score issue? [00:48] fta: BTW, you're an admin of the umd team, so you can add people as well [00:51] ddecator: yes, the PPAs get overloaded when rebuilds happened and they're not always scored low [00:51] micahg: gotcha [00:52] fta: BTW, I had upstream fix the 1.9.2.8/3.6.8 dailies on Lucid/Maverick so they shouldn't FTBFS again [00:56] fta: thanks :) [00:57] i was looking at the wrong place. on the right, there was just "leave the team", no "add member" === fta_ is now known as fta === yofel_ is now known as yofel === fta_ is now known as fta === fta_ is now known as fta [04:19] ooh... when is chromium-daily/dev gonna get synced? I desperately want that cairo flickering bug fixed. === fta_ is now known as fta === fta_ is now known as fta === fta_ is now known as fta [07:21] update-alternatives: error: alternative path /usr/bin/firefox-4.0 doesn't exist. [07:21] hmm. [07:21] That's on lucid amd64. [07:21] DanaG: yeah, should be fixed in tonight's build [07:21] Sweet. [07:26] ah, meaning it'll be in dailies by tomorrow? If so, sweet. [07:27] Oh yeah, and thunderbird-3.1 in dailies hasn't been updated in, oh, this year. [07:27] Version: 3.1~a1~hg20091221r4576+nobinonly-0ubuntu1~umd1 [07:27] DanaG: yes [07:27] DanaG: yeah, I need to figure that out soon [07:28] And it was a pain trying to find a 64-bit build of lightning to go with it... I eventually just put 32-bit thunderbird-3.1 in /opt [07:29] DanaG: I'm hoping to push lightning with TB3.1 to maverick early next month [07:30] Hmm, how about Lucid? If need be, I can just grab the Maverick packages. [07:30] DanaG: will probably go to Lucid after some testing in Maverick and PPAs [07:31] PPA is good. Thanks. === fta_ is now known as fta === fta_ is now known as fta [09:33] hi chrisccoulson [09:34] good morning :) [09:34] hi micahg [09:34] how are you? [09:34] chrisccoulson: do you remember why we didn't have transitional packages for xul191 in lucid? [09:34] chrisccoulson: tired :), should have been asleep a couple hours ago, but alas, not :) [09:36] chrisccoulson: how are you? [09:36] heh, Xorg just crashed [09:36] i'm good thanks [09:36] for xul191 -> xul192, we shouldn't need transitional packages as their dependencies should pull the new package in, and they are parallel installable [09:37] chrisccoulson: yes, but the current issue is xul191 breaks eclipse in lucid, bdrung was going to add a breaks, but pitti thinks xul192 should handle it so we don't have xuls cluttering up people's systems [09:37] i'm not sure about that [09:38] i think eclipse should be fixed to not use xul191, like we've had to do with other applications in hardy -> karmic [09:38] k, I'm too tired to think too deeply about it, can you chat w/pitti about it? [09:38] yeah, can do [09:38] chrisccoulson: we did that, but it wasn't working right, and then they pulled the code that did that [09:39] they = Debian eclipse maintainers [09:54] chrisccoulson: I have a build fix for the crash reporter which I'll add in when I wake up, it needs a curl dev lib, I'm doing a local test build now [09:54] that'll cut down on the number of bugs for us :) [09:56] chrisccoulson: tty in a few hours [09:56] micahg - cool, ok. will see you later! === fta_ is now known as fta === fta_ is now known as fta [10:49] !test [10:49] hrm? [10:49] thx [10:57] asac - i had a look at your comments from yesterday about splitting the xulrunner package for UNE [10:57] and then i thought.... [10:57] ....whatever we do won't work anyway [10:58] as applications still need to discover where mozjs is (either by doing the xulrunner --gre-version hack or by using the xpcom glue) [10:58] and so they still need xulrunner to exist on the system [10:58] so, bundling mozjs in couchdb might be the better option [11:01] chrisccoulson: not sure what you mean. if we package mozjs independent we can maintain the soname [11:01] etc. [11:01] and put it in /usr/lib [11:02] asac - so, debian are maintaining there own SO name? (and just bumping it every time there's an ABI change) [11:02] yes [11:02] there is a bug that would just export public symbols [11:03] with that we might have some chance to maintain it [11:03] debian currently tries to maintain it without that fix [11:03] but well. i really think its setting wrong signal until mozilla agrees that they want this to be exported [11:03] so lets try to include mozjs in couch [11:03] and get over it [11:04] could also be statically linked in there [11:05] hmm. android checkout is 4.2G ;) [11:05] asac - ok, i'll have a chat with the couchdb guys too [11:06] who's looking after that? [11:06] they are crazy [11:06] ;) [11:06] chrisccoulson: you mean the couchdb package? thats kenvandine most likely [11:06] it was forced into ubuntu through online services ;) [11:06] and then got picked up by many things unfortunately :(( [11:07] i think upstream already includes mozjs in their tree [11:07] so we just dont need to strip it ;) [11:07] and ship it _the ugly way_ tm [11:08] maybe latest couchdb can already be built with javascriptcore? [11:09] hmm. seems not :( [11:09] at least not what is in maverick according to configure [11:12] asac - are you still going to be looking after ubufox btw? [11:12] i've got a search plugin to add here ;) [11:13] chrisccoulson: baidu? [11:13] asac - yeah [11:13] we shouldnt put it in ubufox [11:13] rather in the langpack-o-matic database [11:13] for cn [11:14] hmm. damn i forgot the hostname of that ;) [11:15] was that macquearie [11:15] or something [11:15] asac - oh, i wasn't sure whether we want to add it for all users, or just chinese users [11:15] chrisccoulson: look at http://www.baidu.com/ [11:15] i dont think anyone wants that ;) [11:15] yeah, i saw already ;) [11:15] unless you want to encourage ubuntu users to read chinese ;) [11:16] lol [11:16] so yeah. we want to add it for chinese folks only ... and make it only for chinese the default [11:16] the default change we could do in ubufox [11:16] but isnt mozilla even shipping baidu by default for cn upstream? [11:17] asac - i'm not sure what they ship [11:17] we can discuss that at the sprint next week anyway [11:17] yep. lets do that [11:17] i have to wait for arne input on the hostname anyway ... just cant remember :(( === fta__ is now known as fta === fta_ is now known as fta === fta_ is now known as fta === fta_ is now known as fta === fta_ is now known as fta === fta_ is now known as fta === fta_ is now known as fta === fta_ is now known as fta [15:13] chrisccoulson: another sec update building ;)? [15:13] just saw that you block the builders :-P [15:13] asac - yeah, i'm just hogging the builders ;) [15:13] ah its build2 [15:13] cool [15:13] you are on top :-P [15:14] and i'm about to rebase openjdk from doko's latest upload to lucid-proposed, which fixes an armel build failure [15:14] so, the builders are going to be loaded with openjdk too ;) [15:14] they've got all weekend to catch up though [15:14] they already are [15:14] 3698369 [building] Building i386 build of openjdk-6 6b18-1.8-4ubuntu2 in ubuntu lucid PROPOSED [15:14] yeah, that's doko's latest upload [15:14] chrisccoulson: you could upload it late EOD ;) [15:15] chrisccoulson: problem is that security ppa has now a higher default build score than normal archive [15:15] so you have to be a bit more sensible ;) ... otherwise its not fair [15:15] yeah, i'll wait until EOD for openjdk ;) [15:16] else i will be popular for all the wrong reasons ;) [15:17] chrisccoulson, is your openjdk in sync with debian? [15:17] i need a fix in there [15:17] fta - i'm not sure, i don't maintain openjdk [15:17] doko will know that though [15:18] if its not in sync it would show up in MoM, no? [15:18] no merges for openjdk ... guess its blacklisted ;) [15:18] bug 529242 [15:18] hell. i have many outstanding merges ;) [15:18] Launchpad bug 529242 in chromium-browser (Mandriva) (and 4 other projects) "chromium doesn't recognize icedtea6-plugin (affects: 13) (dups: 1) (heat: 105)" [Unknown,Won't fix] https://launchpad.net/bugs/529242 [15:19] i need the fix for debian 576361 [15:19] Debian bug 576361 in icedtea6-plugin "Please don't link the plugin against libxpcom, libxul, etc." [Normal,Fixed] http://bugs.debian.org/576361 [15:19] oh right. i did a all-main rebuild for armel last cycle :-P [15:21] fta: otherwise chromium cannot use jdk? [15:22] !info openjdk [15:22] !info openjdk maverick [15:22] !info openjdk openjdk-6-jre [15:22] !info openjdk-6-jre maverick [15:22] man i suck ;) [15:22] asac: Package openjdk does not exist in lucid [15:22] asac: Package openjdk does not exist in maverick [15:22] asac: 'openjdk-6-jre' is not a valid distribution: hardy, jaunty, karmic, lucid, maverick [15:22] asac: openjdk-6-jre (source: openjdk-6): OpenJDK Java runtime, using Hotspot JIT. In component main, is extra. Version 6b20~pre1-0ubuntu2 (maverick), package size 252 kB, installed size 816 kB [15:22] !info openjdk-6-jre debian [15:22] asac: 'debian' is not a valid distribution: hardy, jaunty, karmic, lucid, maverick [15:23] should be fixed [15:23] we are at b20 ... debian fix was at b18 [15:23] 6b18-1.8-4 [15:23] but you never know if doko synched the packaging too [15:23] but i would assume he did that [15:25] asac, i asked in the bug, got no answer [15:26] fta: what makes you believe its not fixed? [15:26] i would assume its fixed unless you have strong evidence that its not [16:06] asac, "what makes you believe its not fixed?" => users complaining :) [16:06] it's needed in lucid++ [16:06] fta: what are the symptoms [16:06] ? [16:08] it's in the bug: the plugin doesn't load (because it's linked against libxul while it shouldnt and the lib is unreachable) [16:08] asac, ^^ [16:09] fta: stupid chromium bug [16:09] open a new clean profile [16:09] changed proxy settings in one of them [16:09] it changes ALL of the profiles [16:11] BUGabundo_remote, did you file a bug? [16:11] not yet [16:12] just found it 3h ago [16:12] but I'm scrathing my head [16:12] on HOW is this even possbile [16:45] chrisccoulson: I had to kill my test build as I ran out of disk space, should I try it again before commiting the fix for the crashreporter? [16:48] micahg - don't worry about it, i'll do a test build before the next upload anyway [16:48] (and it will be in the dailies too) [16:48] chrisccoulson: k, I didn't know if you wanted to upload today, that's why, ok, I'll commit to 3.6 and 4.0 branches [16:49] chrisccoulson: any difference with using openssl curl vs gnutls curl? [16:56] micahg - i'm not sure about any difference. they are both on the CD though [16:56] chrisccoulson: k, I"m using the openssl library then === fta_ is now known as fta [17:15] chrisccoulson: k, pushed [17:16] hmm, firefox 4.0 still broken. [17:16] micahg - cool, thanks [17:18] DanaG, broken in what way? [17:18] the PPA builders are a little behind, if you need fixes, then you could try building yourself from the packaging in bzr [17:18] * micahg looks at the builds [17:18] micahg - 8 hours ;) [17:18] chrisccoulson: yeah [17:21] micahg - so, firefox sync doesn't actually work in our FF4.0 builds (using the extension from upstream) [17:21] it fails to initialize NSS [17:21] Ah. So how much longer to wait? [17:21] DanaG, at the moment, ages ;) [17:21] "still broken" referring to the same "alternative doesn't exist" thing. [17:21] ah, then I'll just apt-get source and dpkg-buildpackage. [17:21] DanaG, install the firefox-3.7 package [17:22] that should work around that [17:22] the symlinks are in the wrong package [17:22] Ah, thanks. [17:22] That worked. [17:24] chrisccoulson: idk why sync doesn't work [17:25] micahg - "couldn't open library" at let nsslib = ctypes.open(nssfile.path); [17:25] in WeaveCrypto.js [17:26] chrisccoulson: do we need the nss symlink is xul20? [17:26] oh, apparantly it writes that path to the console, 1 second [17:27] hmmm, perhaps it doesn't [17:27] it's a shame this happens as the component registers, i don't think i can catch that in venkman can i? [17:27] * micahg doesn't know [17:28] yeah, i need to insert a break there when it loads, which can only happen when the browser opens [17:28] that sucks [17:28] chrisccoulson: gdb? [17:29] micahg - i need to debug the JS code really [17:29] although [17:29] i suppose i could look at strace [17:29] and see where it's searching for the NSS components [17:30] http://hg.mozilla.org/services/fx-sync/file/f5293dc0b4dd/services/crypto/WeaveCrypto.js#l109 ? has logging on line 134 if you toggle the bit on line 60 [17:30] asac: Hello, are you there ? [17:30] also: probably wherever libxul lives, and not system nss [17:31] why does Ubuntu build nspluginwrapper for i386 arch ? [17:32] Mook_sb, thanks. i got confused there. i saw "debug : true" as being enabled, and then wondered why i didn't get any output ;) [17:32] i'll change that and try again [17:32] oh, hang on [17:32] true = enabled isn't it? [17:32] chrisccoulson: huh, no, then it should be enabled, yes - http://hg.mozilla.org/services/fx-sync/file/f5293dc0b4dd/services/crypto/WeaveCrypto.js#l95 [17:33] you have the browser.dom.window.dump.enabled pref set to true? [17:33] Mook_sb, i'll check [17:33] (either way, it should also be showing up in -jsconsole ...) [17:49] jdstrand: would you be able to look at my package set addition request today? [17:52] fta: doko uploaded your fix from Debian to lucid-proposed [18:00] great [18:08] anyone notice firefox-4.0 is not a command. at least it says that for me. I removed 3.7 (just to keep system clean [18:08] ) [18:10] micahg: what are you referring to? [18:10] * jdstrand feels dense [18:10] jdstrand: I opened a bug for the archive admins requesting 6 packages be added to the mozilla package set [18:11] gnomefreak: yeah, install firefox-3.7 until the next update [18:11] gnomefreak: I goofed in teh packaging and we haven't had a good build yet [18:12] micahg: ok thanks [18:13] jdstrand: bug 605453 [18:13] Launchpad bug 605453 in weave (Ubuntu) (and 5 other projects) "Please add to the Mozilla Package Set (affects: 1) (heat: 10)" [Undecided,New] https://launchpad.net/bugs/605453 === fta_ is now known as fta [18:37] micahg: so I am lookiing at the bug, and I don't know what you are talking about [18:37] jdstrand: it's a request to add packages to the package set, I was told the archvie admins were delegated this task [18:38] micahg: I don't what a 'package set' is. sorry [18:38] * jdstrand feels dense [18:39] micahg: oh, you mean for the whole archive reoganization? [18:39] * jdstrand is getting it [18:39] jdstrand: well, part of it, I have a Mozilla package set now [18:39] so that ubuntu-mozillateam can fiddle with these [18:39] jdstrand: so I can upload a limited number of packages [18:40] jdstrand: well, ubuntu-mozilla-uploaders, but yes :) [18:40] micahg: can you point me to a wiki page that tells me how to do this? [18:41] * micahg doesn't know if there is one...cjwatson set it up originally [18:41] https://wiki.ubuntu.com/ArchiveAdministration should list how to do it, but it does not [18:41] ok anyone have a clue on how to add boot options? [19:11] * gnomefreak wonders if redhat has bash scripting classes. i know simple bash and i understand more advanced but not how to impliment it === fta_ is now known as fta [20:03] jdstrand: what do I do, should I subscribe the TB? [20:04] micahg: possibly, though I think asking cjwatson how to proceed would be best [20:05] jdstrand: sorry for the trouble [20:05] np [20:11] micahg - the FF-sync issue is because the extension tries to load libnss3.so from the GRE folder [20:12] but i don't really want to add symlinks in there, after the issues we had with similar symlinks in thunderbird [20:12] chrisccoulson: ugh, seems like we need a patch for something then [20:12] yeah, i'll try and see what i can do ;) [20:22] fta, you there? [20:22] hey chris [20:25] anyway, firefox 4.0 packages are still a bit messed up. http://paste.ubuntu.com/464672/ [20:25] LLStarks: after the dailies finish building they shoudl be fine [20:25] LLStarks: install firefox-3.7 [20:25] not touching firefox 3.7 or 4.0 again until that addon menu dies [20:26] or extension compatability can be overriden [20:26] LLStarks: after tonight, it should be fine, but some files accidentally got into firefox-3.7 [20:26] oh [20:26] * micahg would update topic, but will be resolved w/in 7 hrs [20:28] micahg: 3.1 is Maverick or are we waiting for 11.04? [20:37] gnomefreak: 3.1 will be in Maverick early next month hopefully [20:37] micahg: ok thanks. let me know when testing can start [20:38] gnomefreak: as soon as you see it in the dailies :) === fta_ is now known as fta [20:38] oops makes alot of sense [20:39] gnomefreak: there's an installer bug ATM where the binaries aren't installed [20:39] damn === fta_ is now known as fta === fta_ is now known as fta === fta_ is now known as fta